Dishonest Bankers – What is the Test?
by Charles Kuhn (Barrister) and Tom Bushnell (Paralegal) from Hickman & Rose
It is trite nowadays to say that the public doesn’t trust bankers. A 2013 YouGov-Cambridge report found that 58% of interviewees agreed that the industry was “at best unprofessional, and at worst dishonest”; whilst a study published in Nature magazine in November 2014 found bankers were more likely to cheat in a coin tossing game than other professionals.
Unsurprisingly, the FCA, in common with most regulators, has the power to ban bankers who are found to be “dishonest” or “lack integrity” (s 56 Financial Services and Markets Act 2000, FIT 1.3 and FIT 2.1). The legal test for the latter is tolerably clear (see, for example, the recent exposition in Tariq Carrimjee v Financial Conduct Authority  UKUT 0079 (TCC), FS/2013/0003, at  to ). However, its more serious cousin, dishonesty, remains the topic of disagreement. This article will contrast the potential tests; chart the existing authorities; argue that the criminal law’s “Ghosh test” is appropriate; and explore why there is a pressing need for certainty.
Nowadays, the civil law and criminal law adopt different tests for dishonesty. Criminal practitioners will be familiar with the “Ghosh test,” taken from Lord Lane’s judgment in R v Ghosh  2 QB 1053. It is a two stage test: would ordinary and reasonable people consider the action dishonest? If so, did the defendant realise that what he was doing was dishonest by those standards? It is also known as a “combined” test, for its use of objective and subjective limbs.
Civil lawyers have faced a more complicated history of authorities, many found in cases of dishonest assistance in breach of trust. The modern day position is, at heart, an objective test: would an honest person regard the conduct as dishonest, in light of the actual knowledge of the individual? (see Abou-Rahmah v Abacha  EWCA Civ 1492).
Which test has been chosen in the world of financial services regulation? The authorities are relatively clear in their endorsement of the combined test. The Financial Services and Markets Tribunal in Hoodless and Blackwell v FSA  FSMT 007 was encouraged by both sides to adopt such a test. The Tribunal’s Chairman considered briefly whether this approach was right in principle, and concluded it was.
This position was recently affirmed in Amir Khan v FSA FS/2013/002. Mr Khan was the subject of a prohibition, fine and withdrawal of approval, following his involvement in various mortgage applications with false or misleading information. He asked the Upper Tribunal to reduce the size of the fine, on grounds of serious financial hardship. The Tribunal needed to characterise various acts as dishonest, or reckless, or negligent. Judge Herrington said:
“It is common ground that the test for dishonesty to be applied in cases of this nature is the test established by Lord Lane in the criminal case of R v Ghosh  2 QB 1053.” (at )
The Tribunal did not explain what was meant by “cases of this nature,” but it would seem odd to apply one test for dishonesty in proceedings concerned with serious financial hardship, and another for other regulatory or disciplinary proceedings.
We understand, however, that the FCA’s Enforcement and Market Oversight Division holds a different view, and prefers the Abou-Rahmah objective test. How can the FCA’s apparent change of position, from Hoodless and Blackwell to today, be explained? The answer lies in the civil law position during 2003, when Hoodless and Blackwell was litigated. The leading civil law case at that time was Twinsectra Ltd v Yardley  UKHL 12, in which a majority of their Lordships appeared to prefer a combined test. This was cited by counsel for the Authority in Hoodless and Blackwell. In 2005, however, the Privy Council in Barlow Clowes International Ltd v Eurotrust International Ltd  UKPC 37 “reinterpreted” Twinsectra. It held that the appropriate test was objective. This was applied in Abou-Rahmah. This line of civil cases seems to have been followed by the FCA.
It is not immediately apparent why the FCA adopts this position. The Upper Tribunal (and its predecessor) has consistently preferred a combined test, despite ample opportunity to reconsider. The civil cases make no mention of the FCA, banking, or regulation.
The Ghosh test
The Ghosh test is desirable for four further reasons.
First, it would ensure coherence with the test laid down by the Upper Tribunal for a lack of integrity. Hoodless and Blackwell explained that lack of integrity represented a lower bar than dishonesty, and “A person lacks integrity if unable to appreciate the distinction between what is honest or dishonest by ordinary standards” (at ). This echoes the Ghosh test and is clearly designed to catch an individual who is not dishonest under the Ghosh test’s second limb (i.e. their conduct is objectively dishonest but they did not realise). To adopt the Abou-Rahmah objective test for dishonesty would render Hoodless & Blackwell virtually otiose.
