Patel v Mirza

“No court will lend its aid to a man who founds his cause of action upon an immoral or an illegal act.” So spoke Lord Mansfield in Holman v Johnson (1775) 1 Cowp 341, 343, ushering in two centuries and more of case law about the extent and effect of this maxim. He stated that the reason was one of public policy:

Illegality has the potential to provide a defence to civil claims of all sorts, whether relating to contract, property, tort or unjust enrichment, and in a wide variety of circumstances.

Take the law of contract. A contract may be prohibited by a statute; or it may be entered into for an illegal or immoral purpose, which may be that of one or both parties; or performance according to its terms may involve the commission of an offence; or it may be intended by one or both parties to be performed in a way which will involve the commission of an offence; or an unlawful act may be committed in the course of its performance. The application of the doctrine of illegality to each of these different situations has caused a good deal of uncertainty, complexity and sometimes inconsistency.

Holman v Johnson involved a claim for the price of goods which the plaintiff sold to the defendant in Dunkirk, knowing that the defendant’s purpose was to smuggle the goods into England. The plaintiff was met with a defence of illegality. The defence failed. Lord Mansfield held that knowledge on the part of the plaintiff that the defendant intended to smuggle the goods did not affect the plaintiff’s entitlement to recover the price of the goods, since he was not himself involved in the smuggling. By contrast, in Pearce v Brooks (1866) LR 1 Ex 213 a claim by a coachbuilder against a prostitute for the hire of what was described in the law report as an “ornamental brougham” was held to be unenforceable for illegality after the jury found that the defendant hired it for the purpose of prostitution and that the plaintiff knew that this was her purpose. It would seem that the difference between Holman v Johnson and Pearce v Brooks had to do with the type of goods supplied, because in both cases the plaintiff knew that the defendant was entering into the contract for an illegal or immoral purpose. In JM Allan (Merchandising) Ltd v Cloke [1963] 2 QB 340, 348, Lord Denning MR endeavoured to rationalise the authorities by saying that “active participation debars, but knowledge by itself does not”. However, the Law Commission commented in its discussion of the subject in its Consultation Paper on Illegal Transactions: the Effect of Illegality on Contracts and Trusts, LCCP 154 (1999) that the case law lacks clear guidance on what amounts to “participation” in this context.

It is unclear to what extent the doctrine of illegality applies to a contract whose object includes something which is in some respect unlawful, or the performance of which will involve some form of illegality, but not in a way which is central to the contract. In St John Shipping Corpn v Joseph Rank Ltd [1957] 1 QB 267, 288, Devlin J said:

As to illegality in the manner of performance of a contract, Mance LJ observed in Hall v Woolston Hall Leisure Ltd [2001] 1 WLR 225, 246, that the conceptual basis on which a contract not illegal nor prohibited at the time of its formation may become unenforceable due to the manner of its performance is open to debate. In Anderson Ltd v Daniel [1924] 1 KB 138 a claim for the price of goods was held to be unenforceable because the seller had failed to give the buyer an invoice containing details which the seller was required to give him by statute. In the St John Shipping case Devlin J rejected the interpretation that the claim in Anderson Ltd v Daniel failed because in the course of performing a legal contract the plaintiff had done something illegal. The correct interpretation, he said, was that “the way in which the contract was performed turned it into the sort of contract that was prohibited by the statute”: [1957] 1 QB 267, 284. In the St John Shipping case the claim was for freight under a charter party. In the course of taking on bunkers the vessel was overloaded and the master thereby committed an offence, for which he was prosecuted and fined £1,200. The extra freight earned by the overloading was £2,295 and to that extent the ship owners stood to profit from their wrong. The cargo owners refused to pay that part of the freight. Devlin J rejected their defence. He held that since the goods had been delivered safely, the ship owners had proved all that they needed. He was not prepared to construe the statute as having the effect of making the contract prohibited. If it had been otherwise, the ship owners would not have been entitled to any freight and would therefore have suffered an additional penalty, much greater than that provided for by Parliament, for conduct which might have been unintentional.

In Ashmore, Benson, Pease and Co Ltd v Dawson [1973] 1 WLR 828 the Court of Appeal adopted a different approach. Manufacturers of heavy engineering equipment entered into a contract of carriage with road hauliers. There was nothing illegal in the formation of the contract, but the hauliers overloaded the vehicles which were to transport the load, in breach of road traffic regulations, and one of the lorries toppled over during the journey as a result of the driver’s negligence. The manufacturers’ transport manager was present when the goods were loaded and was aware of the overloading. A claim by the manufacturers for the cost of repair of the damaged load was rejected on grounds of illegality. The Court of Appeal did not perform the same analysis as had Devlin J in the St John Shipping case. They held simply that the manufacturers participated in the illegal performance of the contract and were therefore barred from suing on it.

These and other cases led the Law Commission to describe the effect that unlawful performance has on the parties’ contractual rights as very unclear. (Consultative Report on the Illegality Defence, LCCP 189 (2009), para 3.27.)

In this case the issue is whether Lord Mansfield’s maxim precludes a party to a contract tainted by illegality from recovering money paid under the contract from the other party under the law of unjust enrichment (to use the term now generally favoured by scholars for what used previously to be labelled restitution and, before that, quasi-contract). On one side it is argued that the maxim applies as much to such a claim as to a claim in contract, and that the court must give no assistance to a party which has engaged in any form of illegality. On the other side it is argued that such an approach would not advance the public policy which underlies Lord Mansfield’s maxim, once the underlying policy is properly understood.

With that introduction I turn to the facts of Mr Patel’s claim and how it was decided in the courts below: see paras 11-16. A central part of their judgments, and of Mr Mirza’s argument, concerns the doctrine of reliance applied by the House of Lords in Tinsley v Milligan [1994] 1 AC 340: see paras 17-20. That decision led to the Law Commission conducting a comprehensive review of the law of illegality and making proposals for addressing what the Commission perceived to be its unsatisfactory features: see paras 21-49. Paras 33-39 concern European law and its potential impact on our domestic law. The approach adopted in Australia, Canada and the USA is considered at paras 50-66. Paras 67-81 address developments since the Law Commission’s report, including three recent decisions of this court which laid bare a division of opinion about the framework for deciding issues of illegality. Paragraphs 82-94 contain a section entitled “The law at a crossroads”. This leads to the critical part of the judgment, which considers the way forward and ends in a summary and proposal for the disposal of this appeal: paras 95-121. The reader who is more interested in what the judgment has to say about the future than the past will no doubt wish to concentrate on the final section.

The essential facts can be shortly told. Mr Patel transferred sums totalling £620,000 to Mr Mirza for the purpose of betting on the price of RBS shares, using advance insider information which Mr Mirza expected to obtain from RBS contacts regarding an anticipated government announcement which would affect the price of the shares. Mr Mirza’s expectation of a government announcement proved to be mistaken, and so the intended betting did not take place, but Mr Mirza failed to repay the money to Mr Patel despite promises to do so. Mr Patel thereupon brought this claim for the recovery of the sums which he had paid. The claim was put on various bases including contract and unjust enrichment. A fuller account of the facts is given in the judgments of the courts below and in the judgment of Lord Neuberger.

The agreement between Mr Patel and Mr Mirza amounted to a conspiracy to commit an offence of insider dealing under section 52 of the Criminal Justice Act 1993. In order to establish his claim to the return of his money, it was necessary for Mr Patel to explain the nature of the agreement.

A defendant’s enrichment is prima facie unjust if the claimant has enriched the defendant on the basis of a consideration which fails. The consideration may have been a promised counter-performance (whether under a valid contract or not), an event or a state of affairs, which failed to materialise. (See Professor Andrew Burrows’ A Restatement of the English Law of Unjust Enrichment, 2012, p 86, para 15). In Sharma v Simposh Ltd [2013] Ch 23, at para 24, the Court of Appeal cited with approval Professor Birks’ summary of the meaning of failure of consideration in his revised edition of An Introduction to the Law of Restitution (1989), p 223:

Applying the “reliance principle” stated in Tinsley v Milligan [1994] 1 AC 340, the judge held that Mr Patel’s claim to recover the sum paid was unenforceable because he had to rely on his own illegality to establish it, unless he could have brought himself within the exception of the doctrine known, misleadingly, as locus poenitentiae; and that he could not bring himself within that exception since he had not voluntarily withdrawn from the illegal scheme.

In the Court of Appeal the majority agreed with the judge on the reliance issue, but disagreed with him on the application of the locus poenitentiae exception. They held that it was enough for the claim to succeed that the scheme had not been executed. Gloster LJ agreed with the majority that Mr Patel’s claim should succeed but she took a different approach to it. She began her thoughtful analysis with a cri de coeur (para 47):

On the issue of reliance, Gloster LJ did not consider it necessary for Mr Patel to establish that the intended betting on RBS shares was to be done with the benefit of insider information; it would have been enough for him to establish that the funds had been paid for the purpose of a speculation on the price of the shares which never took place. If, however, she were wrong on that issue, she agreed with the other members of the court on the locus poenitentiae issue.

The facts of Tinsley v Milligan are well known. Miss Tinsley and Miss Milligan each contributed to the purchase of a home. It was vested in Miss Tinsley’s sole name, but on the mutual understanding that they were joint beneficial owners. It was put in her sole name so as to assist Miss Milligan to make false benefit claims from the Department of Social Security (DSS), which she did over a number of years with Miss Tinsley’s connivance. The money obtained from the DSS helped them to pay their bills, but it played only a small part in the acquisition of the equity in the house. Eventually Miss Milligan confessed to the DSS what she had done and made terms with it, but the parties fell out. Miss Tinsley gave Miss Milligan notice to quit and brought a claim against her for possession. Miss Milligan counterclaimed for a declaration that the property was held by Miss Tinsley on trust for the parties in equal shares.

The Court of Appeal by a majority decided in favour of Miss Milligan by applying the test whether it would be “an affront to the public conscience” to grant the relief claimed by her. The House of Lords unanimously rejected the “public conscience” test, but by a three to two majority upheld the Court of Appeal’s decision. The leading speech was given by Lord Browne-Wilkinson. His starting point was that title to property can pass under an unlawful transaction; but he held that the court would not assist an owner to recover the property if he had to rely on his own illegality to prove his title. The Court of Appeal had recognised that distinction in Bowmakers Ltd v Barnet Instruments Ltd [1945] KB 65 in a case concerning personal property, referred to in more detail at para 111 below, and Lord Browne-Wilkinson held that the same applied to real property in which the claimant had a beneficial interest. Lord Browne-Wilkinson held that it was enough for Miss Milligan to show that she had contributed to the purchase of the property and that there was a common understanding that the parties were joint owners. She did not have to explain why the property had been put into Miss Tinsley’s sole name. If the relationship between them had been that of daughter and mother, and each had contributed to the purchase of a property in the daughter’s name, the result would have been different, because there would then have been a presumption of advancement in the daughter’s favour. The mother would in those circumstances have had to rely on the illegal nature of the transaction to rebut the presumption, and her claim would therefore have been defeated by the doctrine of illegality. Lord Browne-Wilkinson acknowledged the procedural nature of this approach at [1994] 1 AC 340, 374:

Lord Goff, in the minority, held at p 356 that if A puts property in the name of B intending to conceal A’s interest for a fraudulent or illegal purpose, neither law nor equity will allow A to recover the property, and equity will not assist him in asserting an equitable interest in it. It made no difference whether A’s case could be advanced without reference to the underlying purpose. He recognised, at p 363, the resulting hardship and said that he did not disguise his unhappiness at the result, but he did not regard it as appropriate for the courts to introduce a discretion. He considered, at p 364, that reform should be instituted only by the legislature, after a full inquiry by the Law Commission, which would embrace not only the advantages and disadvantages of the present system, but also the likely advantages and disadvantages of a discretionary system. He added that he would be more than happy if a new system could be evolved which was both satisfactory in its effect and capable of avoiding the kind of result which in his judgment flowed from the established rules in cases such as Tinsley v Milligan.

Tinsley v Milligan has been the subject of much criticism in this and other jurisdictions, for its reasoning rather than its result, but this is the first time in this jurisdiction that its reasoning has been directly called into question. Two decades have since passed since the decision and it is right to trace the developments which have occurred in that period.

After the decision in Tinsley v Milligan the Law Commission included the illegality defence in its Sixth Programme of Law Reform (1995) (Law Com 234). It undertook a full inquiry of the kind which Lord Goff envisaged. It published its first consultation paper, Illegal Transactions: The Effect of Illegality on Contracts and Trusts (LCCP 154), in 1999. The responses, and developments in the case law, led the Commission to re-consider the problems and its proposals for reform. In 2009 it issued a further public consultation paper, The Illegality Defence: A Consultative Report (LCCP 189). In 2010 it issued its final confirmatory report, The Illegality Defence (Law Com 320). In relation to trust law, it proposed statutory reform and it produced a draft bill. In relation to the law of contract and unjust enrichment, the Commission considered that there were serious problems but that they were capable of being, and could best be, tackled by the process of judicial development. In 2012 the government announced that it did not intend to take forward the Commission’s recommendation for statutory reform of the law relating to trusts, because it did not consider reform of this area of the law to be a pressing priority for the government.

From its study of the case law and academic writing, the Commission identified the principal policy rationales for the illegality doctrine as 1) furthering the purpose of the rule infringed by the claimant’s behaviour, 2) consistency, 3) prevention of profit from the claimant’s wrongdoing, 4) deterrence and 5) maintaining the integrity of the legal system. It observed that these rationales were not mutually exclusive but overlapped to a greater or lesser degree. A sixth possible rationale, punishment, was controversial. The large majority of consultees considered that punishment was a matter for the criminal courts (to which one might add regulators) and should not be invoked in determining parties’ civil disputes. (LCCP 189, paras 2.5-2.31.)

The conclusion that the illegality defence presented serious problems represented the overwhelming view of academic commentators and consultees generally. The Commission analysed the problems under four heads - complexity, uncertainty, arbitrariness and lack of transparency. It did not suggest that the problems resulted generally in unsatisfactory outcomes, but it was critical of the way in which they were reached. It said that, on the whole, the case law illustrated the judges threading a path through the various rules and exceptions in order to reach outcomes which for the most part would be regarded as fair between the parties involved, although there were instances of results which the Commission considered to be unduly harsh, for example in unlawful employment cases. Generally, the courts managed to avoid unnecessarily harsh decisions either by creating exceptions to the general rules or by straining the application of the relevant rules on the particular facts so as to meet the justice of the case. Seldom was there an open discussion in the judgments of the considerations which led the court to its decision. (LCCP 189, paras 3.50-3.60.)

The Commission considered that Tinsley v Milligan, and cases following it, exemplified the problems of arbitrariness, uncertainty and potential for injustice. The rule applied in that case was arbitrary in that the question whether the illegality affected the recognition or enforcement of the trust depended not on the merits of the parties, nor the policies underlying the illegality defence, but on a procedural issue. Moreover the effect of applying the reliance principle in cases involving the presumption of advancement gave that presumption an overriding importance which it was never intended to have. It led to uncertainty because there was much confusion over what exactly amounted to “reliance”, particularly when the claimant was seeking to establish an equitable interest under a constructive trust. It had the potential to force the court into unjust decisions because, by focusing on procedural matters, the reliance principle precluded the court from paying attention to the policies that justified the existence of the defence, or taking into account such matters as the seriousness of the illegality and the value of the interest at stake. (Law Com 320, paras 2.13-2.15.)

The Commission examined the law in other jurisdictions, European law and European human rights law. In its first consultation paper in 1999 the Commission’s proposed recommendation was to introduce statutory reform on the lines of the New Zealand model. The New Zealand Illegal Contracts Act 1970, section 7, provides that the court may grant to any party to an illegal contract “such relief by way of restitution, compensation, variation of the contract, validation of the contract in whole or part or for any particular purpose, or otherwise howsoever as the court in its discretion thinks just”. In its 2009 consultative report the Commission noted that the operation of this provision had been widely heralded as a success; that it had not created the deluge of litigation that was feared by some commentators; and that this model of reform, with slight variations, had been recommended by the law reform bodies of several other Commonwealth jurisdictions (LCCP 189, para 3.81). Nevertheless, in its 2009 consultative report and in its final report the Commission did not recommend statutory change (except in relation to trusts) for a combination of reasons. Although the proposal for statutory reform in the 1999 consultation paper had been supported by a majority of consultees, a minority had made critical comments which persuaded the Commission that judicial reform was a better way forward, and the Commission found difficulties in drafting a satisfactory statutory model. Most importantly, developments in the case law and the critical responses of consultees led the Commission to conclude that it was open to the courts to develop the law in ways that would render it considerably clearer, more certain and less arbitrary.

Among domestic authorities, the Commission referred to the decisions of the House of Lords in Bakewell Management Ltd v Brandwood [2004] 2 AC 519 and Gray v Thames Trains Ltd [2009] AC 1339.

Bakewell bought an area of land registered as a common. Owners of neighbouring properties had for years driven across the land to reach the public highway. Bakewell brought an action to prevent them from continuing to do so. The defendants claimed to have acquired rights of way by prescription, but by driving across the land without the owner’s consent they had committed offences under the Law of Property Act 1925. So to establish their property rights the defendants had to rely on conduct which was criminal. This, Bakewell submitted, they were not entitled to do. Its argument was rejected. The House of Lords held that public policy did not prevent the defendants from acquiring an easement where the landowner could have made a grant which would have removed the criminality of the user. Lord Walker, with whom Lord Bingham and Lady Hale agreed, said at para 60:

Gray v Thames Trains Ltd was a case in tort. Mr Gray developed post-traumatic stress disorder through being involved in a major railway accident, which caused him to suffer depression and a substantial personality change. He was previously of unblemished character but two years after the accident, and while under medical treatment, he pursued and stabbed to death a man who had stepped in front of his car. His plea of guilty to manslaughter on the ground of diminished responsibility was accepted and he was ordered to be detained in a mental hospital. He sued the train operator for negligence and liability was admitted. His claim for damages included compensation for his loss of liberty, damage to reputation and loss of earnings during his detention. The House of Lords held that public policy precluded him from recovering damages under those heads. The leading opinion was given by Lord Hoffmann, with whose reasoning Lord Phillips (subject to certain additional observations) and Lord Scott agreed.

Lord Hoffmann observed, at paras 30-32, that the maxim ex turpi causa expresses not so much a principle but a policy based on a group of reasons, which vary in different situations. The courts had therefore evolved varying rules to deal with different situations. Because questions of fairness and policy were different in different cases and led to different rules, one could not simply extrapolate rules applicable to one situation and apply them to another. It had to be assumed that the sentence was what the criminal court regarded as appropriate to reflect Mr Gray’s personal responsibility for the crime he had committed. It was therefore right to apply the rule that he could not recover damages for the consequences of the sentence, reflecting an underlying policy based on the inconsistency of requiring someone to be compensated for a sentence imposed because of his personal responsibility for a criminal act. It was also to right to apply a wider rule that you cannot recover damage which is the consequence of your own criminal act, reflecting the idea that it is offensive to public notions of the fair distribution of resources that a claimant should be compensated (usually out of public funds) for the consequences of his own criminal conduct.

Lord Phillips said, at para 15, that he would reserve judgment as to whether the ex turpi causa maxim should apply if it were clear from the judge’s sentencing remarks that the claimant’s offending behaviour played no part in the decision to impose a hospital order, or, where the claimant’s criminal act demonstrated a need to detain him both for his own treatment and for the protection of the public, if the judge made it clear that he did not believe that the claimant should bear significant personal responsibility for his crime. Lord Brown agreed with Lord Phillips’ reservations.

Lord Rodger said, at paras 78-83, that the civil court must assume that the order made by the criminal court was appropriate to reflect Mr Gray’s personal responsibility for the crime he had committed. The right approach on the facts of the case was that the court must “cleave to the same policy as the criminal court”. However, he considered that the approach might well be different if the offence of which he had been convicted was trivial but revealed that he was suffering from a mental disorder, due to the defendant’s fault, which made a hospital order appropriate.

The Law Commission drew from the various judgments a readiness on the part of the judges to examine the policy reasons which justified the application of the illegality defence and to explain why those policies applied to the facts of the case.

The Commission also considered the question how far illegal conduct may deprive claimants of rights under European Union law (LCCP 189, paras 3.82-3.89). Some contractual rights are now derived from EC directives. For example, the right to equal pay granted by the Equal Pay Directive (directive 75/117/EEC) is implied as a term into the employment contract. In other cases, such as the Sale of Consumer Goods Directive (directive 99/44/EC), EU law provides remedies that depend on the existence of a contract. The issue may therefore arise whether a national illegality doctrine which prevents a party from enforcing a contract is compatible with the EU law from which the contractual right arose.

