Proving dishonesty in a breach of trust (Group Seven Ltd v Notable Services LLP)

The Court of Appeal considered the test for dishonesty and specifically the application of that test in the context of dishonest assistance claims following the Supreme Court’s findings in Ivey v Genting Casinos (UK) Ltd.  This article comments on the implications of this decision.

Group Seven Ltd (a company incorporated under the laws of Malta) and another company v Notable Services LLP and another [2019] EWCA Civ 614, [2019] All ER (D) 02 (May)

What are the practical implications of this case?

This judgment is interesting if for no other reason than showing the impact of developments in appellate case law on ongoing matters being tried at first instance. Morgan J’s decision at first instance came some weeks prior to the key Supreme Court case of Ivey v Genting Casinos (UK) Ltd [2017] UKSC 67, [2018] 2 All ER 406. Although the Supreme Court’s decision in Ivey was based in part on earlier cases, including Barlow Clowes International Ltd v Eurotrust International Ltd [2005] UKPC 37, [2006] 1 All ER 333, many practitioners consider that Ivey clarified the law on dishonesty, but Morgan J obviously would not have had the benefit of the Ivey decision at the time he gave judgment in this case. At the very least, Ivey provided a clarification of the test and the process for applying it, which would have potentially avoided the ‘compartmentalised’ approach that the Court of Appeal considered Morgan J had erred in applying when considering dishonesty.

In addition, this decision of the Court of Appeal confirms the need to avoid that compartmentalised approach and confirms the need instead for the trial judge (and, therefore, those advising the parties in such a case) to ‘take a step back’ and ask the question, ‘in all the circumstances, was the defendant’s conduct honest or dishonest according to the standards of ordinary decent people?’. While ‘the circumstances’ will include what the defendant actually knew (or should have known or found out based on clear suspicions), the question remains a high-level, impressionistic one. As the Court of Appeal noted, it is essentially a jury question.

That does, of course, make it harder to advise in certain cases which might be thought to be borderline. There remains room for legitimate disagreement about whether conduct is dishonest ‘by the standards of ordinary decent people’—as, of course, it impliedly involves divining what those standards are in the first place.

What was the background?

This appeal primarily concerned claims for dishonest assistance in a breach of trust made by multiple claimants against multiple defendants (with some defendants also being claimants against other defendants). The first instance decision was by Morgan J in Group Seven Ltd v Nasir [2017] EWHC 2466 (Ch), [2017] All ER (D) 46 (Oct). It arose from what the Court of Appeal described as a ‘brazen’ fraud by which Allseas Group SA (Allseas), a company registered in Switzerland and the parent company of one of the claimants, was defrauded of €100m which it had transferred to its subsidiary, the first claimant, Group Seven LLP (Group Seven) in order to lend monies (unknowingly) to the fraudsters. Those monies were then laundered (or attempted to be laundered) through the client account of a London firm of solicitors, Notable Services LLP (Notable), which firm was a multi-disciplinary limited liability partnership and was one of the defendants.

Among the defendants were Martin Landman, an accountant and partner in Notable who facilitated the money laundering, and a bank employee, Othman Louanjli (employed by LLB Verwaltung (Switzerland) AG (LLB)), who dishonestly provided information to Notable in support of one of the main fraudsters. Other defendants included Notable and LLB who were accused of vicarious liability for the acts of Mr Landman or Mr Louanjli.

Notable had paid around €15m out of its client account, including £170,000 to a Panamanian company (Nisroy) for the benefit of Mr Landman. Police action had, however, managed to secure the recovery of circa €88m of the stolen monies.

Morgan J had found that:

• Mr Landman deliberately and knowingly broke the Solicitors Regulation Authority Accounts Rules 2011 (SRA AR 2011) and his conduct in relation to some requests for payments was dishonest. Objectively, he knew facts which would have shown an honest and reasonable man that the proposed recipient was not entitled to the €100m. However, he did not actually know or have ‘blind eye knowledge’ that the fraudsters were not the beneficial owners of the €100m nor entitled to use that money as if it were their own, and in the absence of such knowledge his conduct did not constitute dishonest assistance of the alleged breaches of trust and fiduciary duty. Notable and Mr Landman were not liable for dishonestly assisting a breach of trust. Mr Landman was, however, liable for unconscionable receipt of the £170,000 bribe paid to Nisroy
• Mr Louanjli dishonestly assisted in the fraudsters’ breach of trust and conspired to injure Group Seven by unlawful means. He had solid grounds for suspicion that the fraudsters had come by the money dishonestly and that he was assisting in money laundering. Mr Louanjli’s statements to Notable influenced Notable’s behaviour in a relevant way and assisted the fraudsters to commit a breach of trust
• the chain of causation between Mr Louanjli’s statements and the losses to Group Seven was not broken by Notable’s conduct in breaking SRA AR 2011 or by Mr Landman’s dishonesty
• LLB was vicariously liable to Group Seven for Mr Louanjli’s wrongdoing in dishonestly assisting the fraudsters’ breach of trust and in conspiring to injure Group Seven by unlawful means
• Mr Louanjli was liable to Group Seven for his unconscionable receipt of the sum of €561,860, being his share of a particular €1m paid out by Notable
• based on these findings, judgment was entered in the Group Seven proceedings against Mr Louanjli and another for €9,179,850.48 and against Mr Landman for £173,000

In fact, the initial recipient of the loan from Group Seven was a company called Larn Ltd which was owned and controlled by one of the fraudsters. Larn Ltd brought its own claims within the same proceedings, some of which also succeeded but which were not the subject of this appeal.

