As the quintessential deep-pocket defendant, financial institutions are often a target of litigation when victims of fraud are casting around to seek to recoup their losses. In 2017 and 2018 there have been a series of attempts to make financial institutions liable, in three common law jurisdictions (England, Cayman and Gibraltar), for dishonestly assisting in frauds. This article considers this unusual spate of recent cases, why they are generally unsuccessful, and the key lessons from the case law for financial institutions and their advisers.
The three relevant frauds were:
(1) In the English case of Singularis v. Daiwa  EWCA Civ 84, as the well-known frauds involving AHAB and the Saad group were being exposed by the credit crunch, the man at the centre of those allegations, Maan Al-Sanea, defrauded Cayman company Singularis of US$204 million, by instructing (as one of its directors and its sole beneficial owner), its brokerage, Daiwa, to pay Singularis’ funds to other companies in the Saad Group, in eight transfers over a period of six weeks. One of the transfers alone was for the sum of US$180 million.
(2) In the Cayman case of Ritter v. Butterfield, Grand Court, before Justice Williams on May 29, 2018, a local Cayman director, Mr. Self, had stolen US$875,000 from a captive insurance company’s bank account by forging his co-director’s signature on wire transfers on eight occasions over a period of two years. Mr. Self was subsequently convicted and imprisoned for theft for five years.
(3) In the Gibraltarian cases of Jyske and RBSI (Gibraltar Court of Appeal, Jan. 15 and June 12, 2018), the relevant fraud was that of the partners in a formerly well respected law firm, Marrache & Co. The three brothers who were principals in the firm had conspired to defraud clients by misusing their monies (which ought to have been segregated and held on trust in client accounts in the usual way) by treating those funds as the firm’s funds, or even the brothers’ own personal funds. Total losses as a result of the fraud were estimated to be GBP 28.4 million. The firm was quickly wound up, the three brothers were made bankrupt, and were each convicted and sentenced for terms of imprisonment of between 8 and 11 years.
In each case, the victims of the fraud or their representatives claimed that employees of financial institutions used by the company or partnership were sufficiently aware of a range of suspicious features of the various transactions such that they were acting dishonestly, and through them the financial institution was (vicariously) liable to account to the victims as a constructive trustee for this dishonest assistance.
Although, as is well known, that cause of action requires three elements to be satisfied: (1) breach of a fiduciary obligation or trust by the fraudster (2) assistance of the fraudster by the defendant and (3) that the assistance was dishonest in the way explained and developed in the leading cases of Tan, Twinsectra and Barlow Clowes; in reality, the battleground in all of these cases is whether anyone at the relevant financial institution had acted dishonestly.
In each of the four recent cases, the dishonesty claim ultimately failed.
(1) In Singularis, two employees of Daiwa were said to be dishonest, Mr. Metcalfe and Mr. Hudson, working in the brokerage’s credit risk and compliance departments respectively. It was said they had approved payments dishonestly by ignoring the many red flags on the account at the time, and without proper enquiry. The red flags included: a freezing order against Mr. Al-Sanea, the removal of the Saad Group’s credit ratings by the leading agencies, US$80 million having been received into the account days after the freezing order with no explanation, as well as deeply suspect documentation produced like “a rabbit out of a magician’s hat”. There was a suggestion they had dishonestly ignored all this as Singluaris was Daiwa’s most profitable relationship between 2007 and 2009. Despite approving the payments without inquiry in these circumstances, Messrs Metcalfe and Hudson were not held to have been dishonest: Mr. Hudson had not had it explained to him by management what other checks he ought to be performing other than working through his standard AML/terrorist financing and sanctions list queries. As for Mr. Metcalfe, concerns held by management in respect of the Singularis account had not been passed on to him, he was relatively inexperienced, he had no motive to act dishonestly, he reasonably trusted Mr. Al-Sanea and his advisers as honest and reputable people, and he had sought to involve others in the decision to approve the payments, such action being inconsistent with any dishonest intention.