In the 1990s various breweries let pubs to tenants on terms containing beer ties. These were found to be unenforceable because they breached article 81 (previously article 85) of the European Community Treaty. The issue then arose whether the fact that the tenant had been party to an illegal contract precluded him from claiming damages from the brewery. In Gibbs Mew plc v Gemmell [1999] 1 EGLR 43, 49 the Court of Appeal held that this was so, because “English law does not allow a party to an illegal agreement to claim damages from the other party for loss caused to him by being a party to the illegal agreement” (per Peter Gibson LJ).

In Courage Ltd v Crehan (Case C-453/99) [2002] QB 507, the Court of Appeal referred the question to the European Court of Justice, which took a different view. Advocate General Mischo expressed the view, at paras 38-43, that although the individuals protected by article 81 were primarily third parties (consumers and competitors), a rule which automatically excluded a party to the agreement from the protection of article 81 was “too formalistic and does not take account of the particular facts of individual cases”; and that a party which was too small to resist the economic pressure imposed on it by the more powerful undertaking had more in common with a third party than with the author of the agreement. (The potential parallel with the relationship in some cases between an employer and an employee is obvious.)

The court agreed with the Advocate General. It held that where a contract was liable to restrict or distort competition, community law did not preclude a rule of national law from barring a contracting party from relying on his own illegal actions, if it was established that that party bore significant responsibility for the distortion of competition. In that context the matters to be taken into account by the national court included the respective bargaining power and the conduct of the parties to the agreement in the economic and legal context in which they found themselves. It was for the national court to ascertain whether the party who claimed to have suffered loss through concluding such a contract was in a markedly weaker position than the other party, such as seriously to compromise or even eliminate his freedom to negotiate the terms of the contract. An absolute bar to an action being brought by a party to a contract which violated the competition rules would not advance the full effectiveness of the prohibition contained in the Treaty, but rather the reverse.

The effect of the court’s decision was not to treat article 81 as intended for the protection of parties who infringed it, as a class, but to treat it as a matter for the national court to determine whether on the facts of a particular case a party should be regarded as sinned against rather than sinning, and therefore entitled to damages for the consequences of the offending provision of the agreement.

The potential impact of European law was referred to, obiter, by Mance LJ in Hall v Woolston Hall Leisure Ltd [2001] 1 WLR 225. The claimant was dismissed from her employment as a chef when her employer became aware that she was pregnant. She brought a claim in the industrial tribunal for compensation under the Sex Discrimination Act 1975. The Act pre-dated the Equal Treatment Directive (76/207/EEC) but gave effect to its provisions. Mrs Hall succeeded on liability, but it emerged during the remedies hearing that her employer was defrauding the Inland Revenue by falsely pretending that her net salary of £250 per week was her gross salary. She was aware of the fraud, because she was given pay slips which showed her gross pay as £250, deductions of £63.35 and net pay of £186.65. She knew that this was untrue, but when she raised the matter with her employer she was told that this was the way in which they did business. The tribunal held that the contract was tainted by illegality and that she had no right to compensation under the Act. Its decision was upheld by the appeal tribunal but reversed by the Court of Appeal, which held that her acquiescence in the employer’s conduct was not causally linked with her sex discrimination claim and that public policy did not preclude her from enforcing her statutory claim. Mance LJ observed additionally that the Act should as far as possible be read as providing the same scope of protection as the Directive. Mrs Hall’s position fell within the wording and purpose of the Directive despite the tribunal’s finding of her knowledge of the fraud on the Inland Revenue.

That case did not involve the direct enforcement of a contractual obligation, but in cases where European Union rights depend on the existence of a contract (for example, in the consumer context), the Law Commission doubted whether the Court of Justice would be content with a system of domestic illegality rules which were formalistic and did not allow room for a proportionate balancing exercise to be carried out on the basis of clear principles of public policy (LCCP 189, para 3.89).

Where the terms or performance of a contract involve breach of a legislative provision, it is rare (as the Commission noted) for the statute to state expressly what are to be the consequences in terms of its enforceability. (For an example of an express statutory unenforceability provision, see section 127(3) of the Consumer Credit Act 1974, which arose for consideration in Wilson v First County Trust Ltd (No 2) [2004] 1 AC 816.) It is to be noted that in the present case, as Gloster LJ pointed out, section 63(2) of the Criminal Justice Act 1993 stipulated that “No contract shall be void or unenforceable by reason only of section 52”, presumably because of a concern that if a contract which involved insider dealing contrary to section 52 were void, there could be undesirable consequences for parties down the line. The question whether a statute has the implied effect of nullifying any contract which infringes it requires a purposive construction of the statute, as illustrated by the decision of the Court of Appeal in Hughes v Asset Managers plc [1995] 3 All ER 669 which the Commission commended.

If a contract involving prohibited conduct is not void as a matter of statutory construction, the Commission recommended that in deciding whether a claim arising from it should be disallowed by reason of illegality, the court should have regard to the policies that underlie the doctrine. It stressed that it was not advocating a general discretion, but a principled evaluation recognising (as Lord Walker put it in the Bakewell case, at para 60) that the maxim ex turpi causa must be applied as an instrument of public policy and not in circumstances where it would not serve the public interest. The Commission identified a number of potentially relevant factors: most importantly, whether allowing the claim would undermine the purpose of the rule which made the relevant conduct unlawful, and, linked to that question, the causal connection between the illegality and the claim (including how central the illegality was to the contract), the gravity of the conduct of the respective parties and the proportionality of denying the claim. (LCCP 189, para 3.142) The Commission recommended a broadly similar approach to the maxim ex turpi causa in cases of unjust enrichment, tort and enforcement of property rights.

The Commission considered that it was within the power of the courts to develop the law in that direction and that there were signs of willingness to do so. The underlying principles were already to be found in the case law and courts were in practice influenced by them in reaching their decisions, in some cases more openly than in others.

In relation to the application of the illegality defence to claims of unjust enrichment, the Commission carried out a detailed review in its 1999 consultation paper (LCCP 154, paras 2.32-2.56) and a further review in its 2009 consultative report (LCCP 189, paras 4.1-4.62). An unjust enrichment claim may simply be to unwind the transaction by repayment of moneys paid and restoration of the parties to their original position, or it may take the form of a claim for recompense for benefits provided by one party to the other (a quantum meruit claim).

The Commission observed that one might have expected to find that illegality has little role to play as a defence to a claim for unjust enrichment, since the claimant is not seeking to execute the contract. However, after a more liberal start, the courts adopted a much tougher stance, applying the ex turpi causa maxim to such claims unless the claimant could bring himself within certain recognised exceptions. These were a) duress, b) possibly ignorance of a fact or law that rendered the contract illegal, c) possibly membership of a vulnerable class protected by statute and d) locus poenitentiae. The locus poenitentiae exception has given rise to difficult and conflicting case law, which was meticulously analysed in the judgments of the courts below in the present case with different conclusions. I do not propose to repeat their analysis because I do not consider it necessary to do so. The topic has only acquired importance because of the strictness of the basic rule which the courts have applied.

Not every case, however, has received such strict treatment. In Mohamed v Alaga & Co [2000] 1 WLR 1815 the Court of Appeal took a more flexible approach. The plaintiff, a Somali translator and interpreter, sued the defendant solicitors for breach of a contract by which he was to introduce Somali refugees to the firm, and assist in the preparation and presentation of their asylum claims, in consideration for a half share of the legal aid fees received by the firm. Alternatively, he claimed payment for his professional services as a translator and interpreter on a quantum meruit. His claim was struck out on the ground that the alleged fee sharing contract contravened rules which had statutory force under the Solicitors Act 1974 and that he was therefore precluded by the doctrine of illegality from claiming payment for services provided under the contract. The Court of Appeal restored the claim for payment on a quantum meruit.

Lord Bingham CJ (with whom the other members of the court agreed) differentiated between the claims for breach of contract and quantum meruit. As to the former, he held that the purpose of the prohibition in the statutory rules was the protection of the public, and that it would defeat the purpose of the prohibition if a non-solicitor party to the agreement could invoke the court’s aid to enforce the agreement. As to the quantum meruit claim, Lord Bingham acknowledged that on one view of the case the plaintiff was seeking to recover part of the consideration payable under an unenforceable contract. But he preferred to view it as a claim for a reasonable reward for professional services rendered. He considered it relevant (obviously to the question of the public interest in permitting or disallowing the claim) that the parties were not equal in blameworthiness. The firm could be assumed to know the rules and the likelihood was that it had acted in knowing disregard of them. By contrast, Lord Bingham had no difficulty in accepting that the plaintiff was unaware of any reason why the firm should not make the agreement, which was a common type of agreement in other commercial fields.

Mr Matthew Collings QC for Mr Mirza submitted in this case that Mohamed v Alaga & Co was a one off case and either represents an exception, peculiar to its particular facts, to the general rule that a party is not entitled to payment for services rendered under an illegal contract or was wrongly decided.

The Commission considered that the policies which underlie the illegality defence are less likely to come into play where parties are attempting to undo, rather than carry out, an illegal contract. As in the case of contractual enforcement, it recommended that a decision on disallowing a particular restitutionary claim for illegality should be based openly on the policies underlying the defence, taking into account the same sort of factors (such as the relative conduct of the parties and the proportionality of denying the claim).

I have said that the Commission examined the law of other jurisdictions. Before considering developments in domestic law since the Commission’s final report, it is convenient at this stage to refer to the law in Australia, Canada and the USA.

In Nelson v Nelson [1995] HCA 25; (1995) 184 CLR 538, the High Court of Australia considered essentially the same issues as in Tinsley v Milligan, which it declined to follow. As the widow of a mariner who had served in World War 1, Mrs Nelson was eligible under the Defence Service Homes Act 1918 to buy a house with the benefit of a subsidy from the Commonwealth of Australia, provided that she did not own or have a financial benefit in another house. She provided the money to buy a house in Bent Street, Sydney, but the transfer was taken in the names of her son and daughter. Their common intention was that Mrs Nelson should be the beneficial owner of the house. The reason for putting the Bent Street property in the names of her children was to enable her to buy another property with the benefit of a subsidy under the Act. This she did. One year later the Bent Street property was sold. By this time Mrs Nelson and her daughter had fallen out, and a dispute arose as to who was entitled to the sale proceeds. Mrs Nelson and her son brought proceedings against the daughter for a declaration that the proceeds were held by the son and daughter in trust for their mother. The daughter opposed the claim and sought a declaration that she had a beneficial interest. Under Tinsley v Milligan the daughter would have succeeded, because the illegal purpose of the parties in arranging for the property to be transferred into the names of the children would have prevented Mrs Nelson from rebutting the presumption of advancement in their favour.

The High Court unanimously rejected that approach. The majority (Deane, McHugh and Gummow JJ) held that the court should use its equitable jurisdiction to grant the declaration sought by Mrs Nelson, with the proviso that it should be subject to terms designed to ensure that the benefit wrongly obtained on the purchase of the second property should be repaid to the Commonwealth. The minority (Dawson and Toohey JJ) would have made the declaration without any such proviso, since the Commonwealth was not a party to the proceedings and should in their view be left to decide what action, if any, it wished to take.

Toohey J said at pp 595-597:

McHugh J went on to suggest that except in a case where a statute made rights arising out of a particular type of transaction unenforceable in all circumstances, a court should not refuse to enforce legal or equitable rights on the ground of illegality if to do so would be disproportionate to the seriousness of the conduct or if it would not further the purpose of the statute. He said at 612-613:

Noting the criminal sanctions which were available under the Act (imprisonment for up to two years) and the ability of the Commonwealth to recover any payments wrongly obtained by Mrs Nelson, the court did not consider that it should impose a further sanction by refusing to enforce her equitable rights, particularly when such a refusal would result in a penalty out of all proportion to the seriousness of her conduct (pp 570-571 per Deane and Gummow JJ, 590-591 per Toohey J and 616-617 per McHugh J).

In Hall v Hebert [1993] 2 SCR 159 the owner of a car allowed a passenger to drive it in the knowledge that he had drunk a large amount of beer during the course of the evening. The car overturned and the driver suffered head injuries. The Supreme Court held that the driver’s claim against the owner in negligence was not barred by illegality, but that there should be a reduction in damages for contributory negligence. The judgment of the majority was given by McLachlin J.

She held that the courts should be allowed to bar recovery in tort on the ground of the plaintiff’s illegal or immoral conduct only in very limited circumstances. The basis of the power lay in the duty of the courts to preserve the integrity of the legal system and it was exercisable only where that concern was in issue. It was in issue where a damage award in a civil suit would allow a person to profit from illegal or wrongful conduct, or would permit an evasion or rebate of a penalty prescribed by the criminal law. In such instances the law refused to give by its right hand what it took away by its left hand.

McLachlin J emphasised the importance of defining what was meant by profit when speaking of the plaintiff profiting from his or her own wrong. It meant profit in the narrow sense of a direct pecuniary award for an act of wrongdoing. Compensation for something other than wrongdoing, such as for personal injury, would not amount to profit in that sense. Compensation for the plaintiff’s injuries arose not from the illegal character of his conduct, but from the damage caused to him by the negligent act of the owner in letting him drive. It represented only the value of, or substitute for, the injuries he had suffered by the fault of another. He would get nothing for being engaged in illegal conduct. McLachlin J accepted that there might be cases where a claim should be barred from tort recovery which did not fall within the category of profit, in order to prevent stultification of the criminal law or the evasion of a criminal penalty, but the underlying principle was that the use of the power to deny recovery on the ground of illegality was justified only where the claim would introduce inconsistency into the fabric of the law.

In Still v Minister of National Revenue (1997) 154 DLR (4th) 229 an American citizen lawfully entered Canada and applied for permanent residence status. Pending consideration of her application, acting in good faith, she accepted employment but did so without obtaining a work permit as required by the Immigration Act 1985. She was subsequently laid off and submitted a claim for benefits under the Unemployment Insurance Act 1985. Her claim was rejected on the ground that the employment on which she relied in order to found her claim was prohibited under the Immigration Act. She appealed successfully to the Federal Court of Appeal.

The judgment of the court was given by Robertson JA. The court accepted that her employment without a work permit was expressly prohibited by the Immigration Act. It acknowledged that under what it described as the “classical model” of the illegality doctrine, the fact that the applicant acted in good faith was irrelevant; her employment under an illegal contract could not constitute insurable employment for the purposes of the Unemployment Insurance Act. However, it said at para 24 that in recognition of the rigidity and oft-times unfair application of the classical illegality doctrine, the courts had developed several ways in which a party may be relieved of the consequences of illegality where appropriate. The difficulty with those exceptions arose from “the legal manoeuvring that must take place to arrive at what is considered a just result”. The court examined, at paras 25-36, a line of authorities of the Ontario courts which showed the courts turning from the classical model towards a modern approach. It expressed the view, at para 42, that the classical model had lost its persuasive force, and was now honoured more in the breach than in its observance through the proliferation of so-called judicial “exceptions” to the rule. The new approach involved an examination of the purpose underlying the relevant prohibition, and its rationale was explained by McLachlin J in Hall v Hebert.

After citing McLachlin J’s judgment in Hall v Hebert, the court said at para 49:

The court proceeded to consider the objectives underlying each of the two Acts. As to the policy consideration that a person should not benefit from his or her own wrong, the court regarded it as a critically significant fact that she had not deliberately broken the law but acted in good faith, and it noted that during her employment both the applicant and her employer had contributed to the unemployment insurance fund. Taking account of the objectives underlying each Act and the facts of the case, it concluded that denial of the application was not required in order to preserve the integrity of the legal system and would be disproportionate to the breach involved in failing to have obtained a work permit.

The American Law Institute’s Restatement (2nd) of Contracts (1981) states at para 178(1):

Nizamuddowlah v Bengal Cabaret Inc (1977) 399 NYS 2d 854 provides a practical example in the case of a devious and oppressive employer. The central defendant (the effective owner of the company named as first defendant) met the plaintiff in Bangladesh and offered to employ him at the defendant’s restaurant in New York City. The plaintiff was to work for an initial period of three months without payment, after which he was to be paid a waiter’s salary. The plaintiff accepted the offer. The defendant arranged for the plaintiff’s travel and entry to the USA on a visitor’s visa, and he also promised to obtain a resident visa or “green card” for him. The plaintiff worked for the defendant for 20 months, but he received no payment despite several demands. He also made repeated inquiries about his green card, but the defendant persistently stalled him. The plaintiff eventually managed to obtain a green card through his own efforts and sued the defendant to recover wages under the Minimum Wage Act. The defendant sought the dismissal of the action on the ground that the contract was illegal.

By working in the USA before he obtained a green card the plaintiff violated the immigration laws, and the judge was not prepared to accept his plea of ignorance, since he was warned in his application form for a visitor’s visa that gainful employment would constitute a breach of his visa conditions. The judge concluded that he had willingly fallen in with the defendant’s proposal because of his strong desire to emigrate to the USA. The judge identified the public harm liable to result from the type of conduct exposed by the case: employment of aliens such as the plaintiff in times of high unemployment deprived citizens and legally admitted aliens of jobs; their employment on substandard terms could depress wages and working conditions; and it could diminish the effectiveness of labour unions.

However, the judge found that the defendant was the main perpetrator, intent on evading and taking advantage of the immigration laws. He said that knowing about the immigration laws, and aware that a party to an illegal contract could not ask a court to help him to carry out his illegal objective, the defendant ran his enterprise without fairly compensating his employees. The judge concluded that the equitable course was that the plaintiff should be paid on the basis of unjust enrichment, and he calculated the amount of the award by reference to the statutory minimum wage.

The New York Supreme Court, Appellate Division, upheld the judgment at (1979) 415 NYS 2d 685. Observing that the Minimum Wage Act contained no indication of a legislative intent to protect only American workers, the court said:

The Court of Appeal supported and followed the approach of the Law Commission in Les Laboratoires Servier v Apotex Inc [2012] EWCA Civ 593, [2013] Bus LR 80 and ParkingEye Ltd v Somerfield Stores Ltd [2013] QB 840. In the latter case ParkingEye contracted to provide a system of automated monitoring of car parking at Somerfield’s supermarkets. The system recorded vehicle registration numbers and customers would be charged for staying beyond a set period. The contract was to be for an initial term of 15 months and ParkingEye’s remuneration was to come from the charges levied over that period. Overstayers were to be sent letters of demand in a standard form agreed between the parties in advance of the conclusion of the contract. If the first demand did not result in payment, it was to be followed by a series of further demands in stronger terms. The third pro forma letter was deceptive because it falsely represented that ParkingEye had the authority and intention to issue proceedings against the customer if payment was not made within a stipulated period. Six months into the contract Somerfield repudiated it for reasons unconnected with the letters of demand. By that time the monitoring system had been installed at 17 of its stores. ParkingEye’s claim for damages was met with a defence which included a plea of illegality based on the intended use of deception in the performance of the contract.

The trial judge rejected the defence and awarded ParkingEye damages of £350,000 for loss of profits caused by Somerfield’s repudiatory breach. The Court of Appeal upheld his decision. The legally objectionable letter was only a small part of the intended performance of the contract and was not essential to it. The judge had found that ParkingEye did not appreciate that the letter would be legally objectionable when the parties agreed on its form, and that, if someone had pointed the matter out, the letter would have been changed. When its objectionable nature occurred to Somerfield, the proper and reasonable course would have been for Somerfield to raise the matter with ParkingEye and continue to honour the contract, so long as ParkingEye made the necessary alteration and performed the contract in a lawful manner, as it would have done. The court held that denial of ParkingEye’s claim was not justified by the policies underlying the doctrine of illegality and would have led to a disproportionate result.

In that case I said at paras 52-53:

On the relevance of ParkingEye’s state of mind, I referred at para 66 to Waugh v Morris (1873) LR 8 QB 202. The case arose from a charter party under which a cargo of hay was to be shipped from Trouville to London. On arrival in London the master learned that a few months before the conclusion of the contract an order had been published under the Contagious Diseases (Animals) Act 1869 making it illegal to land hay brought from France. The master refrained from landing the cargo and, after some delay, the charterer transhipped and exported it. Meanwhile the contractual laydays had expired and the owner claimed for detention. The charterer resisted the claim on the ground that the contract was void for illegality, because its purpose was the delivery of the consignment to London, which was prohibited by law. The defence was rejected.