Secondly, an objective test for dishonesty is unnecessary. The FCA has plenty of other mechanisms to ban individuals whose conduct falls short of the standard required, but which doesn’t carry an element of subjective culpability. For instance, as already discussed, integrity is an objective test and lesser standard than dishonesty. Similarly, a finding of a lack of competence (FIT 1.3 (2)) may be appropriate where an individual acts improperly, but without actually realising.
Moreover, dishonesty carries with it a greater degree of censure. A finding that an individual has behaved dishonestly is a judgement on a person’s character itself. It brings with it a heightened sense of moral criticism, and presumably makes it even harder for an individual to regain employment later. Accordingly, it must be right that the regulator should be unwilling to make such a finding without a subjective mens rea, i.e. assessing the person’s actual state of mind.
Finally, the Ghosh test is applied by other similar regulators considering whether to ban professionals – see, for example, the recent Court of Appeal decision in Hussain v GMC  EWCA Civ 2246. Similarly, the Divisional Court in Bryant v Law Society  EWHC 3043 (Admin) adopted a combined, Ghosh style test, arguing:
“it is more appropriate that the test for dishonesty in the context of solicitors’ disciplinary proceedings should be aligned with the criminal test than with the test for determining civil liability for assisting in a breach of a trust… the tribunal’s finding of dishonesty against a solicitor is likely to have extremely serious consequences for him both professionally (it will normally lead to an order striking him off) and personally. It is just as appropriate to require a finding that the defendant had a subjectively dishonest state of mind in this context as the court in R v Ghosh considered it to be in the criminal context.” (at ).
The reasoning is just as powerful when applied to those in financial services.
Why does it matter?
Clarity in this area of law is vital. Not merely because certainty on legal tests is always desirable, but because recent banking scandals, such as LIBOR and Forex, mean that questions of dishonesty will be at the forefront of the RDC’s (and potentially Upper Tribunal’s) minds.
Last year, the FCA issued a number of Warning Notices against individuals for benchmark manipulation. The FSA’s final notices against major banks for LIBOR manipulation made clear that the practices of “trader manipulation” (submitting figures to benefit a bank’s PNL) and “low-balling” (wholesale lowering of submissions to avoid market concerns over creditworthiness) were rife. Members of senior management knew about and condoned practices that the FCA and SFO now say were wrong. Hence individuals subject to FCA proceedings and criminal prosecution are likely to be running defences on the basis that their actions were not dishonest. Those before the RDC (and in time, Upper Tribunal) must know the test to apply. Respondents will undoubtedly argue that, in the climate at the time, their actions were not objectively dishonest. But the subjective limb of the Ghosh test would provide added support: the argument will surely be that a trader, immersed in an environment that did not even question these practices, did not realise that they were dishonest.
The dishonesty test is important at an earlier stage too: s 394(1)(b) FSMA 2000 requires the FCA to disclose material that “might” undermine their case. Under the Ghosh test, where a respondent’s subjective state of mind is crucial, the FCA will find it harder to resist calls by respondents for disclosure of the detailed evidence of a bank’s widespread conduct. It will not suffice (if indeed it ever did) for the FCA to merely “admit” that conduct was widespread without delving into the actual evidence.
The need for a clear, authoritative steer on the correct test for dishonesty is obvious. Signs that the Upper Tribunal endorses the criminal law’s Ghosh test are to be welcomed. It provides a standard that is clear, tried and tested, and commensurate with the severity of such an allegation.
 “Business culture and dishonesty in the banking industry,” Cohn, Fehr and Maréchal, Nature 516, 86-89.
 At .
 The FSA asked the Financial Services and Markets Tribunal to change its position in Vukelic v FSA FIN 0002/2008, and adopt an objective approach. The Tribunal declined to rule on the issue.
 In Williams v Financial Services Authority (2010) FIN/2010/0006, the Upper Tribunal relied upon previous findings of dishonesty made against Mr Williams in the High Court, applying Abou-Rahmah. However the findings of fact made by the High Court made is clear that Mr Williams would have been dishonest under Ghosh too, and the point doesn’t seem to have been argued.
 This passage has been routinely cited with approval, including recently in Tariq Carrimjee v Financial Conduct Authority (2015)
 Note Longmore LJ’s interesting dictum in which he said he was troubled by the Ghosh direction given to the GMC panel, saying “It would have been standard in a criminal case. But this was a professional disciplinary hearing and it seems to me that in future it would be right and proper for the first part of the direction to be adapted to read that the panel should decide “whether according to the standard of reasonable and honest doctors [not people] what was done was dishonest”.”
The views expressed by the Authors of this article are not necessarily those of the proprietor.