Various parties launched appeals, primarily arguing against Morgan J’s finding that Mr Landman and Notable were not liable for dishonest assistance, given the judge’s findings of fact about Mr Landman. Both LLB and Group Seven alleged, in effect, that the judge had mis-stated and mis-applied the test for dishonest assistance. In addition, Mr Louanjli appealed on the basis that the judge should have found a break in the chain of causation, and LLB appealed against the conclusion that it was vicariously liable for Mr Louanjli’s conduct. The Court of Appeal dismissed the latter two appeals. The appeal which caused the greatest discussion in this judgment concerned the test for dishonest assistance.

What did the court decide?

The dishonest assistance issue

By the time of the appeal, it was common ground that all but one of the necessary ingredients of such accessory liability on the part of Mr Landman were present (and that Notable would be vicariously liable if Mr Landman was). What was in issue was whether Mr Landman had been dishonest.

The court reviewed the modern (civil) law test for dishonesty, particularly as part of accessory liability, from Royal Brunei Airlines v Tan [1995] 2 AC 378, [1995] 3 All ER 97 through Twinsectra Limited v Yardley [2002] UKHL 12, [2002] All ER (D) 321 (Mar), Barlow Clowes and Starglade Properties Ltd v Nash [2010] EWCA Civ 1314, [2010] All ER (D) 221 (Nov), being the case law considered by Morgan J and also considered by the Supreme Court in Ivey. The latter judgment having been handed down on 25 October 2017, less than three weeks after Morgan J had delivered his judgment in this case.

The court stated that, in light of Ivey, it is now settled law that the touchstone of accessory liability for breach of trust or fiduciary duty is indeed dishonesty, and there is no room in the application of that test for the now discredited subjective second limb of the R v Ghosh [1982] QB 1053, [1982] 2 All ER 689 test. The subjective knowledge and state of mind of the defendant are not unimportant. The defendant’s actual state of knowledge and belief as to relevant facts forms a crucial part of the first stage of the test of dishonesty set out in Tan. But once the relevant facts have been ascertained, including the defendant’s state of knowledge or belief as to the facts, the standard of appraisal which must then be applied to those facts is a purely objective one. The court has to ask itself whether the defendant’s conduct was honest or dishonest according to the standards of ordinary decent people.

As to blind eye knowledge, it is not enough that the defendant merely suspects something to be the case, or that he negligently refrains from making further enquiries. As the House of Lords made clear in Manifest Shipping Company Ltd v Uni-Polaris Shipping Company Ltd [2001] All ER (D) 92 (Jan), the imputation of blind eye knowledge requires two conditions to be satisfied. The first is the existence of a suspicion that certain facts may exist, and the second is a conscious decision to refrain from taking any step to confirm their existence. The existence of the suspicion is to be judged subjectively by reference to the beliefs of the relevant person, and the decision to avoid obtaining confirmation must be deliberate.

Where the conditions for imputation of blind eye knowledge are satisfied, a person is treated for the purposes ofestablishing liability for dishonest assistance as if he had actual knowledge of the relevant facts. The court did not think it follows from this, however, that suspicions which fall short of constituting blind eye knowledge are wholly irrelevant to the question whether an alleged accessory has acted dishonestly. The first stage of the test, as it is now understood, requires the court to ascertain all the relevant facts, including the knowledge and beliefs of the defendant. Even though knowledge, in this context, must now be taken to be confined to actual and blind eye knowledge, there was no reason in principle why a person’s beliefs may not include suspicions which he harbours, but which in and of themselves fall short of constituting blind eye knowledge.

The existence of such suspicions, and the weight (if any) to be attributed to them, are then matters to be taken into account at the objective second stage of the test. Put another way, the existence of a legal technique for imputing constructive knowledge, if certain conditions are satisfied, should not be taken as implicitly restricting the scope of the subjective enquiry into a person’s state of mind and beliefs at the first stage. The state of a person’s mind is in principle a pure question of fact, and suspicions of all types and degrees of probability may form part of it, and thus form part of the overall picture to which the objective standard of dishonesty is to be applied.