Giving the judgment of the court, Blackburn J said that all that the owner had bargained for was that on the ship’s arrival in London the freight should be paid and the cargo unloaded. He contemplated that it would be landed and thought that this would be legal; but if he had thought of the possibility of the landing being prohibited, he would probably and rightly have expected that the charterer would not violate the law. Blackburn J said at 208:

Since the decisions of the Court of Appeal in Les Laboratoires Servier v Apotex Inc and the ParkingEye case, there have been three decisions by the Supreme Court involving the doctrine of illegality. The first was Hounga v Allen [2014] 1 WLR 2889, a case with features similar to Nizamuddowlah v Bengal Cabaret Inc. Miss Hounga was a 14-year old Nigerian. Mr and Mrs Allen offered to employ her as a home help in the UK in return for schooling and £50 per month. With their help she entered the UK on false identity documents and obtained a six months’ visitor’s visa. The plan was masterminded by Mrs Allen’s brother who lived in Lagos. He drafted an affidavit for Miss Hounga to swear, giving her surname as that of Mrs Allen’s mother and a false date of birth. The affidavit led to the issue of a passport in that name. Mrs Allen’s family then arranged for Miss Hounga to be taken to the British High Commission in Lagos, where she produced a document purporting to be an invitation from Mrs Allen’s mother pretending to invite her granddaughter to visit her in the United Kingdom. The High Commission was duped into issuing her with entry clearance. Mrs Allen’s brother then bought a ticket for Miss Hounga to travel to England. On arrival at Heathrow Miss Hounga confirmed to an immigration officer that the purpose of her visit was to stay with her grandmother. Subsequently a psychologist reported that Miss Hounga, who was illiterate, had low cognitive functioning, a learning disability and a developmental age much lower than her chronological age. Nevertheless she knew that she had entered the UK on false pretences, that it was illegal for her to remain beyond six months and that it was illegal for her to take employment in the UK.

After her arrival Miss Hounga lived at the Allens’ home, looking after their children and doing housework. She was not enrolled in a school or paid any wages. She was told by Mrs Allen that if she were found by the police she would be sent to prison. This caused her extreme concern. Mrs Allen also subjected her to serious physical abuse. After 18 months an incident occurred in which Mrs Allen beat Miss Hounga, threw her out of the house and poured water over her. Miss Hounga slept that night in the Allens’ garden in wet clothes. Next day they refused to let her back in, and she made her way to a supermarket car park, where she was found and taken to the social services department of the local authority.

Miss Hounga brought claims against the Allens in the employment tribunal for unfair dismissal, breach of contract and unpaid wages. They were dismissed on the ground that her contract of employment was unlawful. She appealed unsuccessfully to the appeal tribunal and she did not seek to appeal further. Neither the Court of Appeal nor the Supreme Court therefore had occasion to consider whether she was entitled to be paid for the services which she rendered on a quantum meruit (by analogy with cases such as Mohamed v Alaga & Co and Nizamuddowlah v Bengal Cabaret Inc et al).

Miss Hounga also claimed to have been the victim of the statutory tort of unlawful discrimination under the Race Relations Act 1976, section 4(2)(c), in relation to her dismissal. The tribunal found that she had been dismissed because of her vulnerability consequent upon her immigration status. She was therefore the victim of unlawful discrimination and she was awarded compensation for her resulting injury to feelings. The tribunal’s order was set aside by the Court of Appeal, which held that the claim was tainted by the illegal nature of her employment and that for the court to uphold it would be to condone the illegality, but it was restored by the Supreme Court. The leading judgment was given by Lord Wilson, with whom Lady Hale and Lord Kerr agreed.

Lord Wilson did not consider that the solution of the case lay either in asking whether Miss Allen needed to rely on an illegal contract or in asking whether there was an inextricable link between the illegality to which she was a party and her claim. At the heart of the judgment Lord Wilson set out his approach in para 42:

On the first question, drawing on the judgment of McLachlin J in Hall v Hebert, Lord Wilson addressed the policy consideration of preserving the integrity of the legal system and not allowing persons to profit from their illegal conduct. He concluded that an award of compensation for damage to Miss Hounga’s feelings was not a form of profit from her employment; it did not permit evasion of a penalty prescribed by the criminal law; and it did not compromise the integrity of the legal system. Conversely, he said that application of the defence could encourage those in the situation of Mrs Allen to believe that they could discriminate against people like Miss Hounga with impunity and could thereby compromise the integrity of the legal system. On the second question, Lord Wilson said that the Court of Appeal’s decision ran strikingly counter to the public policy against forms of people trafficking and in favour of the protection of its victims. Weighing the policy considerations, he concluded that insofar as any public policy existed in favour of applying the illegality defence, it should give way to the public policy to which its application would be an affront.

Hounga v Allen was a case in tort, but Lord Wilson’s approach to the illegality defence was applied by the Court of Appeal in R (Best) v Chief Land Registrar [2016] QB 23, where the issue was whether a claim to be registered under the Land Registration Act 2002 (“LRA”) as the proprietor of a residential building by adverse possession was barred by illegality. The circumstances were that part of the relevant period of possession involved the commission of trespass which constituted a criminal offence under section 144 of the Legal Aid, Sentencing and Punishment of Offenders Act 2012 (“LASPOA”).

Sales LJ (with whom McCombe LJ agreed) expressed the view, at para 51, that the best guidance on the relevant analytical framework was to be found in Lord Wilson’s judgment (from which he quoted para 42 and the passage which followed it). Applying that guidance, he examined the public policy considerations underlying the provisions of the LRA governing acquisition of title to land and the public policy considerations underlying section 144 of LASPOA. He concluded that the mischief at which section 144 was aimed was far removed from the intended operation of the law of adverse possession and that public policy did not preclude the claim for registration.

After Hounga v Allen came the decision of the Supreme Court in Les Laboratoires Servier v Apotex Inc [2015] AC 430. The issue of illegality arose in the context of a claim to enforce a cross-undertaking in damages given as a condition of an interlocutory injunction in proceedings which ultimately failed. The claim was therefore akin to a claim in contract. The facts were somewhat complicated but do not matter for present purposes. The court held unanimously that the Court of Appeal had reached the right result, but the majority of this court expressed the view, at para 21, that the Court of Appeal’s decision could not possibly be justified by the considerations put forward by that court, which had in broad terms followed the approach commended by the Law Commission. I expressed a different view, at para 62, observing that the Court of Appeal had adopted a similar approach to that taken by this court in Hounga v Allen.

After Les Laboratoires Servier v Apotex Inc came Bilta (UK) Ltd v Nazir (No 2) [2016] AC 1. There was a sharp division of opinion about the proper approach to the defence illegality between, on the one hand, a strictly rule-based approach and, on the other hand, a more flexible approach by which the court would look at the policies underlying the doctrine and decide whether they militated in favour of the defence, taking into account a range of potentially relevant factors. The majority did not consider it necessary to resolve the difference in that case, since it did not affect the result, but Lord Neuberger said at para 15 that it needed to be addressed as soon as appropriately possible.

In his Restatement of the English Law of Contract (Oxford University Press, 2016), pp 221-222, Professor Andrew Burrows explained the difficulty of attempting to state the law in relation to illegality:

Since the law was at a crossroads, Professor Burrows set out alternative possible formulations of a “rule-based approach” and a “range of factors approach”.

One possible version of a rule-based approach, at p 224, which Tinsley v Milligan and Les Laboratoires Servier v Apotex Inc could be interpreted as supporting, would be a single master rule based on reliance:

An alternative rule-based formulation, at p 225, saw the reliance rule as only one of a number of rules and essentially confined to the creation of property rights. On this approach a formulation of the rules might be:

Professor Burrows identified six criticisms of those rules and, more generally, of a “rule-based” approach to illegality.

First, the difficulty with the Tinsley v Milligan reliance rule, whether as a master rule or as a rule restricted to cases involving the assertion of proprietary rights, was that it could produce different results according to procedural technicality which had nothing to do with the underlying policies. The decision of the Court of Appeal in Collier v Collier [2002] EWCA 1095; [2002] BPIR 1057 provides a good illustration. A father granted a lease of property to his daughter to hold on trust for him in order to deceive creditors. His claim to beneficial title was rejected on the ground of illegality, because it was held that he needed to rely on the illegal purpose in order to rebut the presumption of illegality which arose in favour of the daughter. Mance LJ considered at paras 105-106 what appeared to be the distinction introduced by Tinsley v Milligan between a beneficial interest which could be established by “some objectively provable and apparently neutral fact” and a beneficial interest arising only from an agreement made for an unlawful purpose. He described the effect as “little more than cosmetic” where the court was perfectly well aware of the close involvement of both parties in the illegality. Tempted as he was to adopt a severely limited view of the meaning of reliance (encouraged by the judgment of Dawson J in Nelson v Nelson), he rightly did not consider that it was open to the Court of Appeal on the authorities to do so. He expressed strong sympathy with the criticisms of the law expressed by the Law Commission, and he concluded at para 113 that he had no liking for the result which the court was compelled to reach.

Second, the difficulties with rule 1 were illustrated by the ParkingEye case. The illegality in that case went to the contract as formed, because from the outset it was intended to send out to customers a form of letter of demand which contained some deliberate inaccuracies. The rule as stated did not permit differentiation between minor and serious illegality or between peripheral and central illegality. To have deprived ParkingEye of what would otherwise have been a contractual entitlement to damages of £350,000 would have been disproportionate. Moreover, as Sir Robin Jacob pointed out in that case, at paras 33-34, there was something odd about a rule which differentiated according to whether the intention was formed before or after the contract was made.

Third, as with the criticism of rule 1, the reference in rule 2 to performance that involved illegal conduct drew no distinction between serious criminality and relatively minor breach of a statutory regulation.

Fourth, although a purported advantage of firm rules is greater certainty, the cases do not always fit the rules because courts have often sought ways around them when they do not like the consequence. The flexible approach would not only produce more acceptable results, but would in practice be no less certain than the rule-based approach.

Fifth, although Lord Mansfield made it clear in Holman v Johnson that the illegality defence operates as a rule of public policy and is not designed to achieve justice between the parties, that does not mean that any result, however arbitrary, is acceptable. The law should strive for the most desirable policy outcome, and it may be that it is best achieved by taking into account a range of factors.

Sixth, although it may be argued that if there are deficiencies in the traditional rules, the way forward is to refine the rules to remove the deficiencies by appropriate exceptions, that task is one which has never been satisfactorily accomplished. The reason is that there are so many variables, for example, in seriousness of the illegality, the knowledge and intentions of the parties, the centrality of the illegality, the effect of denying the defence and the sanctions which the law already imposes. To reach the best result in terms of policy, the judges need to have the flexibility to consider and weigh a range of factors in the light of the facts of the particular case before them.

If a “range of factors” approach were preferred, Professor Burrows suggested, at pp 229-230, that a possible formulation would read as follows:

The reference to what is an “appropriate response” brings to the surface the moral dimension underlying the doctrine of illegality, which inevitably influences the minds of judges and peeps out in their judgments from time to time. Tinsley v Milligan caused disquiet to Lord Goff and others precisely because its reasoning jarred with their sense of what was just and appropriate.

In Yarmouth v France (1887) 19 QBD 647, 653, Lord Esher MR said:

The maxims ex turpi causa and in pari delicto are no exception. It is interesting that, according to Professor JK Grodecki, Lord Mansfield himself was “conscious that if the brocard in pari delicto was to be a beneficial rule of jurisprudence it should not be allowed to become rigid and inflexible”: In pari delicto potior est conditio defendentis (1955) 71 LQR 254, 258. Professor Grodecki gave examples including Smith v Bromley (1760) 2 Doug KB 696n; 99 ER 441 and Walker v Chapman (1773) Lofft 342, 98 ER 684.

In Smith v Bromley (the earliest case in which the maxim in pari delicto appears to have been used) Lord Mansfield granted recovery to the plaintiff of money paid by the plaintiff to procure her brother’s discharge from bankruptcy, which was an illegal consideration. As he explained, Lord Mansfield, at p 698, regarded it as in the public interest that the plaintiff should be repaid notwithstanding the illegal purpose of the payment:

In Walker v Chapman the defendant, who was a page to the King, offered to take a bribe of £50 from the plaintiff in return for securing him a place in the Customs. The bribe was paid but the plaintiff did not obtain the appointment and so he sued for the return of his money. It was argued for the defendant that no action would lie, the plaintiff being party to an iniquitous contract, and that the law would not suffer a party to “draw justice from a foul fountain”. Lord Mansfield rejected the defence, distinguishing between a claim to overturn an illegal contract and a claim to obtain benefit from it. Later judges have taken a different and stricter approach.

Looking behind the maxims, there are two broad discernible policy reasons for the common law doctrine of illegality as a defence to a civil claim. One is that a person should not be allowed to profit from his own wrongdoing. The other, linked, consideration is that the law should be coherent and not self-defeating, condoning illegality by giving with the left hand what it takes with the right hand.

Lord Goff observed in the Spycatcher case, Attorney General v Guardian Newspapers Ltd (No 2) [1990] 1 AC 109, 286, that the “statement that a man shall not be allowed to profit from his own wrong is in very general terms, and does not of itself provide any sure guidance to the solution of a problem in any particular case”. In Hall v Hebert [1993] 2 SCR 159 McLachlin J favoured giving a narrow meaning to profit but, more fundamentally, she expressed the view (at 175-176) that, as a rationale, the statement that a plaintiff will not be allowed to profit from his or her own wrongdoing does not fully explain why particular claims have been rejected, and that it may have the undesirable effect of tempting judges to focus on whether the plaintiff is “getting something” out of the wrongdoing, rather than on the question whether allowing recovery for something which was illegal would produce inconsistency and disharmony in the law, and so cause damage to the integrity of the legal system.

That is a valuable insight, with which I agree. I agree also with Professor Burrows’ observation that this expression leaves open what is meant by inconsistency (or disharmony) in a particular case, but I do not see this as a weakness. It is not a matter which can be determined mechanistically. So how is the court to determine the matter if not by some mechanistic process? In answer to that question I would say that one cannot judge whether allowing a claim which is in some way tainted by illegality would be contrary to the public interest, because it would be harmful to the integrity of the legal system, without a) considering the underlying purpose of the prohibition which has been transgressed, b) considering conversely any other relevant public policies which may be rendered ineffective or less effective by denial of the claim, and c) keeping in mind the possibility of overkill unless the law is applied with a due sense of proportionality. We are, after all, in the area of public policy. That trio of necessary considerations can be found in the case law.

The relevance of taking into account the purpose of the relevant prohibition is self-evident. The importance of taking account of the relevant statutory context is illustrated by Hardy v Motor Insurers’ Bureau [1964] 2 QB 745. The Road Traffic Act 1960 required a motorist to be insured against the risk of causing death or personal injury through the use of a vehicle on a road, but a line of authorities established that a contract to indemnify a person against the consequences of a deliberate criminal act is unenforceable. The plaintiff, a security officer at a factory, was injured when he was trying to question the driver of a van, who drove off at speed and dragged him along the road. The driver was convicted of unlawfully causing grievous bodily harm. The driver being uninsured, the plaintiff sued the defendant under an agreement between the defendant and the Minister of Transport, by which the defendant agreed to satisfy any judgment against a motorist for a liability required to be covered under a motor insurance policy. The defendant relied on the maxim ex turpi causa, arguing that a contract purporting to insure the driver against his own deliberate criminal conduct would have been unlawful. The defence was rejected. Diplock LJ said at p 767:

Hounga v Allen and R (Best) v Chief Land Registrar are illustrations of cases in which there were countervailing public interest considerations, which needed to be balanced.

As to the dangers of overkill, Lord Wright gave a salutary warning in Vita Food Products Inc v Unus Shipping Co Ltd [1939] AC 277, 293:

To similar effect Devlin J questioned “whether public policy is well served by driving from the seat of judgment everyone who has been guilty of a minor transgression” in St John Shipping Corpn v Joseph Rank Ltd [1957] 1 QB 267, 288-289.

In Saunders v Edwards [1987] 1 WLR 1116, 1134, Bingham LJ said

In considering whether it would be disproportionate to refuse relief to which the claimant would otherwise be entitled, as a matter of public policy, various factors may be relevant. Professor Burrows’ list is helpful but I would not attempt to lay down a prescriptive or definitive list because of the infinite possible variety of cases. Potentially relevant factors include the seriousness of the conduct, its centrality to the contract, whether it was intentional and whether there was marked disparity in the parties’ respective culpability.

The integrity and harmony of the law permit - and I would say require - such flexibility. Part of the harmony of the law is its division of responsibility between the criminal and civil courts and tribunals. Punishment for wrongdoing is the responsibility of the criminal courts and, in some instances, statutory regulators. It should also be noted that under the Proceeds of Crime Act 2002 the state has wide powers to confiscate proceeds of crime, whether on a conviction or without a conviction. Punishment is not generally the function of the civil courts, which are concerned with determining private rights and obligations. The broad principle is not in doubt that the public interest requires that the civil courts should not undermine the effectiveness of the criminal law; but nor should they impose what would amount in substance to an additional penalty disproportionate to the nature and seriousness of any wrongdoing. ParkingEye is a good example of a case where denial of claim would have been disproportionate. The claimant did not set out to break the law. If it had realised that the letters which it was proposing to send were legally objectionable, the text would have been changed. The illegality did not affect the main performance of the contract. Denial of the claim would have given the defendant a very substantial unjust reward. Respect for the integrity of the justice system is not enhanced if it appears to produce results which are arbitrary, unjust or disproportionate.

The courts must obviously abide by the terms of any statute, but I conclude that it is right for a court which is considering the application of the common law doctrine of illegality to have regard to the policy factors involved and to the nature and circumstances of the illegal conduct in determining whether the public interest in preserving the integrity of the justice system should result in denial of the relief claimed. I put it in that way rather than whether the contract should be regarded as tainted by illegality, because the question is whether the relief claimed should be granted.

I agree with the criticisms made in Nelson v Nelson and by academic commentators of the reliance rule as laid down in Bowmakers and Tinsley v Milligan, and I would hold that it should no longer be followed. Unless a statute provides otherwise (expressly or by necessary implication), property can pass under a transaction which is illegal as a contract: Singh v Ali [1960] AC 167, 176, and Sharma v Simposh Ltd [2013] Ch 23, paras 27-44. There may be circumstances in which a court will refuse to lend its assistance to an owner to enforce his title as, for example, where to do so would be to assist the claimant in a drug trafficking operation, but the outcome should not depend on a procedural question.

In Bowmakers [1945] 1 KB 65 the claim was for conversion of goods which had been obtained by the plaintiffs and supplied to the defendant under transactions which were assumed to be tainted by illegality. The Court of Appeal rightly said, at p 71, that “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”, but it added the qualifying words “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”. The objections to the proviso have already been identified. It makes the question whether the court will refuse its assistance to the claimant to enforce his title to his property depend on a procedural question and it has led to uncertain case law about what constitutes reliance. The court ended its judgment, at p 72, by saying:

In Tinsley v Milligan, even if Miss Milligan had not owned up and come to terms with the DSS, it would have been disproportionate to have prevented her from enforcing her equitable interest in the property and conversely to have left Miss Tinsley unjustly enriched.

Critics of the “range of factors” approach say that it would create unacceptable uncertainty. I would make three points in reply. First, one of the principal criticisms of the law has been its uncertainty and unpredictability. Doctrinally it is riven with uncertainties: see, for example, paras 4-8 above. There is also uncertainty how a court will in practice steer its way in order to reach what appears to be a just and reasonable result. Second, I am not aware of evidence that uncertainty has been a source of serious problems in those jurisdictions which have taken a relatively flexible approach. Third, there are areas in which certainty is particularly important. Ordinary citizens and businesses enter into all sorts of everyday lawful activities which are governed by well understood rules of law. Lord Mansfield said in Vallejo v Wheeler (1774) 1 Cowp 143, 153:

In Tinsley v Milligan Lord Goff considered that if the law was to move in a more flexible direction, to which he was not opposed in principle, there should be a full investigation by the Law Commission (which has happened) and that any reform should be through legislation. Realistically, the prospect of legislation can be ignored. The government declined to take forward the Commission’s bill on trusts because it was not seen to be “a pressing priority for government” (a phrase familiar to the Commission), and there is no reason for optimism that it would take a different view if presented with a wider bill. In Clayton v The Queen (2006) 231 ALR 500, para 119, Kirby J said that waiting for a modern Parliament to grapple with issues of law reform is like “waiting for the Greek Kalends. It will not happen” and that “Eventually courts must accept this and shoulder their own responsibility for the state of the common law”. The responsibility of the courts for dealing with defects in the common law was recently emphasised by this court in R v Jogee [2016] 2 WLR 681, para 85, and Knauer v Ministry of Justice [2016] 2 WLR 672, para 26. In each of those cases the court decided that it should depart from previous decisions of the House of Lords. That is never a step taken lightly. In departing from Tinsley v Milligan it is material that it has been widely criticised; that people cannot be said to have entered into lawful transactions in reliance on the law as then stated; and, most fundamentally, that the criticisms are well founded.