(2) In Ritter, no individual employee of Butterfield was identified as having been dishonest. The implication, however, was that the Bank’s various payment processing staff who had approved each of the eight forged payment transfer requests had all acted dishonestly by ignoring the alleged suspicious nature of the payments, which included what were said to be obviously forged signatures and suspicious payment descriptions. At trial, a further and wider claim was added: that the Bank itself was structurally or systemically dishonest. The Court dismissed the dishonesty claim in its entirety as it had been defectively pleaded by not identifying who was alleged to have been dishonest, and on what particular basis, as well as for not following the strict guidance to pleading such claims set out in the leading case of Lipkin Gorman  1 WLR 1340 (CA) (dishonesty to be pleaded as the primary not alternative claim, with distinct and separate pleading). This was held to be in breach of basic procedural fairness. Having dismissed the claim on this basis, the Court went on to record that there was no evidence of dishonesty in any event, relying on the principle from Re H  AC 563 (HL), that since dishonesty allegations are more serious, more cogent evidence is required in support to establish them to the civil standard on the balance of probabilities. Specifically, the Court held that there was no reason to doubt the honesty of Mr. Self, who was thought to be a reputable insurance manager; that it was exceptionally rare for there to be fraud between two authorised signatories on an account; that none of the signatures were obvious forgeries once hindsight was stripped out; and that none of the payment descriptions were unusual. The Court accepted evidence from Mr. Skinner, the Bank’s Head of Corporate Banking, reliable as to banking practice, when he said that if he had been asked to approve any of the payments at the time, he would have done so. The context of the approval of the payments was also held to be of importance when assessing whether the Bank’s employees had acted improperly, viz. “where the staff members at the Bank have to process around 20-30 transfers per day for corporate clients and around 8,000 wire transfers per month for clients in all divisions of the Bank … appropriate validation enquiries cannot and do not require a detailed review and cross-reference of each and every transaction.”
(3) In Jyske, it was alleged that the account manager for the accounts of Marrache & Co and its associated companies was dishonest by being sufficiently on notice that the brothers were misusing client funds by paying client monies into the firm’s office account, by transferring sums from the office account to the client account, by ignoring large transfers from the client account to the office account to keep the latter within overdraft limits (putting client accounts into overdraft which were then brought into line by payments from a different client account); and by being aware of payments from client accounts to the Marrache brothers’ personal accounts and credit cards ranging from GBP 5,000 to 350,000. Although Mr. Bishop was held dishonest at first instance by Justice Jack, this was overturned on appeal as the judge had imposed monitoring expectations which went beyond the expert evidence of banking practice (generally to confirm proper authority, and review transfers for AML concerns). It was also held that the bank was entitled to trust a highly rated, respectable firm of lawyers, it being “virtually unthinkable” that a partner of law firm would be misappropriating client funds. The Court again referred to the “practical realities” facing Mr. Bishop in conducting his business (20-30 transfers a day, responsible for more than 250 customers with 300 separate accounts, with the impugned transactions being only 70 over a six-year period). Finally, it was held (1) that Justice Jack at first instance had ignored the “inherent improbability of Mr. Bishop having acted dishonestly, coupled with a complete lack of motive to do so”; and (2) that bank employees are not expected to act as policemen, forensically checking across multiple accounts for suspicious patterns.
(4) In RBSI, seven bank employees were alleged to be dishonest: three relationship managers, three credit managers and the Head of Corporate and Financial Institutions. Justice Jack held at first instance that one relationship manager, Mr. Shaw, had been dishonest as he had actually identified a need to investigate potential cross-firing of cheques and misuse of client funds, and had conducted an investigation which the judge held must have revealed the fraud, notwithstanding Mr. Shaw’s denials. The judge found that the reason Mr. Shaw did not report what he found, was in order not to “open a can of worms” for RBSI, and because he lacked the courage to act as a whistle-blower. On appeal, however, all of these findings were overturned. It was held that the basis upon which Mr. Shaw had been found to be dishonest had not been pleaded; and that the findings that Mr. Shaw had lied about deriving comfort in respect of the Marrache accounts from a previous internal investigation, as well as a meeting with the firm’s accountants, had been made on a demonstrably incorrect factual basis. The Court of Appeal also (1) held that Justice Jack had wrongly identified suspicious features on the account (bunching of payments) which neither of the banking experts had considered material (2) accepted Mr. Shaw’s evidence that he was under no obligation to continually monitor his customers’ accounts for suspicious activity (3) referred to the fact Mr. Shaw trusted the Marrache brothers as a respectable firm of solicitors and (4) held that it was an application of hindsight to say that his investigation should have been conducted more widely. Finally, the Court of Appeal found Justice Jack’s case theory on Mr. Shaw’s motive “surprising and implausible”, as “he had nothing to gain … [but] plenty to lose: his job, his reputation and his career.”