The court noted the contrast between the judge’s findings in relation to Mr Meduri (another partner in Notable and a defendant to the claim but who the judge considered not to have been dishonest) and his findings in relation to Mr Landman, namely that:

• he was quite sure that Mr Landman had not been frank with the court about what was happening and what was discussed in the relevant period
• however, the judge’s suspicions as to the possibilities which there might have been during that period for Mr Landman to find out about the source of the €100m were simply not enough for him to make a finding on the balance of probabilities that Mr Landman knew that the money did not belong to Larn
• the judge felt compelled to make the same finding that he had regarding Mr Meduri, namely that Mr Landman believed the loan agreement to be ‘simply a paper transaction between connected companies to allow Larn to have the use of the money in the UK in a way which was tax efficient’
• the judge found assessing blind eye knowledge regarding Mr Landman to be difficult given Mr Landman’s apparent propensity for dishonesty. He also noted that Mr Landman had his own doubts about the underlying fraudster, Mr Nobre, that Notable via Mr Meduri and another had asked some questions about the source of the money. Mr Landman had ultimately allowed them to conduct such enquiries and that despite Mr Landman apparently persistently lying at the trial, and having been found to be dishonest in relation to certain authorisations of payments, he was not persuaded those factors would justify a conclusion on the balance of probabilities that Mr Landman had blind eye knowledge that the money was not beneficially owned by Larn or that Larn was not entitled to use it

The Court of Appeal considered there was a basis to distinguish between Mr Meduri and Mr Landman, namely that the latter had (indirectly) received a £170,000 bribe and that a corrupt payment of that sort could not have formed part of a legitimate tax planning strategy. The Court of Appeal also noted that Mr Landman had concealed the bribe from Notable (and others), a point not dealt with by Morgan J, which concealment facilitated the transaction proceeding (as it would not have done had others known of the bribe).

The court also noted that Morgan J had found Mr Landman liable for unconscionable receipt of the bribe of £170,000 being trust monies (resulting from breach of fiduciary duty) although the judge had found Mr Landman’s knowledge ‘of circumstances which would indicate the facts to an honest and reasonable man’ to be sufficient in that regard (pursuant to the case of Baden v Societe Generale [1993] 1 WLR 509, [1992] 4 All ER 161, notwithstanding the disapproval of the Baden knowledge spectrum/definitions in later cases). Essentially that although he had found that subjectively Mr Landman had thought that Mr Nobre was entitled to the money, objectively an honest and reasonable man would not have reached that conclusion. The Court of Appeal found it hard to reconcile Morgan J’s findings against Mr Landman on unconscionable receipt with his findings on dishonest assistance.

In fact, it said Morgan J’s final analysis on unconscionable receipt could not be faulted, but that in that analysis the judge was engaging in substantially the same two stage analysis as the law requires in relation to dishonest assistance, and finding by application of an objective test that an honest and reasonable person in Mr Landman’s position, and knowing the facts which he did, would have concluded that Mr Nobre was not entitled to the €100m. Accordingly, Mr Landman must have been objectively dishonest whatever he may have subjectively thought, and that the dishonesty extended not only to the payments which were made out of Notable’s client account, but also to the prior receipt of that sum into the client account. If an honest and reasonable person would have concluded that Mr Nobre was not beneficially entitled to the €100m, it would also have been obvious to such a person that Mr Nobre was seeking to use Notable’s client account to launder the money, and that any steps deliberately taken to facilitate that purpose would constitute dishonest assistance of a scheme intended in one way or another to defraud the true beneficial owner of the money.

Morgan J had erred by applying to Mr Landman the same compartmentalised approach as he had adopted in relation to Mr Meduri’s conduct, an approach which may have made sense regarding Mr Meduri, but not in relation to Mr Landman where the distinctions set out above, applied. It was impossible to insulate Mr Landman’s dishonesty in assisting the making of payments out of the account from his equally dishonest conduct in facilitating the initial receipt of the €100m into the account. If Morgan J had stood back and reviewed the position as a whole, the only conclusion to which he could reasonably have come to is that Mr Landman’s whole course of conduct in relation to the €100m, from late October 2011 onwards, was objectively dishonest.

The Court of Appeal was also satisfied that, in all the circumstances, Mr Landman had ‘blind eye knowledge’, having manipulated the process by which Notable had taken anti-money laundering advice, and could not see how Morgan J had concluded that the evidence pointed the other way.

While reminding itself of the need for appellate restraint, especially on issues of fact, the court concluded that the appeal rested not so much on the oral evidence at trial but on the other facts as agreed or established and/or not contested on the appeal.

Notable had asserted that, apart from the objective standard for dishonesty arising under Barlow Clowes and Ivey, there was nevertheless a minimum content of knowledge for there to be dishonest assistance, namely that the accessory must at least know, in broad terms, what is the nature of the underlying breach of trust, or in other words what the design of the principal wrongdoer is. However, having found that Mr Landman had ‘blind eye knowledge’, the Court of Appeal did not need to rule on this point, but gave a provisional view that the simplicity of the modern dishonesty test should not to be complicated by the introduction, as a matter of law, of a minimum content of knowledge which must be satisfied, for a number of reasons, including that it would not be practical to lay down any kind of meaningful test of that sort given the infinite variety of factual matrices that such cases involve.

The Court of Appeal therefore allowed the appeal on the dishonest assistance issue.

This article was first published in LexisNexis.