In the present case I would endorse the approach and conclusion of Gloster LJ. She correctly asked herself whether the policy underlying the rule which made the contract between Mr Patel and Mr Mirza illegal would be stultified if Mr Patel’s claim in unjust enrichment were allowed. After examining the policy underlying the statutory provisions about insider dealing, she concluded that there was no logical basis why considerations of public policy should require Mr Patel to forfeit the moneys which he paid into Mr Mirza’s account, and which were never used for the purpose for which they were paid. She said that such a result would not be a just and proportionate response to the illegality. I agree. It seems likely that Lord Mansfield would also have agreed: see Walker v Chapman. Mr Patel is seeking to unwind the arrangement, not to profit from it.

It is not necessary to discuss the question of locus poenitentiae which troubled the courts below, as it has troubled other courts, because it assumed importance only because of a wrong approach to the issue whether Mr Patel was prima facie entitled to the recovery of his money. In place of the basic rule and limited exceptions to which I referred at para 44 above, I would hold that a person who satisfies the ordinary requirements of a claim in unjust enrichment will not prima facie be debarred from recovering money paid or property transferred by reason of the fact that the consideration which has failed was an unlawful consideration. I do not exclude the possibility that there may be particular reason for the court to refuse its assistance to the claimant, applying the kind of exercise which Gloster LJ applied in this case, just as there may be a particular reason for the court to refuse to assist an owner to enforce his title to property, but such cases are likely to be rare. (At para 110 I gave the example of a drug trafficker.) In Tappenden v Randall (1801) 2 Bos & Pul 467, 471, 126 ER 1388, 1390, a case of a successful claim for the repayment of money paid for an unenforceable consideration which failed, Heath J said obiter that there might be “cases where the contract may be of a nature too grossly immoral for the court to enter into any discussion of it: as where one man has paid money by way of hire to another to murder a third person”. The case was mentioned by the Law Commission (LCCP 189, para 4.53), but there is a dearth of later case law on the point. This is hardly surprising because a person who takes out a contract on the life of a third person is not likely to advertise his guilt by suing. But as a matter of legal analysis it is sufficient for present purposes to identify the framework within which such an issue may be decided. No particular reason has been advanced in this case to justify Mr Mirza’s retention of the monies beyond the fact that it was paid to him for the unlawful purpose of placing an insider bet.

In support of his argument that this purpose was sufficient to disentitle Mr Patel from obtaining the return of his money, Mr Collings relied on cases such as Parkinson v College of Ambulance Ltd [1925] 2 KB 1. In that case the plaintiff made a donation to a charity to secure a knighthood. When the honour failed to materialise he sued for the return of his money. The claim was rejected.

Bribes of all kinds are odious and corrupting, but it does not follow that it is in the public interest to prevent their repayment. There are two sides to the equation. If today it transpired that a bribe had been paid to a political party, a charity or a holder of public office, it might be regarded it as more repugnant to the public interest that the recipient should keep it than that it should be returned. We are not directly concerned with such a case but I refer to it because of the reliance placed on that line of authorities.

Since criticism was made of the Court of Appeal’s decision in Mohamed v Alaga and Co, I would affirm its correctness and reject the view that it should somehow be confined to its own peculiar facts. With hindsight, it is perhaps unfortunate that this court did not have the opportunity of considering a claim by Miss Hounga for a quantum meruit.

The essential rationale of the illegality doctrine is that it would be contrary to the public interest to enforce a claim if to do so would be harmful to the integrity of the legal system (or, possibly, certain aspects of public morality, the boundaries of which have never been made entirely clear and which do not arise for consideration in this case). In assessing whether the public interest would be harmed in that way, it is necessary a) to consider the underlying purpose of the prohibition which has been transgressed and whether that purpose will be enhanced by denial of the claim, b) to consider any other relevant public policy on which the denial of the claim may have an impact and c) to consider whether denial of the claim would be a proportionate response to the illegality, bearing in mind that punishment is a matter for the criminal courts. Within that framework, various factors may be relevant, but it would be a mistake to suggest that the court is free to decide a case in an undisciplined way. The public interest is best served by a principled and transparent assessment of the considerations identified, rather by than the application of a formal approach capable of producing results which may appear arbitrary, unjust or disproportionate.

A claimant, such as Mr Patel, who satisfies the ordinary requirements of a claim for unjust enrichment, should not be debarred from enforcing his claim by reason only of the fact that the money which he seeks to recover was paid for an unlawful purpose. There may be rare cases where for some particular reason the enforcement of such a claim might be regarded as undermining the integrity of the justice system, but there are no such circumstances in this case. I would dismiss the appeal.

For the reasons given by Lord Toulson, with which I completely agree, I consider that this appeal should be dismissed.

The approach commended by Lord Toulson does not involve engaging with “an open and unsettled range of factors” - Lord Mance at para 192 of his judgment. On the contrary, as I see it, Lord Toulson’s judgment outlines a structured approach to a hitherto intractable problem. It is an approach, moreover, which, if properly applied, will promote, rather than detract from, consistency in the law. And it has the added advantage of avoiding the need to devise piecemeal and contrived exceptions to previous formulations of the illegality rule.

Central to Lord Toulson’s analysis is the “trio of considerations” which he identified in para 101 of his judgment. The first of these involves an examination of the underlying purpose of the “prohibition which has been transgressed”. By this, I understand Lord Toulson to mean the reasons that a claimant’s conduct should operate to bar him or her from a remedy which would otherwise be available. That such reasons should be subject to scrutiny is surely unexceptionable. Whether in order to preserve “the integrity of the legal system” (per McLachlin J in Hall v Hebert [1993] 2 SCR 159 at 169) or to allow a proper understanding of the true nature of the public policy imperative for recognising a defence of illegality, the purpose of the denial of a remedy to which the claimant would otherwise be entitled should be clearly understood.

As it happens, McLachlin J disagreed with Cory J’s suggestion that the doctrine of ex turpi causa non oritur actio should be replaced with a power vested in the courts to reject claims on “considerations of public policy” - p 168. But what is the preservation of the integrity of the legal system, if not a public policy consideration? Moreover, the underpinning of the preservation of that integrity (which McLachlin J said was that a person in a civil suit should not be permitted “to profit from illegal or wrongful conduct” or to benefit from “an evasion or rebate of a penalty prescribed by the criminal law”) is par excellence a public policy consideration. And McLachlin J seemed to acknowledge as much when she said (at p 169) that the principle could be described “by an old-fashioned Latin name or by the currently fashionable concept of ‘public policy’”.

It is doubtful that a public policy consideration in the context of the defence of illegality could now be properly described as a “currently fashionable” concept. Indeed, in a number of cases that I will refer to briefly below, the maxim ex turpi causa has been recognised in this country as an expression of policy, rather than a principle. And in Canada it appears to be accepted that the weighing of public policies is the proper approach to take in order to determine whether a defence of illegality should be allowed to prevail. In Still v Minister of National Revenue (1997) 154 DLR (4th) 229 (which is discussed by Lord Toulson in paras 58 et seq of his judgment) the Federal Court of Appeal considered that the doctrine of illegality now “rests on the understanding that it would be contrary to public policy to allow a person to maintain an action on a contract prohibited by statute” (emphasis supplied). On that basis, Robertson JA, who delivered the judgment of the court, said that it was necessary to identify the policy considerations which outweighed the applicant’s right to a remedy. Although this was said in relation to competing policy goals in two items of legislation, there is no reason not to adopt the same approach in evaluating rival policy considerations in the non-statutory context.

To take this case as an example, why should Mr Mirza’s wrongful retention of Mr Patel’s money not be weighed against the undoubted illegality on the part of Mr Patel in entering an agreement to wrongly benefit from Mr Mirza’s claimed ability to obtain access to insider information? If one concentrates on the illegal nature of the contract to the exclusion of other considerations, an incongruous result in legal and moral terms may be produced. This can be avoided by taking into account and giving due weight to the second and third of Lord Toulson’s considerations viz countervailing public policies which would be wrongly discounted by denial of the claim and the proportionality of refusing to acknowledge its legitimacy.

It is, of course, possible to reach the same outcome that a weighing of the competing policy considerations produces by treating this case as one of unjust enrichment which warranted returning the parties to the position that they occupied before the transaction. This is on the basis that the court is not required to give effect to the illegal contract in order to find that Mr Mirza should not be allowed to retain Mr Patel’s money. It would “simply return the parties to the status quo ante where they should always have been.” - Lord Sumption at para 268. That seems to me, however, to be a much more adventitious and less satisfactory route to the proper disposal of the case than that represented by a rounded assessment of the various public policy considerations at stake.

Moreover, if the ex turpi causa axiom is itself no more than an expression of policy, the taking into account of countervailing policy considerations, in order to decide whether to give effect to it in a particular instance, is the only logical way to proceed. That it is, in truth, a policy based rule has been clearly recognised. In Gray v Thames Trains Ltd [2009] UKHL 33; [2009] AC 1339, para 30, Lord Hoffmann said that the maxim expressed, “not so much a principle as a policy” and that it did not have a single basis of justification but was rather based on “a group of reasons which vary in different situations”. And in Stone and Rolls Ltd v Moore Stephens [2009] UKHL 39; [2009] AC 1391, para 25 Lord Phillips expressly endorsed what Lord Hoffmann had said about the public policy nature of ex turpi causa, observing that it was necessary to consider the policy underlying it, in order to decide whether the defence of illegality was bound to defeat a claim.

Finally, in Les Laboratoires Servier v Apotex Inc [2012] EWCA Civ 593; [2013] Bus LR 80, after referring (in para 66) to the Law Commission’s recommendation in its 2010 Report (Law Com 320) to the effect that the illegality offence should be allowed where its application could be firmly justified by one or more of the rationales underlying it existence, Etherton LJ said, at para 73:

Lord Sumption has said in para 262(iii) of his judgment in this case that this court in Les Laboratoires Servier [2015] AC 430 had overruled the view expressed by the Court of Appeal that “an illegal act might nevertheless found a cause of action if it was not as wicked as all that”. That may be so, but I do not understand the judgment of this court in Les Laboratoires Servier to have expressly rejected the notion that whether the defence should be available depends on an examination of the policy considerations which underlie it in any particular instance and those which militate against it. At para 61 of his judgment in Les Laboratoires Servier Lord Toulson quoted with approval the statement of Lord Wilson in Hounga v Allen [2014] 1 WLR 2889 at para 42 to the effect that, in considering whether to allow a defence of illegality, “it is necessary, first, to ask ‘What is the aspect of public policy which founds the defence?’ and, second, to ask ‘But is there another aspect of public policy to which application of the defence would run counter?’” The decision in Hounga was not mentioned in the judgment of the majority in Les Laboratoires Servier.

Lord Sumption did refer to Hounga, however, in the later case of Bilta (UK) Ltd v Nazir (No 2) [2016] AC 1. He sought to explain the decision in Hounga on the basis that Ms Hounga did not rely, and did not need to rely, on the circumstances in which she had entered the United Kingdom (she had entered illegally). This is correct but she did need to rely on the fact of her employment in advancing a claim for unlawful discrimination in her dismissal from that employment. Since the employment was not legally sanctioned, she was therefore confronted with the illegality defence and, indeed, the Court of Appeal had held that the illegality of the contract of employment formed a material part of Ms Hounga’s complaint and that to uphold it would be to condone the illegality. It was held in Hounga that the appellant’s claim was not inextricably linked to her illegal conduct. On that account her action could not be defeated on the basis that her contract of employment was illegal. But Lord Wilson’s discussion of the manner in which competing public policy considerations should be viewed, in calculating whether a defence of illegality should be permitted to defeat an otherwise viable claim, unquestionably forms part of the ratio of the decision.

The way is now open for this court to make its choice between, on the one hand, cleaving to the rule-based approach exemplified by Tinsley v Milligan [1994] 1 AC 340 and, arguably, the decision of the majority in Les Laboratoires Servier, and, on the other, a more flexible approach, taking into account the policy considerations that are said to favour recognising the defence of illegality, those which militate against such recognition and the proportionality of allowing the defence to prevail. In Bilta (UK) Ltd Lord Neuberger said that the proper approach to the defence of illegality needed to be addressed by this court “as soon as appropriately possible” - para 15. This case unmistakably presents us with the opportunity to address the question and for the reasons given by Lord Toulson, I believe that the approach which he commends is plainly to be preferred.

A rule-based approach to the question of the effect of illegality on the availability of a remedy has failed to deliver on what some have claimed to be its principal virtues viz ease of application and predictability of outcome. This case exemplifies the point. There was a sharp but perfectly respectable difference of view in the judgments of the Court of Appeal as to whether the necessary ingredient of reliance on the illegal aspect of the agreement between Mr Mirza and Mr Patel was present. This is hardly surprising. In many situations in which transactions between parties are tainted by some form of illegality, it is not always easy to decide what it is that needs to be relied on when an unravelling of those transactions or some means of dealing with their failure is sought.

On the question of unravelling or unpicking an agreement, I do not consider that Tinsley is an example of the court conducting an unravelling exercise or of its returning the parties to the status quo ante. This much is clear from the speech of Lord Browne-Wilkinson at 376F of the report:

In effect, in Tinsley the majority gave effect to rather than unravelled the illegal agreement made between the parties. The agreement was that the ownership of the house should be shared equally between Miss Milligan and Miss Tinsley, and that they should represent to the Department of Social Security that it was owned solely by Miss Tinsley. It was because Miss Milligan did not need to rely on the illegal component of the agreement (that they make the false representation to the department) that she was able to succeed. This was not, therefore, a case of unravelling the agreement or restoring the parties to the status quo ante. To the contrary, it was an instance of segregating the illegal part of the agreement from that which, it was considered, could be enforced. Reference to or reliance on the objectionable part could thereby be avoided. To claim that such a contrivance produces a predictable, much less a certain, outcome, for such arrangements is, I believe, extremely far-fetched.

Even if the claim to predictability of outcome for the reliance test could be made good, however, it is questionable whether particular weight should be given to this consideration in circumstances where a claimant and defendant have been parties to an agreement which is plainly illegal. Certainty or predictability of outcome may be a laudable aim for those who seek the law’s resolution of genuine, honest disputes. It is not a premium to which those engaged in disreputable conduct can claim automatic entitlement. For the reasons I have given, however, I do not believe that outcomes are easier to forecast on a rule-based approach.

Quite apart from the difficulty in predicting whether a claimant has to rely on the illegal dimension of an agreement in order to advance his claim, there is something unattractive and contrived about the means by which attempts have to be made in order to avoid the spectre of reliance. Professor Burrows in his Restatement of the English Law of Contract (Oxford University Press) outlined what he described as his single reliance master rule at p 224 in this way:

In this case the formation of the contract, its purpose and its performance all involved illegality. Under the single reliance master rule, it is said that all of this can be ignored because it is not necessary to rely on the terms of the agreement, other than to demonstrate that there was no legal basis for the payment of the money to Mr Mirza. So, the looming presence of illegality does not require to be confronted at all. The issue is side-stepped and avoided. This cannot be the correct way in which to deal with the impact of illegality - in fact, under this approach, illegality is not addressed at all. It is surely better and more principled to examine why illegality should or should not operate to deny Mr Patel a remedy.

Returning the parties to the status quo ante likewise side-steps the issue of illegality. This approach proceeds on the basis that the transaction should simply never have taken place or that the parties should be returned to the condition that they ought always to have occupied. The contract is unpicked because it should not have been made. Mr Mirza is deprived of the money because it is unjust enrichment. No examination of the effect that the illegality has is warranted; recognition that there has been unjust enrichment is all that is required.

This is objectionable not only because it effectively ignores the illegality that surrounded the making of the contract but also because it produces an inconsistent result with that which is founded on a breach of contract claim. This leads to what Professor Peter Birks, in an article entitled, “Recovering Value Transferred under an Illegal Contract” (2000) 1 TIL 155, describes as self-stultification. Entitlement to restitution of money paid on foot of an illegal contract on the basis of unjust enrichment makes a nonsense, he says, of refusal to enforce the contract and, at p 160, it is “important that the law as stated in one area should not make nonsense of the law as stated in another”.

Self-stultification can be avoided by adoption of the approach suggested by Lord Toulson. His mode of analysis requires examination of the justification for the defence of illegality in whatever context it arises, not a decision to circumvent the defence because of the type of remedy that is claimed. That appears to me to be a much more principled approach than one which avoids having to engage with the merits of the defence at all. Not having to engage with the merits on the basis that one does not have to rely on the illegality is a matter of fortuity. Because of that incidental circumstance an avenue to an equivalent outcome to that which would result from enforcement of the contract opens up. An examination of the impact of the illegality becomes irrelevant. That this should be a matter of happenstance is deeply unsatisfactory.

Lord Toulson’s solution to this question also permits readier access to investigation of the traditional justifications for the ex turpi causa maxim - preservation of the integrity of the legal system and preventing profit from wrongdoing. If, on examination of the particular circumstances of the case, these can be shown to weigh heavily in the balance, it is more likely that the defence will be upheld. Carving out an exception to the application of the defence on the basis that it does not affect a claim for unjust enrichment where the illegality of the claimant does not require to be relied on does nothing to directly protect or uphold these values.

For these reasons and those given by Lord Toulson, I would dismiss the appeal.

The present appeal concerns a claim for the return of money paid by the claimant to the defendant pursuant to a contract to carry out an illegal activity, and the illegal activity is not in the event proceeded with owing to matters beyond the control of either party.

In such a case, the general rule should in my view be that the claimant is entitled to the return of the money which he has paid. In the first place, such a rule (“the Rule”) is consistent with the law as laid down in the 18th century by two eminent judges, one of whom is regarded as the founder of many aspects of the common law, including illegality; in addition it has support from some more modern cases. Secondly, the Rule appears to me to accord with policy, which is particularly important when illegality arises in the context of a civil claim. Thirdly, the Rule renders the outcome in cases in one area of a very difficult topic, that of contracts involving illegality, and the maxim ex turpi causa non oritur actio (ie that no claim can be based on an illegal or immoral arrangement), relatively clear and certain.

I turn first to the authorities. In Smith v Bromley (1760) 2 Doug KB 696n, the Court of King’s Bench permitted a plaintiff to recover money she had paid to someone who had agreed to procure her brother’s discharge from bankruptcy, which was an illegal consideration. Lord Mansfield CJ said at p 698 in the course of his judgment that, although the payment had been made for an illegal purpose:

In Neville v Wilkinson (1782) 1 Bro CC 543, 547 Lord Thurlow LC approved this approach, and “declared his opinion” that:

In the following century, the same approach was adopted in Taylor v Bowers (1876) 1 QBD 291 (which involved transfer of goods rather than of cash). Cockburn CJ said at first instance at p 295 that it was “well established” that “where money has been paid, or goods delivered, under an unlawful agreement, but there has been no further performance of it”, then “the party paying the money or delivering the goods may repudiate the transaction, and recover back his money or goods”. The Court of Appeal agreed, and at p 300 Mellish LJ, with whom Baggallay JA and Grove J agreed, said this:

It is also fair to say that Fry LJ doubted the correctness of Mellish LJ’s dictum in Kearley v Thomson (1890) 24 QBD 742, 746, and that in some subsequent cases the principle has not been applied. An obvious example is Parkinson v College of Ambulance [1925] 2 KB 1, where a donor was held to be disentitled from recovering a gift to a charity obtained by the charity’s illegal (and dishonest) promise to obtain an honour for the donor. I consider that that case was wrongly decided. It seems to me that the judgment in that case got close to representing what Bingham LJ described as the court “on the first indication of unlawfulness affecting any aspect of a transaction, draw[ing] up its skirts and refus[ing] all assistance to the plaintiff, no matter how serious his loss nor how disproportionate his loss to the unlawfulness of his conduct” (which he considered to be “unacceptable”) - Saunders v Edwards [1987] 1 WLR 1116, 1134. I agree with the view that the decision in Parkinson represented “a new and regrettable extension of the scope of the maxim” of ex turpi causa (to quote from Professor Grodecki’s article (1955) 71 LQR 254, 263), and I consider that it should be overruled.

The Rule also derives some support from the Court of Appeal’s decision in Tribe v Tribe [1996] Ch 107, where the plaintiff was held to be entitled to recover shares which he had transferred to his son in order deceptively to improve his negotiating position in relation to an anticipated claim by his landlord, which in the event did not materialise. The question for the Court of Appeal was whether, following the controversial decision of the House of Lords in Tinsley v Milligan [1994] 1 AC 340 (the details of which are set out in paras 17-19 above), the father could “rebut the presumption of advancement by giving evidence of his illegal purpose”, to quote from Millett LJ’s judgment at pp 129H-130A. It was held that he could, on the basis that “[t]he transferor can lead evidence of the illegal purpose whenever it is necessary for him to do so provided that he has withdrawn from the transaction before the illegal purpose has been wholly or partly carried into effect” - per Millett LJ at pp 134G-H.