The real principles of importance in these claims are not the precise content or definition of the legal test for dishonesty (the subject of debate in extenso in the leading appellate cases concluding with Barlow Clowes). On any analysis, the most important principles guiding the ultimate results are the legal requirements for (1) proper pleading to give fair warning of the nature of the serious case being made against the employee and the financial institution, without which there will be fundamental procedural unfairness, and (2) suitably cogent evidence to establish such serious allegations against employees in financial institutions, who are in all probability not going to have acted dishonestly.
Well-known exceptional examples of staff at financial institutions actually acting dishonestly are not hard to bring to mind (e.g., Nick Leeson at Barings, Jerome Kerviel at SocGen and Kweku Adoboli at UBS). They invariably involve a combination of a clear personal financial motive and wholesale concealment by the rogue employee, features that were notably absent in the cases discussed above. For a recent decision in which the English Courts upheld a dishonest assistance claim against a financial institution, see Group Seven v Notable Services LLP  EWCA Civ 614;  EWHC 2466 (Ch) (in the High Court judgment in that case, Justice Morgan found the relevant banker to have been dishonest in providing misleading banking references in respect of an individual later convicted of money laundering offences. Following the analysis above of the factors which are usually present in successful claims, but absent in claims which fail, Justice Morgan found that the banker (1) was personally financially motivated to act in the way he had (noting that during cross-examination, he “accept[ed] that he expected to benefit financially”  and held that he acted as he did “with a view to receiving a substantial payment” ) and (2) had engaged in concealment from colleagues (the banker accepted at trial that he had mislead his colleague into believing that he had discontinued contact with the individual later found to have been engaged in money laundering, and for whom the banker had provided the dishonest references).
First, whilst the dishonesty risk is usually negligible, there is still – in cases where the claimants include the client of the financial institution – a risk of negligence liability for allowing payments to be made where the Bank was on enquiry as to possible misappropriation (the “Quincecare duty”). Liability was established on this basis for the first time in Singularis (per Vos LJ), even though the dishonesty claim was dismissed. The negligence risk can be minimised by ensuring any concerns in respect of clients are shared amongst all relevant employees, and by training staff on the Quincecare duty and what it requires of them (as seems not to have been done at Daiwa in Singularis). One way to seek to further build on that awareness by using financial institutions’ existing systems is to make clear (e.g. on the relevant reporting form) that when an employee makes an internal SAR to the MLRO, that if there is anything in the report which suggests there might be misappropriation by an authorised party from a corporate, partnership or trust account, payments from the relevant account must be stopped while inquiries are made.
Secondly, where staff have determined to investigate possible misappropriation (as did Mr. Shaw in RBSI) it is advisable for the legal and compliance function to be informed and involved, supported by external advice where justified in the circumstances, to ensure any investigation is conducted with a proper scope and with the relevant legal obligations clearly in mind.
Thirdly, brokerages have more risk than banks as they cannot plead the “practical realities” of the level of transactions that have to be processed, in the same way a bank can (per Singularis).
Finally, whilst it is accepted in the case law that financial institutions are not “detectives” or “policemen”, and that they are not expected to continually monitor across multiple accounts for signs of fraud, as automated systems are increasingly used to perform a wider kind of monitoring than was possible hitherto (say than in the 1980s when these principles were first established in the case law), this has the potential to increase the risk of liability, in particular if such reports are notified to relationship managers, or account managers (the employees who are most typically the target of these claims).
Sebastian Said is a Partner in the Dispute Resolution Practice Group of Appleby (Cayman) Ltd). He appeared as advocate for Butterfield in its successful defence in Ritter.