There is some support in the cases for the notion that different considerations should apply depending whether the claimant’s claim for return of money or property paid pursuant to an unperformed illegal contract is based on a common law claim or a claim in equity (compare the Privy Council decisions in Singh v Ali [1960] AC 167 and Chettiar v Chettiar [1962] AC 294). I do not consider that such a distinction is appropriate (and it may be that in that connection I differ from Millett LJ in Tribe at p 129G - although see at p 130E). I agree with Lord Browne-Wilkinson’s observation in Tinsley at p 371, where he said that “[i]f the law is that a party is entitled to enforce a property right acquired under an illegal transaction, … the same rule ought to apply to any property right so acquired, whether such right is legal or equitable”.

That proposition is supported, as I see it, by the second reason supporting the Rule, namely policy. As Millett LJ said in Tribe at p 133F, “the justification for this rule [which precludes the court from lending its assistance to a man who founds his cause of action on an illegal or immoral act] is not a principle of justice but a principle of policy”, citing Lord Mansfield in Holman v Johnson (1775) 1 Cowp 341, 343. That approach is also supported by Lord Hoffmann in Gray v Thames Trains Ltd [2009] AC 1339, para 30, where he went on to say that the “policy is not based upon a single justification but on a group of reasons, which vary in different situations”. Similarly, in Bakewell Management Ltd v Brandwood [2004] 2 AC 519, para 60, Lord Walker said that “the maxim ex turpi causa must be applied as an instrument of public policy, and not in circumstances where it does not serve any public interest”.

More broadly, it appears to me that policy supports the Rule, in part for the simple reasons given in the passages cited in para 147 above. Further, as Lord Mance points out, there is obvious attraction in the notion that, if all transfers made pursuant to an unexecuted illegal contract are re-transferred, then the parties are back in the position that they were, ie as if there had been no illegal contract, which again would seem to comply with policy.

It also appears to me that the Rule is consistent with the approach adopted in McLachlin J’s analysis in the Canadian Supreme Court case Hall v Hebert [1993] 2 SCR 159, 176. She explained that the basic justification for refusing relief to a plaintiff who relied on an illegal contract was that “to allow recovery … would be to allow recovery for what is illegal. It would put the courts in the position of saying that the same conduct is both legal, in the sense of being capable of rectification by the court, and illegal. It would, in short, introduce an inconsistency in the law”. Later at pp 179-180, she suggested that the court’s “power” to refuse relief in a claim where illegality is involved “is a limited one” and that the use of the power “is justified where allowing the plaintiff’s claim would introduce inconsistency into the fabric of the law, either by permitting the plaintiff to profit from an illegal or wrongful act, or to evade a penalty prescribed by criminal law”. This approach (which is more fully analysed by Lord Mance) finds an echo in Lord Rodger’s observation in Gray, para 82, that “the civil court should cleave to the same policy as the criminal court”.

For some time, it was assumed that the Rule could be invoked not merely when the anticipated illegal purpose not been proceeded with at all, but with the super-added requirement that it had not done so because of the “repentance” of the plaintiff who was seeking to get his money back. Like Millett LJ in Tribe at p 135D-E, I would reject that notion. As he said, “[j]ustice is not a reward for merit”, and in any event the notion that repentance is needed “could lead to bizarre results”. Further, a claimant’s repentance may be born of, or combined with, self-interest, in which case, if repentance is the essential factor, the court would face a real difficulty. In my opinion, the notion that the application of a rule should depend on whether or not the claimant has repented typifies the inappropriately moralistic approach of some courts when they have to deal with claims based in some way on illegality, which was rightly criticised by Bingham LJ in Saunders [1987] 1 WLR 1116, 1134. Rather, the courts should adopt a more objective and analytical approach like that of McLachlin J in Hall.

Quite apart from principle, it appears to me that the Rule would establish, or maintain, a degree of clarity and certainty in relation to one aspect of the law on the vexed topic of the effect of illegality on contractual claims. One thing which is clear from reading only some of the large number of judgments on the law on that topic over the past 350 years is the inconsistency of reasoning and outcome in different cases. Those responsible for making and developing the law in any area must strive to achieve as much clarity and as much certainty as are consistent with principle and practicality.

There is, I acknowledge, some attraction in the point that the need for certainty in this area is diminished by the fact that parties to an arrangement which is illegal have less cause for complaint if the law is uncertain. However, criminals are entitled to certainty in the law just as much as anyone else. In any event, third parties are often affected by the enforceability of rights acquired or lost under contracts, and innocent third parties, it could be said with force, are in a particularly strong position to expect certainty and clarity from the law. Quite apart from this, there is a general public interest in certainty and clarity in all areas of law, not merely because it is a fundamental aspect of the rule of law, but also because the less clear and certain the law on any particular topic, the more demands there are on the services of the courts.

That leaves open two issues. First, the issue of what constitutes an illegal contract for the purpose of the Rule. In my view, as a general proposition, the rule would apply to any contract where the illegality would result in the court (if it could otherwise do so) not being able to order specific performance of the contract or damages for its breach. It would thus normally apply to any contract whose performance would inevitably involve the commission of a crime (i) because the whole purpose of the contract was the commission of a crime (eg a contract killing), or (ii) because it was a contract whose essential ingredient was the commission of a crime (the present case is an example), or (iii) because the contract could not be performed without the commission of a crime. In practice, of course, category (iii) would cover the other two categories, but setting out the three categories may help to illustrate the extent of the application of the rule.

As to category (iii), I have difficulties in seeing how a court could order specific performance of a contract which necessarily involved one or other of the parties committing a crime (even a minor crime). Requiring the contract to be performed would involve the court ordering a party to act illegally: that cannot be a course open to a court. For the same reason I have difficulties in seeing how a court could normally award damages for breach of such a contract. Conceptually, damages are a substitute for non-performance, and performance is not something the court can award; and it seems inconsistent with the court’s function to penalise a defendant in damages for not doing something illegal or to compensate a claimant for not having a benefit which would have required either or both of the parties doing something illegal. For the court to make an order for specific performance or damages in such cases would seem to infringe the principle of consistency discussed in the judgment of McLachlin J in Hall.

The second issue is foreshadowed by the fact that I have described the Rule as being generally applicable. That is because the need for certainty must, particularly given the importance of policy, yield to the fact that, in this difficult and potentially multi-faceted area, there will inevitably be exceptions. Experience and common sense both suggest that any attempt to lay down a clear and inflexible rule on even one aspect of the topic of illegality in the context of contractual claims will lead to difficulties. (Both the majority and the minority reasoning in Tinsley are a good example of this). Indeed, the very fact that the approach of the courts in cases on this topic is based on policy suggests that strict immutable rules are inappropriate. Nonetheless, that does not negate any attempt to identify principles such as that suggested by McLachlin J and general rules such as that described in the cases mentioned in paras 147 and 148 above. The fact that the approach of the law to contracts with an illegal aspect is based on policy does not discharge judges from the normal duty of ensuring that the law on any topic is as clear and certain as it can be.

By way of example, I would mention two possible exceptions. First, where one of the parties, especially the defendant, is in a class which is intended to be protected by the criminal legislation involved, it may well be inappropriate to invoke the Rule. Secondly, there could well be no recovery (or only partial recovery) by a plaintiff where the defendant was unaware of the facts which gave rise to the illegality - especially if he had received the money and had altered his position so that it might be oppressive to expect him to repay it. There will no doubt be other exceptions, but I do not think that that undermines the usefulness of having the Rule as the prima facie or presumptive approach. (I discuss in paras 172-175 below with the test for determining whether it is appropriate to apply the Rule in any particular case).

In the present case, Mr Patel paid £620,000 to Mr Mirza pursuant to a contract, under which Mr Mirza was to use the money to trade in RBS shares with the benefit of inside information for their common benefit. That was a contract whose agreed fundamental purpose was illegal. In fact, the anticipated inside information was not forthcoming and the contract effectively lapsed. I can see no good reason on these simple facts for not applying the Rule and accordingly I consider that Mr Patel is entitled to the return of the £620,000.

The majority, and indeed Lord Mance and Lord Sumption, would go wider in their judgments on this appeal, by laying down some wider and more general principles or rules relating to the effect of illegality on contracts. There is considerable attraction in doing so, not least because the law is in a state of uncertainty. The reasoning of the majority in the most recent decision of the House of Lords, Tinsley, is generally thought to be unsatisfactory: for a convincing analysis see the judgments in the decision of the High Court of Australia in Nelson v Nelson (1995) 184 CLR 538 (discussed more fully in paras 50-54 above). And the result arrived at by the minority in Tinsley is plainly unsatisfactory. I fear that the different approaches adopted by members of this court in the recent cases of Hounga v Allen [2014] 1 WLR 2889, Les Laboratoires Servier v Apotex Inc [2015] AC 430 and Bilta (UK) Ltd v Nazir (No 2) [2016] AC 1 have left the law on the topic in some disarray. As I said in Bilta, para 15, “the proper approach to the defence of illegality needs to be addressed by this court (certainly with a panel of seven and conceivably with a panel of nine Justices) as soon as appropriately possible”.

Nonetheless, there are arguments for not looking more widely at the issue of illegality in the contractual context. Thus, in all three recent Supreme Court cases (as in the present one), while there are some fairly sharp differences of opinion as to the proper approach, there is no real dispute as to the outcome. More broadly, the common law traditionally develops on a case by case basis, and there are self-evident dangers for a court to paint on an unnecessarily broad canvas, particularly bearing in mind that it is proceeding by reference to the facts of one particular case. And that can be said to be particularly true in the area of illegal contracts, where, as already mentioned, experience has shown that it is a topic fraught with difficulties, as is evidenced by the fact that the reasoning and outcomes in a number of cases concerning contracts affected by illegality over the past 300 years are hard to reconcile. Reading those cases also shows that it would be impossible to envisage, and therefore to cater for, every type of problem which might arise in this field.

Nonetheless, it seems to me right to venture further in this case, essentially for the reasons summarised in para 164 above.

The first general point I should make is that, in my view, even where the contemplated illegal activity has been performed in part or in whole, it would be right to apply the Rule in appropriate cases. Thus, in the case of an illegal contract where money is paid by the claimant to the defendant, and the contract is then partly or wholly performed by the defendant paying a lesser sum to the claimant, I do not see why, at least in the absence of good reasons to the contrary, the court should not order that the claimant should recover the money that he paid the defendant, albeit reduced by the lesser sum which the claimant subsequently received from the defendant. Similarly, where the contract is wholly performed.

In effect, the reasons supporting the application of the Rule in cases where the illegal activity has not occurred, apply for the same reasons to contracts where the contemplated illegal activity has been wholly or partly performed. And there is the added reason of consistency with a case where the contract has not been performed. Thus, in my view, if the defendant in this case had only been able to purchase just a few shares on inside information and had accounted to the claimant for the proceeds of sale of those shares in the sum of, say £10,000, the contract would have been partly performed, but I consider that the claimant could have successfully sued to recover the £620,000 he had paid, less the £10,000 which he had received.

There are, I think, three arguments against such a conclusion. The first is that there are a number of judgments, including those in and Taylor v Bowers 1 QBD 291, Kearley v Thomson 24 QBD 742 and Tribe v Tribe [1996] Ch 107, where it has been expressly stated that the Rule only applies where the “illegal purpose has [not] been wholly or partly performed” to quote from Millett LJ in Tribe at p 124E. However, perhaps particularly once one strips away the notion that repentance is irrelevant, I can see no good reason for not extending the rule to partly or even wholly performed contracts where restitutio in integrum can be achieved in practical terms and would be consistent with policy and proportionality. In the present case, for example, it would seem to be penal on the claimant that he could be deprived of £610,000 (and by the same token it would seem absurdly gratuitous that the defendant could benefit to the tune of £610,000) simply because the contract had been performed to a small extent.

Secondly, it may be argued that, once the contract has been partly performed, the basis for restitutio in integrum has gone. But that argument is only right if the basis of the Rule is total failure of consideration. In my view, that is not necessarily the correct analysis (unless the illegal consideration for which the money was paid is treated in law as no consideration, because it is illegal). Indeed, in the end, the correct analysis is not the centrally important issue, given that the question as to how the court deals with illegal contracts is ultimately based on policy. The ultimate function of the courts in common law and equity is to formulate and develop rules of a clear and practical nature. Now that the judiciary (rightly) pay more attention than we did to legal books and articles, we judges can look to legal academics not only to identify what they think are judicial inconsistencies and errors, but also to develop and modify their analyses of legal principles when we consider it necessary to change, develop or clarify the law.

Thirdly, it may be said that application of the Rule would result in the court sometimes getting precious close to enforcing an illegal contract - a course which the court most certainly cannot take, as already mentioned. I accept that application of the Rule would sometimes involve the court making an order whose effect in practice is similar to performance of the illegal contract. But there is nothing in that point. If a particular outcome is correct, then the mere fact that the same outcome could have been arrived at on a wrong basis does not make it the wrong outcome. Indeed, it is worth noting that the outcome in Tribe was precisely what it would have been if the contract in question had been enforced. The father had transferred the shares on the basis that it would help him avoid a threatened claim and that they would be transferred back when the claim was no longer threatened; he sought an order for the retransfer after the threat had gone away, and application of the rule resulted in that order.

That, of course, leaves open what would constitute “an appropriate case” for the application of the Rule and “good reasons to the contrary” for these purposes. The exceptions which I have referred to in para 162 above would be examples of where it might not be appropriate to invoke the Rule. However, it seems to me to be clear that there could be many other circumstances where application of the Rule would not be appropriate in circumstances where the illegal activity has been wholly or partly put into effect.

In that connection, some assistance can be obtained from the guidance given by McLachlin J. Beyond that, it may be that some or all of the factors identified by Professor Burrows in the passage quoted by Lord Toulson in para 93 above could be relevant depending on the facts and issues in any particular case. However, I am not convinced that it is helpful to list all the potentially relevant factors and say that it is a matter for the court in each case to decide which of those factors apply in that case and what weight to give them. Once a judge is required to take into account a significant number of relevant factors, and the question of how much weight to give each of them is a matter for the judge, the difference between judgment and discretion is, I think, in practice pretty slight.

I have come to the conclusion that the approach suggested by Lord Toulson in para 101 above provides as reliable and helpful guidance as it is possible to give in this difficult field. When faced with a claim based on a contract which involves illegal activity (whether or not the illegal activity has been wholly, partly or not at all undertaken), the court should, when deciding how to take into account the impact of the illegality on the claim, bear in mind the need for integrity and consistency in the justice system, and in particular (a) the policy behind the illegality, (b) any other public policy issues, and (c) the need for proportionality.

I must admit that I was initially not attracted by this approach because it seemed close to giving a discretion to judges when it comes to deciding how to deal with a claim based on a contract with an illegal element. However, on further reflection, it appears to me that, unlike the multi-factorial approach proposed by Professor Burrows, the structured approach proposed by Lord Toulson is not akin in practice to a discretion, and, in any event, it is the best guidance that can sensibly be offered at the moment. Experience shows that it is simply not possible to identify a more helpful or rigorous test. When considering whether it is possible to give more specific or firm guidance, I have considered some examples, which ultimately have helped to persuade me that greater clarity, strictness or specificity is simply not possible, at any rate at this stage, and they have served to conform the aptness of the approach set out in para 101 above.

A simple example is a case where the consideration for which the claimant paid or owed money was inherently illegal, rather than happening to involve an illegal act in order to be achieved. In such cases, it seems to me that considerations of certainty and policy indicate that the claimant should generally be able to refuse to pay any money which is due under the contract and, indeed, to recover the money he had paid. Thus, if the claimant paid a sum to the defendant to commit a crime, such as a murder or a robbery, it seems to me that the claimant should normally be able to recover the sum, irrespective of whether the defendant had committed, or even attempted to commit, the crime. If the defendant had not attempted the crime, the Rule would generally apply. If he had actually succeeded in carrying out the crime, he should not be better off than if he had not done so. I suppose one could justify that conclusion on the ground that the law should not regard an inherently criminal act as effective consideration.

That example might appear to suggest that more specific guidance could be given. However, even in relation to cases of the type described in para 176, there could be exceptions such as those mentioned in para 162 above. And, bearing in mind the enormous number of different crimes and different factual circumstances which could arise, it would be little short of foolhardy to imagine that there could not be other cases of this type where it would be inappropriate to apply the Rule.

Further, different considerations would often, I suspect very often, apply where the contract was not inherently illegal, but necessarily involved an illegal action. An extreme case might be where an employer employed a builder to carry out construction work which they both knew would inevitably require the builder to park illegally - say on a double red line. As already explained in para 160 above, if the defendant refused to carry out the work, the contract could not be enforced prospectively by the employer, but he would be entitled to recover any money he had paid. However, if the builder carried out the work, the employer would not be able to avoid liability to pay in full: the fact that the defendant could not perform his obligations under the contract without committing a relatively technical and incidental crime would not deprive him of the right to payment in full for such performance.

However, greater problems and uncertainties could arise in other cases - eg where the nature of the criminal activity was more serious and/or more central to the activity involved, where the illegal activity was expressly included in the contract, or where one of the parties did not know or intend that the activity in question to be carried out was illegal but the other did, or where the proceedings arose out of the fact that such a contract had only been partly performed.

Further, where a claimant has performed his part of a contract which was inherently lawful but was unlawful for some other reason, there is real room for debate in any particular case whether he should be entitled to claim payment on a quantum meruit basis, even though he cannot enforce his right to contractual payment - compare Mohamed v Alaga & Co [2000] 1 WLR 1815 and Taylor v Bhail [1996] CLC 377. While it would be possible to lay down a general rule as to whether or not a claimant could recover in such a case, it seems to me to be more satisfactory for the outcome to turn on the factors mentioned in para 174 above.

Similarly, it seems to me that the justification for the decision of the majority in Tinsley was, as Lord Toulson says, that it would have been disproportionate to have refused to enforce Miss Milligan’s equitable interest in the relevant property on the grounds of her illegal activity, and the policy behind the law making the activity in question illegal was not infringed by acceding to her claim.

It is also worth referring back to the two examples set out in para 162 above. If the purpose of rendering an activity illegal is to protect a class of persons which includes only one of the parties to the contract, then, absent any other argument based on policy or proportionality, it would seem appropriate that that party should not be disadvantaged by the illegality, and/or should be entitled to rely on the fact that the activity is illegal, as against the other party. And, if a claimant seeks recovery of money paid to a defendant under a contract which can only be performed illegally, and has not been performed, proportionality and policy may well justify the court refusing repayment if the defendant has spent the money and was unaware of the facts giving rise to the illegality at the time he spent it.

I would make three concluding points. First, quite apart from being persuaded by the reasons which justify the approach I have summarised in para 174 above, I consider that the fact that it is consistent with judgments of the courts in Australia and Canada, as explained by Lord Toulson in paras 50-61 above is a good reason for adopting the approach. When considering how to characterise, or whether to develop, any fundamental principle of the common law, it is normally sensible for a judge to consider how the principle has been approached in other common law jurisdictions, and it is desirable, if not always achievable, that all common law jurisdictions adopt the same approach.

Secondly, I should briefly address the fact that the criminal law and the Proceeds of Crime Act 2002 (“POCA”) may inevitably have some impact on the rights and duties of parties who have entered into contracts with an illegal connection. The involvement of the criminal law played a very important part in the judgment of McLachlin J in Hall v Hebert. It seems to me to have two main components. First, it is for the criminal law, not the civil law, to penalise a party or parties for entering into and/or performing a contract with an illegal component. Secondly, in so far as the civil law is fashioned by judges in a particular case, they must ensure that it is not inconsistent with the criminal law.

So far as POCA is concerned, it enables the courts, through statutory powers, to do that which a common law judge cannot do, and which many might think was the best outcome in many of the more serious cases involving illegality, namely to ensure that the proceeds of crime are retained by neither party, but are paid over to the Government. This is not the occasion to discuss the effect of POCA, save to say that I would take some persuading that the common law should be influenced by the fact that POCA is or is not being invoked in any particular case, although the civil courts should not make any order, or at least permit the enforcement of any order, if its effect would run counter to the provisions of POCA or to any step which was being contemplated under POCA by the relevant authorities.

Finally, I should say that, although my analysis may be slightly different from that of Lord Toulson, I do not think that there is any significant difference between us in practice. I agree with his framework for arriving at an outcome, but I also consider that there is a prima facie outcome, namely restitution in integrum.

That the law of illegality, particularly as it results from Tinsley v Milligan [1994] 1 AC 340, merits at the highest level the consideration now being given to it, I would be among the first to accept. I indicated as much as a party to the unsatisfying decision which the Court of Appeal had to reach in Collier v Collier [2002] BPIR 1057: see in particular para 106. Whether it is, however, appropriate to abandon basic principles going back nearly 250 years, resting on the sound appreciation of as a great a judge as Lord Mansfield CJ and approved and elucidated by the Supreme Court of Canada in an authoritatively reasoned judgment in 1993, is a different matter.

The basic problem, identified clearly and succinctly by Lord Mansfield in Holman v Johnson (1775) 1 Cowp 341, is that there are at least three potential interests when questions of illegality arise for consideration: those of two parties and the public interest. It is, as he said, for reasons of public interest that an otherwise good cause of action may sometimes fail, where there has been illegality. In the absence of any relevant statutory power, the court has no direct power to mediate between these three interests, by for example requiring the public interest to be satisfied by a payment to the public purse. It does not even have the power, conferred by statute in New Zealand, to vary or validate an illegal contract in part “or otherwise howsoever” (New Zealand Illegal Contracts Act 1970, section 7).

The application of the principle stated by Lord Mansfield was expanded in scope after his day (notably by Lord Eldon in Muckleston v Brown (1801) 6 Ves 52 as described by Lord Browne-Wilkinson in Tinsley v Milligan at p 372F. But, more recently it has diminished, Tinsley v Milligan being itself actually an example of this, in so far as it confirmed both that legal title to property could pass under an illegal contract and that equitable title was capable of recognition. The court’s recognition of the equitable title was, however, made subject to the (problematic) pre-condition that the claimant could avoid reliance on illegality by relying on a procedural presumption. The court was able, in Tinsley v Milligan, to derive this presumption from the objectively demonstrable contribution made by Miss Milligan to the cost of acquiring the property. At the same time the court was prepared to ignore the fact, perfectly well-known to it, of the parties’ illegal intentions.

In common with Lord Toulson (paras 100-101), I consider that valuable insight into the appropriate approach to the significance of illegality under today’s conditions is found in the judgment of McLachlin J (as she was) writing for the majority the Supreme Court of Canada in Hall v Hebert [1993] 2 SCR 159. The case concerned a claim in tort by a passenger against the owner of a car, who lost the keys when they fell out of the ignition when the car stalled and who decided in these circumstances that his passenger (who he knew to have drunk 11 or 12 bottles of beer) should drive while he tried to push-start the car. Unsurprisingly, the manoeuvre led to the passenger losing control, the car turning over and the passenger being injured. The Canadian Supreme Court upheld the passenger’s claim, subject to contributory negligence.

The majority in the Canadian Supreme Court rightly regarded the case as one of “great importance”. A number of points emerge with great clarity from McLachlin J’s judgment:

First, rejecting Cory J’s suggestion that “a power to reject claims on considerations of public policy” should replace the maxim ex turpi causa non oritur action, McLachlin J expressed her concern that public policy would provide no “clear guidance as to when judges could exercise this draconian power and upon what grounds”. She went on:

Second, she saw tort, not contract, as the real problem area in relation to illegality, expressing the view that:

Third, after examining authorities where the maxim applied to prevent claimants from profiting or obtaining exemplary damages in circumstances of illegality, she identified its rationale in today’s world, in terms which have equal relevance to contract and tort:

Fourth, McLachlin J said that such compensatory damages as were claimed in Hall v Hebert are

Finally, she concluded that:

In my opinion, what is called for is a limited approach to the effect of illegality, focused on the need to avoid inconsistency in the law, without depriving claimants of the opportunity to obtain damages for wrongs or to put themselves in the position in which they should have been. This will offer the opportunity of resolving such problems as have, rightly, been identified in the present law, without replacing it wholesale with an open and unsettled range of factors. The latter might, in McLachlin J’s words, “prove more problematic than has the troubled doctrine of ex turpi causa” itself.

McLachlin J’s emphasis on the admissibility of compensatory claims leads me to the principle traditionally addressed under the head of locus poenitentiae. This principle in fact had a relevant role in the Tinsley v Milligan in so far as it was recognised as demonstrating that “the effect of illegality is not to prevent a proprietary interest in equity from arising or to produce a forfeiture of such right: the effect is to render the equitable interest unenforceable in certain circumstances”: per Lord Browne-Wilkinson, p 374D. But its true significance is considerably greater. Where it applies, it fulfils a not dissimilar function to a claim for damages in tort. It puts the parties back in the position that they should have been in, in this case but for the entry into of the contract which was or became affected and unenforceable by reason of the illegality.

In early authorities the principle was put in wide terms. Smith v Bromley (1760) 2 Doug KB 696n was a case where the plaintiff was able to recover money she had paid to procure her brother’s discharge from bankruptcy, which was an illegal payment. The primary reason was that the law making it illegal was for the protection of bankrupts and their families (so that the plaintiff and the defendant were non in pari delictu). An editor’s footnote (F7) on p 697 gives this as one of two exceptions to the principle that, in a case of illegality, matters are left to lie where they fall (potior est conditio defendentis). But Lord Mansfield CJ reinforced this reason by the more general consideration at p 698, that, although the payment had been made for an illegal purpose:

In Neville v Wilkinson (1782) I Bro Ch 543, 547, Lord Thurlow LC approved this approach, and “declared his opinion” that:

In Taylor v Bowers (1876) 1 QBD 291 possession of goods had been passed by the plaintiff, their owner, to A, in exchange for fictitious bills of exchange, in order to deceive creditors. But no compromise was achieved with creditors, “the illegal transaction was not carried out”, and “it wholly came to an end” (p 300). In these circumstances, the plaintiff successfully sought recovery of the goods:

Like Lord Sumption (paras 245-252), I see this principle of rescission as having become unduly limited with time, particularly in 20th century authority, and I consider that it should be restored to its former significance and generalised. Further, I consider that there is no reason why rescission should necessarily be restricted, as it was even in these earlier authorities, by reference to a test of execution or carrying out of the illegal purpose. The logic of the principle is that the illegal transaction should be disregarded, and the parties restored to the position in which they would have been, had they never entered into it. If and to the extent that the rescission on that basis remains possible, then prima facie it should be available.

In addition, as at present advised, I would not see any necessary objection to permitting rescission after part performance, by making, where possible, appropriate adjustments for benefits received. Equally, picking up points in Lord Neuberger’s judgment (para 162) which I have read since writing the bulk of this judgment, I would not as at present advised see an imbalance or lack of parity of delict between the parties as a necessary or even probable bar to rescission, though I would agree that, in accordance with general principle, factors such as change of position could well preclude rescission. Complications may also arise in a context where a benefit received under an illegal transactions is capable of forfeiture under the Proceeds of Crime Act 2002. We did not hear submissions on the position in such circumstances, and I express no opinion on it.

On the above basis, reliance on illegality remains significant as a bar to relief, but only in so far as it is reliance in order to profit from or otherwise enforce an illegal contract. Reliance in order to restore the status quo is unobjectionable. The result is, as I see it, not dissimilar to that which (leaving aside the potential effects of section 7) results under section 6(1) of the New Zealand Illegal Contracts Act 1970, providing that:

The approach I adopt avoids unsatisfactory results such as that reached in Collier v Collier, where it would have been entirely possible to achieve rescission even though the illegal scheme had been in some measure “executed” or “carried out”. The father there could require the restoration of the property of which he had for an illegal purpose allowed his daughter to have the legal title. Similarly, in a situation like that in Tinsley v Milligan, it should be possible to avoid reliance on the artificial procedural concept of a presumption of a resulting trust. Such a presumption was available in that case to give effect to (though without necessarily referring to) the parties’ actual intentions regarding equitable ownership or the reason (although the court was well aware of it) for structuring the transactions as they were. But, as Collier v Collier demonstrates, an artificial procedural presumption of this nature cannot be relied upon to be available in every case.

In future, Miss Milligan should simply be able to reverse the effect, as between herself and Miss Tinsley, of the property transactions which they arranged for the illegal purpose, which they carried out, of deceiving public authorities. Because the court would be reversing, rather than enforcing the illegal transactions, the court could take into account both the objective fact of joint contributions and the parties’ actual and, by itself, legal purpose of joint ownership. Setting on one side the transactions by which they sought to achieve their illegal purpose, the underlying equitable interests, which they shared based on their contributions and intentions, would be enforceable as such. The court could on that basis order the property to be registered in the joint names of Miss Tinsley and Miss Milligan.

It follows from what I have so far said that I cannot accept Lord Toulson’s view (para 116) that it is unnecessary to consider the scope of locus poenitentiae. The underlying concept behind locus poenitentiae is restitutionary. It recognises that neither an admission of nor reliance on illegality is a bar to relief involving the reversal of an illegal transaction. In the full restitutionary sense I have discussed, the concept must be seen as an integral part of the overall principle governing illegality, and as the corollary of McLachlin J’s limited rationalisation of that principle. Understood in that sense, free of early 20th century moralising, it restores the position to what it would and should have been, without any illegality. It avoids windfall benefits and disproportionate losses, without involving the positive enforcement of or the recovery of profits based on illegal bargains. No doubt, however, it would be desirable to avoid the moral undertones of the Latin brocard, and to encapsulate the full width of the modern principle, by referring in future simply to parties’ normal entitlement to reverse the effects of an illegal transaction, where possible, even though the transaction may have been wholly or in part executed or carried into effect.

It also follows that in the present case I consider that no problem exists about recognising that Mr Patel is entitled to require Mr Mirza to return the stake which Mr Patel put up for the illegal purpose of use by Mr Mirza to make profits for their joint benefit by misuse of inside information. The claim does not seek to enforce or profit by the illegality. It seeks merely to put the position back to where it should have been, and would have been had no such illegal transaction ever been undertaken. I add that, having written the above and read Lord Neuberger’s judgment in draft, it seems to me that, thus far, my analysis is essentially the same as that which Lord Neuberger describes in his judgment as “the Rule”.

Before leaving the case, I must however return to the suggestion, unnecessary in my view for the resolution of this appeal, that the law of illegality should be generally rewritten. The new approach is advocated primarily by Lord Toulson, but Lord Neuberger appears, unless I have misunderstood him, to suggest that it could serve both as a potential modification or qualification of the Rule and as an approach to be adopted to claims positively to enforce a contract, and to claims for damages for breach of contract or a quantum meruit for services rendered under an illegal contract (see his paras 174-175 and 178-180). The new approach is ostensibly based by Lord Toulson on Hall v Hebert, but it is transmuted by the statement (by Lord Toulson in para 101) that:

Under consideration c), it is then indicated (paras 107 and 108) that:

What is apparent is that this approach, would introduce not only a new era but entirely novel dimensions into any issue of illegality. Courts would be required to make a value judgment, by reference to a widely spread mélange of ingredients, about the overall “merits” or strengths, in a highly unspecific non-legal sense, of the respective claims of the public interest and of each of the parties. But courts could only do so, by either allowing or disallowing enforcement of the contract as between the two parties to it, unless they were able (if and when this was possible) to adopt the yet further novelty, pioneered by the majority of the Australian court in Nelson v Nelson [1995] HCA 25, (1995) 184 CLR 538, of requiring the account to the public for any profit unjustifiably made at the public expense, as a condition of obtaining relief.

Although other jurisdictions are invoked, it is notable how slender the basis for doing so is. It comes down to the New Zealand statute and the Australian authorities of Nelson v Nelson and Fitzgerald v F J Leonhardt Pty Ltd [1997] HCA 17, (1997) 189 CLR 215. We have no idea or information as to whether or not the approach there has proved unproblematic for the profession or the courts. What we do however have is an authoritative decision of the Canadian Supreme Court in Hall v Hebert, which does not in any way support the wholesale abandonment of a clear cut test, but rather explains and redefines the principle ex turpi causa in a manner which (consistently with the way in which the common law usually develops) offers every prospect of avoiding the evident anomalies which an over formalistic approach has in the past evidenced.

Lord Toulson also starts his judgment with a series of paragraphs (1 to 9) instancing what are supposed to be problems existing under the present law. I would only say as to Holman v Johnson (1775) 1 Cowp 341 and Pearce v Brooks (1866) LR 1 Ex 213 that the question what constitutes knowing participation sufficient to render a contract unenforceable is a discrete problem, which is unlikely to be resolved any more simply under the “range of factors” approach now advocated. Likewise, the St John Shipping case [1957] 1 QB 267 and Ashmore Benson Pease & Co Ltd v A V Dawson Ltd [1973] 1 WLR 828 arose in areas where the purpose and effect of statutory provisions were central to the decision (as it was in cases such as Hall v Woolston Hall Leisure Ltd [2001] 1 WLR 225, Still v Minister of National Revenue (1997) 154 DLR (4th) 229 and Nizamuddowlah v Bengal Cabaret Inc (1977) 399 NYS 2d 854, mentioned by Lord Toulson in paras 6, 58-61 and 63-66). Questions as to the effect of collateral or minor illegality (such as parking on a double red line, instanced by Lord Neuberger in para 178) on the enforceability of contractual rights have not, I believe, led to real difficulty in achieving just solutions under these and other authorities (compare also McLachlin J’s view cited in para 191(ii) above) - and certainly not to such difficulty as to justify tearing up the existing law and starting again. Again, the new approach now advocated, with its wide range of additional factors, over and above statutory purpose and effect, would be unlikely to avoid similar analysis of statutory policy and similarly nice issues. More importantly, these are problems in areas far removed from the present, and do not to my mind throw any light on the issues we have to decide on this appeal.

For the reasons I have given, which correspond with those given by Lord Clarke and Lord Sumption, I would dismiss this appeal.

As I see it, there is no disagreement between members of the court as to the correct disposal of this appeal. It is that the appeal must be dismissed because Mr Patel is entitled to restitution of the £620,000 that he paid to Mr Mirza on the basis that otherwise Mr Mirza would be unjustly enriched. As it seems to me, the application of orthodox principles of unjust enrichment, rescission and restitutio in integrum leads to this conclusion. Those principles are consistently set out by Lord Mance and Lord Sumption. Although Lord Sumption sets out a broader statement of principle, he agrees with Lord Mance and vice versa. As it seems to me, there is no difference between their approach and the application by Lord Neuberger of what he calls “the Rule”, which he defines in paras 145 and 146, as the right to return of money paid by the claimant to the defendant pursuant to a contract to carry out an illegal activity and the illegal activity is not in the event proceeded with owing to matters beyond the control of either party. Lord Sumption, at para 252, emphasises that the Rule arises automatically and by operation of law; a “right to restitution that in principle follows from the legal ineffectiveness of the contract …”. I do not understand Lord Neuberger or Lord Mance to disagree with that. As Lord Neuberger says in para 146, the Rule is consistent with authority and with policy and renders the outcome in cases of contracts involving illegality and the maxim ex turpi causa non oritur action relatively clear and certain.

As Lord Neuberger says at para 154, in agreement with Lord Mance, there is obvious attraction in the notion that, if all transfers made pursuant to an unexecuted illegal contract are re-transferred, then the parties are back in the position they were, ie as if there had been no illegal contract, which would seem to comply with public policy.

This approach does not require any balancing of a series of different factors. It simply applies the principles derived from the authorities to the facts of the case. Lord Neuberger, Lord Mance and Lord Sumption have referred in detail, and (so far as I can see) consistently, to the authorities over very many years. None of them supports a balancing of the kind suggested by Lord Toulson. To my mind the most important sources are the judgments of Lord Mansfield CJ in Holman v Johnson (1775) 1 Cowp 341 and McLachlin J (now CJ) in Hall v Hebert [1993] 2 SCR 159.

Lord Mance sets out in para 191 what he calls a number of points which emerge with great clarity from McLachlin J’s judgment. I will not repeat those passages here. The critical point for present purposes is that she stressed the importance of having a firm doctrinal foundation for what she described as a narrow principle. She was concerned, at p 169, that public policy would provide “no clear guidance as to when judges could exercise this draconian power and upon what grounds”. The draconian power was “a power to reject claims on considerations of public policy”. On the facts of Hall v Hebert she concluded that such compensatory damages as were claimed in that case were not properly to be regarded as awarded as compensation for an illegal act but only as compensation for personal injury. Then, as Lord Mance says, finally she concluded that:

I entirely agree with that approach. I have always thought that the power of the court to deny recovery on the ground of illegality should be limited to well defined circumstances. I agree with Lord Mance in para 192 that, in the absence of such circumstances, claimants should not be deprived of the opportunity to obtain damages for wrongs or to put themselves in the position in which they should have been. As I see it, there is no need to replace that approach with what he calls an open and unsettled range of factors.

I agree with Lord Sumption’s opinion in this regard. As he puts it at para 257, the search for principle which led McLachlin J to identify consistency as the foundation of this area of the law was a response to Cory J, who had favoured a more flexible approach which would have depended upon whether the relevant public policy required that result on the facts of each case. The majority, including McLachlin J, did not agree. In para 258 Lord Sumption draws attention to the similar opinion of Lord Goff in Tinsley v Milligan [1994] 1 AC 340 at 358E-F, where he objected to the “public conscience test” adopted in the Court of Appeal, under which the court must “weigh, or balance, the adverse consequences of respectively granting or refusing relief”. Lord Goff added that that was “little different, if at all, from stating that the court has a discretion whether to grant or refuse relief” and that it was very difficult to reconcile with the principle of policy stated by Lord Mansfield in Holman v Johnson. As Lord Sumption observes, on this point Lord Goff was supported by the whole of the Appellate Committee.

Between paras 259 and 265 Lord Sumption considers what he calls the “range of factors” approach and gives his reasons for rejecting it. I agree with him, and will not repeat his reasoning here, save for the following passage at para 262(iv):

In para 265 Lord Sumption says that he cannot agree with the conclusion of Lord Toulson (at para 109) that the application of the illegality principle should depend on

Lord Mance expresses much the same conclusion in paras 204 to 207, with which I also agree. It is to my mind noteworthy that Lord Toulson puts his conclusion thus in para 109:

The striking feature of that approach is as I see it that it puts the question, not whether the contract should be regarded as tainted by illegality but whether the relief claimed should be granted. That seems to me to be essentially a question of discretion, or at least a consideration of all the relevant factors in order to decide where the balance should be struck. As I see it, there is no support in any of the authorities for that approach and it is directly contrary to many of the cases referred to by Lord Sumption and Lord Mance, in particular the reasoning of the majority in Hall v Hebert and of the House of Lords in Tinsley v Milligan, where it was expressly rejected by Lord Goff. It would be close to reviving the public conscience test. In my opinion the question posed in para 109 is the wrong question.

I recognise that common law principles develop from time to time. Two such developments are relevant here. The first is this. Lord Sumption and Lord Mance both focus on the scope of the principle of restitutio in integrum, as does Lord Neuberger. For example, Lord Neuberger first sets out the basis of the Rule, which seems to me to be consistent with the principles identified by Lord Sumption and Lord Mance. Thus in much of his judgment, notably in paras 145 to 160, Lord Neuberger stresses that the Rule supports the importance of certainty in the law. He then gives some examples of possible extensions of the Rule. So, for example, he says in paras 167 to 169 that the Rule may apply where the illegal contract is wholly or partly performed by the plaintiff paying a lesser sum to the defendant. I agree, but that is on the basis that it is essentially ordering restitution so far as appropriate in accordance with the underlying principle embodied in the Rule. As Lord Neuberger puts it in para 169, there is no good reason for not extending the Rule to partly or even wholly performed contracts where restitutio in integrum can be achieved in practical terms and would be consistent with policy and proportionality. As I read his judgment, save at the very end his approach is orthodox and contemplates a development of the legal principles identified by Lord Sumption and Lord Mance.

The second relevant development is this. It is now recognised that some of the reasoning in Tinsley v Milligan can no longer stand: see in particular Lord Sumption at paras 236 to 239 and Lord Mance at paras 199 to 201. It is I think now accepted on all sides that, if Collier v Collier [2002] BPIR 1057 came before the courts today it would be decided differently. That is not however because the court will adopt the proposals of Lord Toulson but because the relevant legal principles have developed in a normal way.

Finally, I should note that it is not in dispute that the appeal should be dismissed on conventional principles.

I recognise that Lord Neuberger has expressed some support for the approach of Lord Toulson but I am not persuaded by his reasoning that it is appropriate.

Two questions arise on this appeal. The first is whether the contract between these parties is affected by the principle of public policy ex turpi causa non oritur actio (the illegality principle, as I shall call it). The second is whether, if so, Mr Patel is entitled to restitution of the £620,000 that he paid to Mr Mirza.

The first question has divided the courts below. The Deputy Judge (David Donaldson QC) and the majority of the Court of Appeal (Rimer LJ and Vos LJ) thought it plain that Mr Patel’s claim was founded on an illegal agreement and could not be sustained unless he could invoke a special exception for executory agreements. They considered that there was such an exception. Gloster LJ on the other hand declined to see the problem in terms of rule and exception. At the risk of a rather crude summary of her thoughtful analysis, I would summarise her reasons as follows. Her first and main point (paras 67, 69-70, 72, 79-80) was that the rationale of the illegality rule did not require Mr Patel to be denied restitutionary relief, because it did not involve enforcing his contract with Mr Mirza or enabling him to derive any benefit from it. Mr Patel’s right to restitution was, she considered, “collateral”. Second, that Mr Mirza and Mr Patel were not equally blameworthy because Mr Mirza was a finance professional while Mr Patel was not, and would not necessarily have known that insider dealing was illegal. Third, section 63(2) of the Criminal Justice Act 1993 provided that no contract should be void or unenforceable by reason of the prohibition of insider dealing in section 52. The fourth was that Mr Patel did not need to rely on the illegal character of his agreement with Mr Mirza in order to recover the money. It was enough that he had paid it for a speculation that never occurred.

The present appeal exposes, not for the first time, a long-standing schism between those judges and writers who regard the law of illegality as calling for the application of clear rules, and those who would wish address the equities of each case as it arises. There are recent statements of this court in support of both points of view: see Les Laboratoires Servier v Apotex Inc [2015] AC 340 and Hounga v Allen [2014] 1 WLR 2889, paras 44-45. It also raises one of the most basic problems of a system of judge-made customary law such as the common law. The common law is not an uninhabited island on which judges are at liberty to plant whatever suits their personal tastes. It is a body of instincts and principles which, barring some radical change in the values of our society, is developed organically, building on what was there before. It has a greater inherent flexibility and capacity to develop independently of legislation than codified systems do. But there is a price to be paid for this advantage in terms of certainty and accessibility to those who are not professional lawyers. The equities of a particular case are important. But there are pragmatic limits to what law can achieve without becoming arbitrary, incoherent and unpredictable even to the best advised citizen, and without inviting unforeseen and undesirable collateral consequences.

Ancient as it is, the classic statement of the principle as it has traditionally been understood remains that of Lord Mansfield CJ in Holman v Johnson (1775) 1 Cowp 341:

There was a time when the courts approached the application of the illegality principle on the footing that a court should not be required to sully its hands by dealing with criminal ventures. In Everet v Williams (1725), noted at (1893) LQR 197, the notorious case in which two highwaymen sought an account of the division of their profits, the court not only dismissed the action but fined the plaintiff’s solicitors for the indignity visited upon it. There are periodic echoes of this attitude in later cases, notably Parkinson v College of Ambulance Ltd [1925] 2 KB 1, 13, in which Lush J thought that no adjudication on a contract to procure an honour could be undertaken with “propriety or decency”. This notion has sometimes been thought to derive support from Lord Mansfield’s reference to the court withholding its aid. But the truth is that it has rarely risen above the level of indignant judicial asides. There are many purposes for which courts must necessarily inquire into the illegal acts of litigants. There are principled exceptions to the illegality principle, which may entitle a party to base a claim on an illegal act. There are statutory schemes of apportionment which may require liability for dishonest acts to be distributed among the wrongdoers. The notion of judicial abstention could never be unqualified, nor has it been historically. The law, as Bingham LJ observed in Saunders v Edwards [1987] 1 WLR 1116, 1134, must

In its consultative report of 2009, The Illegality Defence (LCCP 189), at para 2.5, the Law Commission identified six policy rationales for the rule, which could be found in the case-law and the academic literature. They were: (1) furthering the purpose of the rule which the claimant's illegal behaviour has infringed; (2) consistency; (3) the need to prevent the claimant profiting from his or her own wrong; (4) deterrence; (5) maintaining the integrity of the legal system; and (6) punishment.

By “maintaining the integrity of the legal system” (rationale (5)), the Law Commission meant sparing the judiciary from involvement in serious wrongdoing: see para 2.24. I have given my reasons for rejecting this rationale. The Law Commission itself (paras 2.28-2.29) rejected rationale (6), punishment, on the ground that although rules of civil law might have a punitive effect, this was no part of their purpose. With very limited exceptions, such as certain rules of causation in fraud cases or the rare occasions for awarding punitive damages, I think that this is correct. The other four rationales overlap. All of them to my mind are subsumed in no (2), the principle of consistency. The most influential statement of that principle is to be found in the much admired judgment of McLachlin J delivering the judgment of the majority of the Supreme Court of Canada in Hall v Hebert [1993] 2 SCR 159, 169:

In practice the illegality principle has almost invariably been raised as a defence to a civil claim based on a breach of the criminal law. In Les Laboratoires Servier v Apotex Inc [2015] AC 430, this court held that with immaterial exceptions the defence is available only in such cases. This conclusion tends to reinforce the significance of the principle of consistency as a rationale. The civil courts of the state cannot coherently give effect to legal rights founded on criminal acts which are contrary to the state’s public law. There is no reason to regard this as any less important according to whether the civil claim lies in contract or tort.

The English courts have taken a broader view than McLachlan J did of what constitutes “profiting” from an illegal act, but that is by the way. Her rationalisation of the illegality principle as being based on the consistency and internal coherence of the law has been consistently adopted in England in tort and contract cases alike by this court and by the Appellate Committee of the House of Lords before it: see R v Islam [2009] AC 1076, para 38 (Lord Mance); Stone and Rolls Ltd v Moore Stephens [2009] 1 AC 1391, paras 128 (Lord Walker), 226 (Lord Mance); Hounga v Allen [2014] 1 WLR 2889, para 43 (Lord Wilson); Les Laboratoires Servier v Apotex Inc [2015] AC 430, para 24 (Lord Sumption); Bilta (UK) Ltd v Nazir (No 2) [2016] AC 1, para 172 (Lord Toulson and Lord Hodge). In Gray v Thames Trains [2009] 1 AC 1339, Lord Hoffmann (with whom the rest of the Appellate Committee agreed) put forward the principle of consistency as the rationale of what he called the “narrower rule” precluding the recovery of damages representing loss directly arising from the sentence of a criminal court. He was inclined to think that the “wider rule” that a person cannot recover for damage which is the consequence of his own criminal act was based on a different principle concerned with “public notions of the fair distribution of resources”: para 51, and cf Lord Rodger at para 84. Certainly, the inconsistency of awarding damages representing loss arising from a criminal sentence is more obvious and direct than it is when the claimant is claiming other damages causally flowing from his commission of a crime. But it seems to me, as it did to McLachlan J and those who have adopted her approach more generally, that the internal coherence of the law is also the reason why it will not give effect in a civil court to a cause of action based on acts which it would punish in a criminal court. As Lord Hughes put it in Hounga v Allen (para 55), a dissenting judgment but not on this point, “the law must act consistently; it cannot give with one hand what it takes away with another, nor condone when facing right what it condemns when facing left.”

The starting point is that the courts exist to provide remedies in support of legal rights. It is fundamental that any departure from that concept should have a clear justification grounded in principle, and that it should be no more extensive than is required by that principle. The underlying principle is that for reasons of consistency the court will not give effect, at the suit of a person who committed an illegal act (or someone claiming through him), to a right derived from that act.

The test which has usually been adopted for determining whether this principle applies is the reliance test. The question is whether the person making the claim is obliged to rely in support of it on an illegal act on his part. The reliance test is implicit in Lord Mansfield’s statement of principle, which assumes that the plaintiff’s action is “founded on” his illegal act. But the modern origin of the test is the decision of the Court of Appeal in Bowmakers Ltd v Barnet Instruments Ltd [1945] 1 KB 65, which concerned a hire purchase agreement illegal under wartime regulations. When the hirer disposed of the goods, the owner was held entitled to damages for conversion notwithstanding the illegality, because his right of action was based on his ownership. He could establish that without relying on the illegal hire purchase agreement. The reliance test was subsequently approved by the Privy Council in Singh v Ali [1960] AC 167 and Chettiar v Chettiar [1962] AC 294 and by the House of Lords in Tinsley v Milligan [1994] 1 AC 340. All of these decisions, were about title to property, real or personal. But in Clunis v Camden and Islington Health Authority [1998] QB 978 the Court of Appeal applied it to a claim in tort. In St John Shipping Co Ltd v Joseph Rank Ltd [1957] 1 QB 267, 291-292, Devlin J had applied it to a claim for freight under a contract of carriage. In Hewison v Meridian Shipping Services PTE Ltd [2003] ICR 766, the Court of Appeal applied it to a concurrent claim for damages in contract and tort in which the measure of damages depended on the terms of a contract. The claimant’s action for damages against his employer for an injury at work failed because in order to prove his loss of earnings he had to show that he would have continued to deceive his employer about his fitness to operate machinery, as he had in the past.

There is, as these decisions suggest, nothing about the reliance test that limits its relevance to certain causes of action. But the test may apply in different ways, depending on what it is that the law regards as illegal. In a tort case or a property case it is generally enough to identify the illegal act and demonstrate the dependence of the cause of action upon the facts making it illegal. In a contract case, the position is less straightforward. A contract may be affected by illegality because terms lawful in themselves are intended to be performed in an illegal way or for an illegal purpose not apparent from the contract itself. This does not mean that contracts vitiated by this circumstance can be enforced simply by putting the case without reference to the illegal purpose or proposed mode of performance. It is enough to give rise to the defence that the claimant must rely on a contract which is in fact illegal, whether that is apparent from the terms or not.

The problem about the reliance test is not so much the test itself as the way in which it was applied in Tinsley v Milligan. The facts of that case are well known. Ms Tinsley and Ms Milligan contributed in approximately equal shares to the cost of buying a house in which both of them intended to live and run their lodging rooms business. They decided that it would be conveyed into the sole name of Ms Tinsley so as to enable Ms Milligan to defraud the Department of Social Security by pretending that she did not own her home and paid rent. Ms Tinsley claimed an order for possession on the footing that she was the sole owner. The Appellate Committee held by a majority that Ms Milligan was entitled to assert a 50% interest in the house notwithstanding the illegal purpose for which it had been conveyed into Ms Tinsley’s sole name. There were two stages in the reasoning of Lord Browne-Wilkinson, who delivered the leading speech for the majority. The first was that where property is transferred for an illegal purpose, the transferee nevertheless obtains a good title, notwithstanding that the transaction being illegal it would not have been specifically enforced. This is so whether the title in question is legal or equitable. The decision of the majority on this point settled a question on which there had been inconsistent authorities dating back to the beginning of the 19th century. It did so in a way which reflected the law’s traditional reluctance to disturb settled titles. The result represents a notable difference between the law relating to the creation of legal or equitable titles and the law relating to contractual obligations generally. It means that although a contract may be vitiated by its illegal purpose or the illegal way in which it was intended to be performed, this is not true of title to property. It followed in that case that Ms Tinsley had a good title to the disputed property. The second stage of the reasoning was that an equitable interest in the property would also be recognised, provided that the person claiming it was not “forced to plead or rely on the illegality” (p 376E). In Ms Milligan’s case, equity presumed a resulting trust in her favour by virtue only of her contribution to the purchase price. She did not therefore have to plead or prove the reasons why the property had been conveyed into Ms Tinsley’s sole name. It followed that she could make good her claim to an interest.

The problem about this is that it makes the illegality principle depend on adventitious procedural matters, such as the rules of pleading, the incidence of the burden of proof and the various equitable presumptions. If Ms Tinsley had been a man and Ms Milligan had been his daughter, the decision would have gone the other way because the presumption of resulting trust would have been replaced by a presumption of advancement. She would have had to rebut it by reference to the actual facts. This is what the Privy Council decided in Chettiar v Chettiar [1962] AC 294 and the Court of Appeal in Collier v Collier [2002] BPIR 1057, in both of which property was gratuitously transferred for an illegal purpose by a father to his son or daughter. The father was accordingly unable to establish his interest. Yet the distinction between these cases and Tinsley v Milligan is completely arbitrary. This is because the equitable presumptions operate wholly procedurally, and have nothing to do with the principle which the court is applying in illegality cases.

In Nelson v Nelson (1995) 184 CLR 538, the majority’s analysis in Tinsley v Milligan was criticised on this ground in the High Court of Australia: see pp 579-580 (Dawson J), 592-593 (Toohey J), 609-610 (McHugh J). In my opinion, these criticisms are justified, although I would not go as far as McHugh did in Nelson v Nelson. He, alone among the judges of the High Court of Australia, would have jettisoned the reliance test altogether. What then is the true principle? In property cases, as the House held in Tinsley v Milligan, title is not vitiated by an antecedent illegal arrangement. An equitable interest in property may accordingly arise from a tainted scheme. Whether an equitable interest exists depends on the intentions of the parties. The true principle is that the application of the illegality principle depends on what facts the court must be satisfied about in order to find an intention giving rise to an equitable interest. It does not depend on how those facts are established. Ms Milligan was entitled to the interest which she claimed in the property because she paid half of the price and there was no intention to make a gift. That was all that the court needed to be satisfied about. Likewise, if Collier v Collier were to come before the courts today, the result should be the same notwithstanding that the equitable presumption went the other way. Mr Collier leased his property to his daughter for an illegal purpose, namely to deceive his creditors in the event that he became insolvent. He had an equitable interest in the property because the lease was gratuitous and there was no intention to make a gift. It would make no difference to the recognition of that interest that the purpose of the transaction was illegal. Why he chose to organise his affairs in that way would no doubt emerge in the course of the evidence, but would be irrelevant to the facts which founded his claim. The point was well made by Dawson J in Nelson v Nelson, at p 580:

Shorn of the arbitrary refinements introduced by the equitable presumptions, which in any event apply only in property cases, the reliance test accords with principle. First, it gives effect to the basic principle that a person may not derive a legal right from his own illegal act. Second, it establishes a direct causal link between the illegality and the claim, distinguishing between those illegal acts which are collateral or matters of background only, and those from which the legal right asserted can be said to result. Third, it ensures that the illegality principle applies no more widely than is necessary to give effect to its purpose of preventing legal rights from being derived from illegal acts. The reliance test is the narrowest test of connection available. Every alternative test which has been proposed would widen the application of the defence as well as render its application more uncertain.

This last objection applies in particular to the main alternative test which has been proposed in the case law, namely that the facts relied upon should be “inextricably linked” with the illegal act. The difficulty about inextricable linkage as a test of connection is that it is far from clear what it means. On the face of it, the only link between the illegal act and the claim which is truly “inextricable”, is a link based on causation and necessary reliance. So far as the test of inextricable linkage broadens the required connection more widely, it seems to me to be contrary to principle. Its vices may be illustrated by reference to the decision in Cross v Kirkby [2000] EWCA Civ 426, The Times 5 April 2000, where it was first proposed by Beldam LJ. The facts were that a hunt saboteur started a fight with a hunt follower at a meet and came out of it worst. He ended up with a fractured skull, and sued the hunt follower for damages occasioned by his injuries. The main issue was whether the hunt follower had defended himself with excessive force. Beldam LJ held that he had not. But in case he was wrong about that, he held that the saboteur’s injuries were inextricably linked with the fact that he had started the fight, so that his claim was barred by the illegality principle. Otton LJ agreed generally with Beldam LJ, but Judge LJ agreed only on the primary ground. To my mind, Beldam LJ’s alternative ground was unprincipled. It only arose if the hunt follower responded to the attack with excessive force, and on that footing it was irrelevant who started the fight. The illegality principle served simply to deprive the plaintiff of a proper claim arising from the unlawful use of excessive force against him. The case illustrates the tendency of any test broader than the reliance test to degenerate into a question of instinctive judicial preference for one party over another.

To the principle that a person may not rely on his own illegal act in support of his claim, there are significant exceptions, which are as old as the principle itself and generally inherent in it. These are broadly summed up in the proposition that the illegality principle is available only where the parties were in pari delicto in relation to the illegal act. This principle must not be misunderstood. It does not authorise a general enquiry into their relative blameworthiness. The question is whether they were legally on the same footing. The case law discloses two main categories of case where the law regards the parties as not being in pari delicto, but both are based on the same principle.

One comprises cases in which the claimant’s participation in the illegal act is treated as involuntary: for example, it may have been brought about by fraud, undue influence or duress on the part of the defendant who seeks to invoke the defence. The best-known example is Burrows v Rhodes [1899] 1 QB 816, where the illegality consisted in the plaintiff having enlisted in the defendant’s private army for the Jameson raid, contrary to the Foreign Enlistment Act 1870. The illegality principle was held not to arise because he had been induced to do so by the defendant’s fraudulent misrepresentation that the raid had the sanction of the Crown, which if true would have made it legal. Cases in which the illegality consisted in the act of another for which the claimant is responsible only by virtue of a statute imposing strict liability, fall into the same category: see Osman v J Ralph Moss Ltd [1970] 1 Lloyd’s Rep 313; Les Laboratoires Servier v Apotex [2015] AC 430, para 29. In such cases, however, the construction and purpose of the statute in question will call for careful attention.

The other category comprises cases in which the application of the illegality principle would be inconsistent with the rule of law which makes the act illegal. The paradigm case is a rule of law intended to protect persons such as the plaintiff against exploitation by the likes of the defendant. Such a rule will commonly require the plaintiff to have a remedy notwithstanding that he participated in its breach. The exception generally arises in the context of acts made illegal by statute. In Browning v Morris (1778) 2 Cowp 790, 792, Lord Mansfield expressed the point in this way:

Protective statutes are the plainest examples of rules of law which implicitly exclude the operation of the illegality principle, but they are not the only ones. Some statutes, on their proper construction, are inconsistent with the application of the illegality principle even if they are in no sense protective. The statutory prohibitions against the overloading of ships are wholly directed to the operational safety of ships and their crews. On that ground, among others, Devlin J held in St John Shipping Corpn v Joseph Rank Ltd [1957] 1 QB 267 that a breach of the Merchant Shipping (Safety and Load Line Conventions) Act 1932 did not justify shippers and bill of lading holders in defending an action for freight. For the same reason, the illegality principle has been held to have no application to claims to contribution under the Civil Liability (Contribution) Act 1978. The reason is that this would be inconsistent with the scheme of the Act: K v P [1993] Ch 140. In Stone and Rolls Ltd v Moore Stephens [2009] AC 1391, three members of the Appellate Committee, Lord Phillips, Lord Scott and Lord Mance, regarded the application of the illegality principle to an auditor’s negligence as turning on the purpose of the auditor’s statutory functions, although they reached different conclusions about what that purpose was.

The next question is whether the illegality principle bars an action for the recovery of the money which Mr Patel paid under the contract.

English law does not have a unified theory of restitution. Failure or absence of basis, which supplies such a theory in most civil law systems, was rejected as the overarching rationale of the law of restitution in Woolwich Equitable Building Society v Inland Revenue Comrs [1993] AC 70, 172 (Lord Goff). For the moment, therefore, as Lord Hoffmann observed in Deutsche Morgan Grenfell Group plc v Inland Revenue Comrs [2007] 1 AC 558, para 21, “the claimant has to prove that the circumstances in which the payment was made come within one of the categories which the law recognizes as sufficient to make retention by the recipient unjust.”

It is nonetheless true that failure of basis is the reason (or at least a reason) why the retention of a benefit is treated in some categories of case as unjust. One of these is the category of case in which a money benefit is conferred on the recipient under or in anticipation of a contract and the basis for that transfer has failed, for example by frustration, total failure of consideration or want of contractual capacity or vires on the part of one of the parties. As a general rule, benefits transferred under a contract which is void or otherwise legally ineffective are recoverable: Westdeutsche Landesbank Girozentrale v Islington London Borough Council [1994] 4 All ER 890 (Hobhouse J), approved (obiter) on appeal to the House of Lords [1996] AC 669, 681-682 (Lord Goff), 714 (Lord Browne-Wilkinson), 723 (Lord Woolf). In Guinness Mahon & Co Ltd v Kensington and Chelsea Royal London Borough Council [1999] QB 215, the Court of Appeal held that the ineffectiveness of the transaction was a ground of restitution independent of total failure of consideration, and therefore available even if the contract had been partly performed. The reason, as Morritt LJ observed (p 230) is that: “The bank did not get in exchange for that performance all it expected, for it did not get the benefit of the contractual obligation.”

One would expect the same reasoning to apply where the contract is unenforceable for illegality. In fact, however, the courts have not said this. The reason is that they have treated restitution as being available only where the payer was entitled to a locus poenitentiae in which to withdraw from the transaction. The breadth of this locus has varied with judicial fashion, but for much of the 20th century it was very narrowly interpreted indeed. This approach is not consistent with the recognition of a general right to the restitution of money paid under an illegal contract, in spite of the close analogy with other cases of ineffective contracts.

In one sense, the contract between these parties may be said to have been frustrated by the failure of the inside information to materialise, or to have resulted in a total failure of consideration because as a result the shares were never purchased. But that cannot be an adequate explanation of the reason why someone in Mr Patel’s position may be entitled to restitution even on the limited basis which the concept of a locus poenitentiae allows. That concept permits the recovery of money paid even before (indeed, especially before) the time for performance has arrived, and therefore in many cases before the contract was frustrated or the question of failure of consideration could arise. The ground of restitution in these circumstances can only be that the contract was illegal and that the basis for the payment had failed.

Of course, in order to demonstrate that the basis for the payment had failed, Mr Patel must say what that basis was, which would necessarily disclose its illegality. In my opinion, the reason why the law should nevertheless allow restitution in such a case is that it does not offend the principle applicable to illegal contracts. That principle, as I have suggested above, is that the courts will not give effect to an illegal transaction or to a right derived from it. But restitution does not do that. It merely recognises the ineffectiveness of the transaction and gives effect to the ordinary legal consequences of that state of affairs. The effect is to put the parties in the position in which they would have been if they had never entered into the illegal transaction, which in the eyes of the law is the position which they should always have been in.

The judges who first formulated the modern law of illegality at the end of the 18th century had no difficulty about this. In Smith v Bromley (1760) 2 Doug 696n, 697, one of Lord Mansfield’s earliest statements on this area of law, he thought that restitution of an illegal consideration was “necessary for the better support and maintenance of the law”. In Neville v Wilkinson (1782) Lord Chancellor Thurlow referred to this statement and “declared his opinion, that, in all cases where money was paid for an unlawful purpose, the party, though particeps criminis, might recover at law; and that the reason was, that if courts of justice mean to prevent the perpetration of crimes, it must be not by allowing a man who has got possession to remain in possession, but by putting the parties back to the state in which they were before”. This was the basis on which relief was granted, at any rate by Mellish LJ and Bagallay LJ, in Taylor v Bowers (1876) 1 QBD 291 traditionally regarded as the leading case, and by Lord Atkinson delivering the advice of the Privy Council in Petherpermal Chetty v Muniandi Servai (1908) LR 35 Ind App 98, 103.

In the course of the twentieth century, the law took a different and to my mind less satisfactory turn. The courts began to treat the right of restitution as depending on the moral quality of the plaintiff’s decision to withdraw. They reasoned that if the object of allowing restitution was to encourage withdrawal from an illegal venture, it ought to be withheld if the claimant had withdrawn involuntarily, for example because the other party withdrew first or the venture became impossible or failed for some reason other than his genuine regret. Although there are earlier traces of this notion, it is first overtly expressed in Parkinson v College of Ambulance [1925] 2 KB 1, 16, where Lush J suggested that there was no locus poenitentiae if there was no penitence. It may be said to have reached its high point in the three decisions in Alexander v Rayson [1936] 1 KB 169, Berg v Sadler & Moore [1937] 2 KB 158 and Bigos v Bousted [1951] 1 All ER 92. The concept of penitential withdrawal leads to difficult distinctions and suggests an enquiry into a party’s state of mind of a kind which the law rarely contemplates. It was rejected, rightly to my mind, by Millett LJ in Tribe v Tribe [1996] Ch 107, 135 “Justice is not a reward for merit”, he said: “restitution should not be confined to the penitent.” I agree. But for the same reason I would reject the suggestion that Millett LJ went on to make that the right to restitution should still depend on the voluntary character of the plaintiff’s withdrawal. As with the notion of penitence, this is to put a moral gloss on a principle that depends simply on the right to restitution that in principle follows from the legal ineffectiveness of the contract under or in anticipation of which the money was paid.

The courts’ view about when the right of restitution ceases to be available has closely reflected the way in which they have analysed that right. At the outset, and throughout the 19th century, they held that the right of restitution ceased in contract cases once the contract had been executed at least in part. The reason for this was that they viewed the right of restitution as arising from a principle analogous to rescission for mistake or misrepresentation. They therefore applied to it the then current doctrine that an executed contract could not be rescinded at law except for fraud. In Lowry v Bourdieu (1780) 2 Doug 468, 471, Buller J observed that in this context there was a “sound distinction between contracts executed and executory; and if an action is brought to rescind a contract, you must do it while the contract continues executory”. Lord Mansfield, who sat in that case, presumably agreed, for he had expressed the same view less expansively in Browning v Morris (1778) 2 Cowp 790. Later, when the courts came to regard the locus poenitentiae as depending on the moral quality of the plaintiff’s reason for resiling, they reframed the proposition so as to suggest that the right of restitution ceased to be available when the illegal purpose of the venture had been carried out. This might be the same as the point of time when the contract was executed. But it might be later, as in the numerous cases where a person nominally transferred his property to another with a view to defrauding his creditors. This test seems to me to be practically unworkable. Are we to distinguish between cases where the relevant representation was never made to the creditors and cases where it was but they did not believe it? More fundamentally, it proceeds from the same spurious moral gloss on the legal principle as the notion that the claimant’s withdrawal must have been voluntary or penitent. The rule against rescinding executed contracts has now gone, and the limitation to cases in which the unlawful purpose has not been carried out never was sound. The rational rule, which I would hold to be the law, is that restitution is available for so long as mutual restitution of benefits remains possible. In most such cases, the same facts will give rise to a defence of change of position.

I would also reject the dicta, beginning with Tappenden v Randall (1801) 2 B&P 467, 470 and Kearley v Thomson (1890) 24 QBD 742, 747, to the effect that there may be some crimes so heinous that the courts will decline to award restitution in any circumstances. There are difficulties about distinguishing between degrees of illegality on what must inevitably be a purely subjective basis. But the suggestion is in any event contrary to principle. If I pay £10,000 to a hitman to kill my enemy, he should not kill my enemy and should not have £10,000. The fact that when it comes to the point he is unwilling or unable to kill my enemy does not give him any legal or moral entitlement to keep the £10,000. If he does kill him, the rational response is the same. He should be convicted of murder, but he should never have received the money for such a purpose and by the same token should not be allowed to retain it. Of course, in practice, this is all rather artificial. In a case involving heinous crimes, both parties would be exposed to confiscation orders under the Proceeds of Crime Act 2002. St Thomas Aquinas thought the ideal solution to such a conundrum was that neither party should have the money, which should be paid to charity: Summa Theologica II.2, Q 62, para 5. The courts have no power to order that, but statute has now intervened to produce something like the same result.

I say nothing about cases in which an order for restitution would be functionally indistinguishable from an order for enforcement, as in a case of an illegal loan or foreign exchange transaction. The traditional view is that if the law will not enforce an agreement it will not give the same financial relief under a different legal label: Boissevain v Weil [1950] AC 327. I am inclined to think that the principle is sound, but I should prefer not to express a concluded view on the point. It is not the position here.

I can now return to the judicial schism to which I referred at the outset of this judgment.

A convenient starting point is the Supreme Court of Canada’s decision in Hall v Hebert, to which I have already referred. It is important to remember that the search for principle which led McLachlin J to identify consistency as the foundation of this area of law was a response to the judgment of Cory J in the same case. He had favoured a more flexible test for applying the illegality principle, which would have depended on whether the relevant public policy required that result on the facts of each case: see p 205. That approach was not accepted by the rest of the court. Part of McLachlin J’s concern about it arose from

In Tinsley v Milligan [1994] 1 AC 340, a similar view was taken by Lord Goff. I have cited extensively from this part of his speech in my judgment in Les Laboratoires Servier v Apotex Inc [2015] AC 430, para 16, and the exercise need not be repeated here. In summary, Lord Goff objected to a test for applying the illegality principle which would require the court to “weigh, or balance, the adverse consequences of respectively granting or refusing relief” (p 358E-F). The adoption of such a test, he considered, at p 363, “would constitute a revolution in this branch of the law, under which what is in effect a discretion would become vested in the court to deal with the matter by the process of a balancing operation, in place of a system of rules, ultimately derived from the principle of public policy enunciated by Lord Mansfield CJ in Holman v Johnson”. On this point, Lord Goff was supported by the whole of the Appellate Committee.

For many years, the chief critic of this approach was the Law Commission, which at one stage proposed legislation along the lines of the New Zealand Illegal Contracts Act 1970 to make the application of the illegality principle subject to a broad judicial discretion. More recently, Professor Burrows has proposed that the same solution should be adopted by judicial decision, in his Restatement of the Law of Contract (2016). He would make the application of the illegality principle dependent, at any rate in contract cases, on a “range of factors” approach. This would require the judge to assess whether to deny a remedy would be an “appropriate response” to the claimant’s conduct, taking account where relevant of eight factors. These factors are for the most part derived from the Law Commission’s Consultative Report (paras 8.3, 8.11). They are: (a) how seriously illegal or contrary to public policy the conduct was; (b) whether the party seeking enforcement knew of, or intended, the conduct; (c) how central to the contract or its performance the conduct was; (d) how serious a sanction the denial of enforcement is for the party seeking enforcement; (e) whether denying enforcement will further the purpose of the rule which the conduct has infringed; (f) whether denying enforcement will act as a deterrent to conduct that is illegal or contrary to public policy; (g) whether denying enforcement will ensure that the party seeking enforcement does not profit by the conduct; (h) whether denying enforcement will avoid inconsistency in the law, thereby maintaining the integrity of the legal system. Lord Toulson, in his judgment on the present appeal, supports this approach while suggesting that yet further factors may also be relevant.

With the arguable exception of (a) and (d) all of the considerations identified by Professor Burrows have been influential factors in the development of the rules of law comprised in the illegality principle as it stands today. Thus (b) is reflected in the requirement that, except where the making of the contract was itself illegal, there should have been some degree of participation by the claimant in the illegal act. It is also reflected in the exception for cases in which he was liable for the acts of another by virtue only of a rule imposing strict liability. As to (c), the purpose of the reliance test is to confine the illegality principle to cases in which the illegal act was truly central. Factor (e) is the basis of the exception discussed earlier in this judgment for cases in which the application of the illegality principle would be inconsistent with the legal rule which makes the act illegal, for example because its object is the protection of someone in the position of the claimant. It is also the basis on which claims are allowed for the restitution of money paid under an illegal contract. As to (f) and (g), there can be no doubt that historically the hope of deterring illegal conduct and depriving those responsible of any benefit arising from it have been important factors in the development of the illegality principle, although I personally doubt whether any but the best-advised litigants have enough knowledge of the law to be deterred by it. Factor (h), as I have suggested, is the most widely accepted rationale for the illegality principle in the modern law.

The real issue, as it seems to me, is whether the “range of factors” identified by the Law Commission and Professor Burrows are to be regarded (i) as part of the policy rationale of a legal rule and the various exceptions to that rule, or (ii) as matters to be taken into account by a judge deciding in each case whether to apply the legal rule at all. As matters stand, the former approach represents the law. The latter would require the courts to “weigh, or balance, the adverse consequences of respectively granting or refusing relief” on a case by case basis, which was the very proposition that the House of Lords unanimously rejected in Tinsley v Milligan. We are entitled to change the law, but if we do that we should do it openly, acknowledging what we are doing and assessing the consequences, including the indirect consequences, so far as we can foresee them. In my opinion, it would be wrong to transform the policy factors which have gone into the development of the current rules, into factors influencing an essentially discretionary decision about whether those rules should be applied. Neither party contended for such a result, and their reticence was in my view entirely justified. It would be unprincipled and uncertain, and far from confining the ambit of the illegality principle to its essential minimum, it could only broaden it beyond its proper limits. Perhaps most important of all, justice can be achieved without taking this revolutionary step.

The reason why the application of the “range of factors” test on a case by case basis is unprincipled is that it loses sight of the reason why legal rights can ever be defeated on account of their illegal factual basis. It is I think right to make four points:

Whatever rationale one adopts for the illegality principle, it is manifestly designed to vindicate the public interest as against the interests and legal rights of the parties. That is why the judge is required to take the point of his own motion even if the parties have not raised it, as the deputy judge did in this case. The operation of the principle cannot therefore depend on an evaluation of the equities as between the parties or the proportionality of its impact upon the claimant.

The “range of factors” test largely devalues the principle of consistency, by relegating it to the status of one of a number of evaluative factors, entitled to no more weight than the judge chooses to give it in the particular case. The criminal law, which is in almost every case the source of the relevant illegality, is a critical source of public policy. It is the prime example of the “positive law” (Lord Mansfield’s phrase) which has always moulded the law of illegality in civil proceedings. The courts cannot consistently or coherently recognise legal consequences for an act which the law treats as punishable. Gloster LJ, for example, thought it relevant that there was no finding that Mr Patel knew that insider dealing was illegal. Yet that would have been of no relevance in a criminal court, and it is difficult to see why it should be any more relevant in a civil one. Professor Burrows’ factor (f) (whether denying enforcement will ensure that the party seeking enforcement does not profit by the conduct) is surely fundamental to the principle of consistency, and not just a factor to be weighed up against others.

The main justification for the “range of factors” test has always been that it enables the court to avoid inflicting loss on the claimant disproportionate to the measure of his badness. This was the instinct that led the Court of Appeal in Euro-Diam Ltd v Bathurst [1990] 1 QB 1 to propose that the illegality principle should be applied only where the alternative would be shocking to the public conscience. That concept was rejected in Tinsley v Milligan. Since then, it has been suggested that there may be cases at the opposite end of the spectrum of gravity, in which the offence was too trivial to engage the illegality principle: see Gray v Thames Trains Ltd, at para 83 (Lord Rodger). One would expect most if not all such offences to be covered by the exception for cases in which the application of the illegality principle would be inconsistent with the legal rule which makes the act illegal. But, extremes apart, it is difficult to reconcile with any kind of principle the notion that there may be degrees of illegality, as Professor Burrows’ factor (a) seems to envisage. If the application of the illegality principle is to depend on the court’s view of how illegal the illegality was or how much it matters, there would appear to be no principle whatever to guide the evaluation other than the judge’s gut instinct. This was why this court recently rejected the view expressed by the Court of Appeal in Les Laboratoires Servier v Apotex Inc [2013] Bus LR 80 that an illegal act might nevertheless found a cause of action if it was not as wicked as all that.

The “range of factors” test discards any requirement for an analytical connection between the illegality and the claim, by making the nature of the connection simply one factor in a broader evaluation of individual cases and offering no guidance as to what sort of connection might be relevant. I have already observed that the reliance test is the narrowest test available. If it is no longer to be decisive, the possibility is opened up of an altogether wider ambit for the illegality principle, extending to cases where the relevant connection was remote or non-existent but other factors not necessarily involving any connection at all, were thought to be compelling.

The reason why the adoption of a “range of factors” test on a case by case basis would be uncertain is obvious in the light of these considerations. An evaluative test dependent on the perceived relevance and relative weight to be accorded in each individual case to a large number of incommensurate factors leaves a great deal to a judge’s visceral reaction to particular facts. Questions such as how illegal is illegality would admit of no predictable answer, even if the responses of different judges were entirely uniform. In fact, it is an inescapable truth that some judges are more censorious than others. Far from resolving the uncertainties created by recent differences of judicial opinion, the range of factors test would open a new era in this part of the law. A new body of jurisprudence would be gradually built up to identify which of a large range of factors should be regarded as relevant and what considerations should determine the weight that they should receive. No one factor would ever be decisive as a matter of law, only in some cases on their particular facts. The size of the authorities bundles in this and other recent appeals to this court on the illegality principle is testimony to the volume of litigation which the principle has generated in every period of its history. I do not suppose that those who are about to enter into an illegal transaction are in the habit of studying the decisions of the courts on the point, but those who advise them after the event do, and the resultant uncertainty is likely to generate a great deal of wasteful and unnecessary litigation. I would readily accept that certainty is not the only value, or even necessarily the most important. But we are concerned in this case with the law of contract, an area in which the value of certainty is very great. It is one thing to say that a legal right may be overridden by a rule of law. It is another thing altogether to make a legal right, and particularly a contractual right, dependent on a judge’s view about whether in all the circumstances it ought to be enforced.

Finally, I would point out that the adoption of such a revolutionary change in hitherto accepted legal principle is unnecessary to achieve substantial justice in the great majority of cases. The unsatisfactory features of the illegality principle as it has traditionally been understood have often been overstated, in part because of the way in which they were emphasised by Lord Goff in Tinsley v Milligan. It was, he said, “not a principle of justice; it is a principle of policy, whose application is indiscriminate and so can lead to unfair consequences as between the parties to litigation” (p 355B-C). That observation, however, reflected his view that no equitable interest in property could ever be claimed where the legal title had been vested in another for dishonest purposes. The law had been stated in this way by Lord Eldon at the beginning of the 19th century: see Muckleston v Brown (1801) 6 Ves 52 and Curtis v Perry (1802) 6 Ves 739. But Lord Eldon’s approach, although adopted by Lord Goff, was rejected by the majority of the Committee. When the law of illegality is looked at as a whole, it is apparent that although governed by rules of law, a considerable measure of flexibility is inherent in those rules. In particular, they are qualified by principled exceptions for (i) cases in which the parties to the illegal act are not on the same legal footing and (ii) cases in which an overriding statutory policy requires that the claimant should have a remedy notwithstanding his participation in the illegal act. Properly understood and applied, these exceptions substantially mitigate the arbitrary injustices which the illegality principle would otherwise produce. At the same time, the wider availability of restitutionary remedies which will result from the present decision will do much to mitigate the injustices which have hitherto resulted from the principle that the loss should lie where it falls.

For these reasons, I regret that I cannot agree with the conclusion of Lord Toulson (para 109) that that the application of the illegality principle should depend on

Against that background it is in my view entirely clear that the transaction into which these parties entered was affected by the illegality principle. The agreement pleaded, and found by the deputy judge to have been made, was not simply that Mr Mirza would place bets on movements of RBS shares for the joint account of himself and Mr Patel, but that he would do so with the benefit of inside information. Subject to immaterial exceptions, section 52 of the Criminal Justice Act 1993 makes it an offence for a person in possession of inside information to deal or encourage another person to deal in “securities”, including contracts for differences. This was accordingly an agreement for Mr Mirza to commit a criminal offence. It was also a criminal conspiracy to that end.

Section 63(2) of the 1993 Act provides that: “No contract shall be void or unenforceable by reason only of section 52.” The contracts affected by section 52 are contracts by way of dealing in securities. It follows that if Mr Mirza had placed the spread bets with IG Index, as he had conditionally promised to do, the contract would have been enforceable as between himself and IG Index. But Mr Patel could not have obtained specific performance of the distinct contract between himself and Mr Mirza or damages for breach of it. This is because, first, he would have had to rely on the contract, which provided as one of its terms that the dealing should be carried out with the benefit of inside information. Mr Patel could not have avoided this result by simply characterising it as an agreement to speculate in RBS shares without referring to the basis on which it was agreed that that should happen. Secondly, none of the possible exceptions apply. The parties were on the same legal footing. Both would be liable to conviction for conspiracy in a criminal court, and any difference in the degree of their fault would be relevant only to the sentence. Section 52 of the Criminal Justice Act 1993 is not a statute designed to protect the interests of persons entering into an agreement to commit the offence of insider dealing, and there is no other overriding statutory policy which requires their participation in the offence to be overlooked when it comes to determining its civil consequences.

However, restitution still being possible, none of this is a bar to Mr Patel’s recovery of the £620,000 which he paid to Mr Mirza. The reason is simply that although Mr Patel would have to rely on the illegal character of the transaction in order to demonstrate that there was no legal basis for the payment, an order for restitution would not give effect to the illegal act or to any right derived from it. It would simply return the parties to the status quo ante where they should always have been. The only ground on which that could be objectionable is that the court should not sully itself by attending to illegal acts at all, and that has not for many years been regarded as a reputable foundation for the law of illegality. This was Gloster LJ’s main reason for upholding Mr Patel’s right to recover the money. Although my analysis differs in a number of respects from hers, I think that the distinction which she drew between a claim to give effect to a right derived from an illegal act, and a claim to unpick the transaction by an award of restitution, was sound.

In the circumstances, Mr Mirza’s only arguable defence was that he had paid the money to Mr Georgiou, the intermediary who had proposed the deal. But the judge declined to make a finding to this effect, and rejected a defence of change of position on the ground that even if it was true, Mr Mirza had had no reason to repay the money to anyone but Mr Patel from whom he had received it.

The Court of Appeal gave judgment for Mr Patel for £620,000 with interest. For the reasons which I have given, which correspond to those given by Lord Mance and Lord Clarke, I would dismiss the appeal against that order.