Keywords: criminal law – theft – fraud – dishonesty – Ghosh – Ivey.
In October 2017, the Supreme Court in Ivey overruled the Ghosh test for dishonesty, a test which has been a feature of the legal landscape for over 35 years. Dishonesty is an element of a number of offences which, collectively, are among those most prosecuted. Since Ivey was a breach of contract case, consideration of Ghosh as the standard of dishonesty in criminal cases was not required to decide the appeal. The Supreme Court took the opportunity to tackle Ghosh in Ivey because of a perception that the law was wrong. A lack of argument as to the basis and merits of Ghosh and the replacement test mean the decision in Ivey is open to criticism on a number of grounds.
In Ivey v Genting Casinos (UK) Ltd t/a Crockfords,1 the Supreme Court overruled the test for criminal dishonesty set out in R v Ghosh2 by Lord Lane. Dishonesty was first used as an element for some offences in the Theft Act 1968, most notably theft. It has since featured in other offences, most significantly those in the Fraud Act 2006. Why it took 50 years to settle what dishonesty means and why Ghosh has been overruled after 35 years are legitimate questions. In deciding to formulate a new test of dishonesty in Ivey, Lord Hughes undoubtedly sought to change the law to be fairer, simpler and more consistent. My argument here is that it is not clear this has been achieved and the lack of argument in Ivey as to the basis and merits of Ghosh may have led to what might be described as ‘judicial overreach’.
The Ghosh test consists of two limbs. The Crown Court Compendium – guidance for Crown Court judges – formulated the Ghosh direction as follows:
You must first decide whether you are sure that D’s behaviour was dishonest by the ordinary standards of reasonable and honest people. If you are sure it was, you must then ask whether D must have realised that it was dishonest by those standards. If you are sure about that as well, the prosecution will have proved that D acted dishonestly and your verdict will therefore be ‘Guilty’, whether or not D personally thought his behaviour was dishonest.
But if you are not sure that D’s behaviour was dishonest by those standards, or not sure that D realised that it was, the prosecution will not have proved that D acted dishonestly, and your verdict will therefore be ‘Not Guilty’.3
Although that direction is for juries, the principles are, of course, identical for a legal advisor assisting a bench of magistrates or a district judge directing him or herself. The effect of the first limb is to protect defendants with unusually high moral standards: their conduct is assessed against the ordinary standards of reasonable people. And the second limb protects defendants who have failed to recognise those standards, preserving the principle that only those with guilty minds should be punished.
In his judgement in Ivey, Lord Hughes sets out six ‘serious problems’ with Ghosh.4 These problems and the context in which they were addressed are discussed below. Before that, though, it is necessary to understand Lord Hughes’ new test of dishonesty. Even if one accepts there are serious problems with Ghosh, any new test is an improvement only to the extent it has fewer or less significant problems.
In Ivey, Lord Hughes approved Lord Hoffmann’s test of dishonesty in civil actions from Barlow Clowes International Ltd v Eurotrust International Limited:
Although a dishonest state of mind is a subjective mental state, the standard by which the law determines whether it is dishonest is objective. If by ordinary standards a defendant’s mental state would be characterised as dishonest, it is irrelevant that the defendant judges by different standards.5
As to the application of that test in criminal cases, Lord Hughes gave additional guidance in Ivey:
When dishonesty is in question the fact-finding tribunal must first ascertain (subjectively) the actual state of the individual’s knowledge or belief as to the facts. The reasonableness or otherwise of his belief is a matter of evidence (often in practice determinative) going to whether he held the belief, but it is not an additional requirement that his belief must be reasonable; the question is whether it is genuinely held. When once his actual state of mind as to knowledge or belief as to facts is established, the question whether his conduct was honest or dishonest is to be determined by the fact-finder by applying the (objective) standards of ordinary decent people. There is no requirement that the defendant must appreciate that what he has done is, by those standards, dishonest.6
It appears that in assessing whether the defendant was dishonest, the fact-finding tribunal takes account of a defendant’s knowledge or belief as to a fact or facts, save where that belief is as to what ordinary decent people consider dishonest. It is submitted that the ostensibly simple formulation in Ivey raises some difficult issues.
2. Ivey v Genting Casinos (UK) Ltd t/a Crockfords
Ivey was a breach of contract case. In August 2012, the professional gambler Phil Ivey started with a stake of £1 million and won almost £8 million from Crockfords Casino in Mayfair, London playing the Punto Banco variant of Baccarat. Crockfords refused to pay and Ivey sued. Punto Banco is a game of chance played with the house, which ordinarily enjoys a slight advantage. Ivey shifted the odds in his favour by edge-sorting or exploiting subtle visible differences on the edge of the playing cards to differentiate between ‘good’ and ‘bad’ cards before they were dealt from the shoe.
There was no need on the facts and arguments in Ivey to revisit Ghosh as the test of dishonesty where it is an element of criminal offences. Despite this, Lord Hughes engaged in a comprehensive review and concluded that the problems identified provided ‘convincing grounds for holding that the second leg of the test propounded in Ghosh does not correctly represent the law and that directions based upon it ought no longer to be given.’7 As the conclusion was obiter, it might not be followed. Since Ivey, however, the question has been addressed in DPP v Patterson8 and the November 2017 update to the Crown Court Compendium.
In Patterson, Sir Brian Leveson, President of the Queen’s Bench Division and Head of Criminal Justice, noted that Lord Hughes’ observations were obiter but said ‘Given the terms of the unanimous observations of the Supreme Court expressed by Lord Hughes, who does not shy from asserting that Ghosh does not correctly represent the law, it is difficult to imagine the Court of Appeal preferring Ghosh to Ivey in the future.’9 The Crown Court Compendium has been updated to include Ivey and states: ‘Directions based on Ghosh should no longer be given’ and ‘Judges will want to consider the guidance given at para 74 of Ivey, which sets out an alternative two-stage test with subjective and objective elements.’10
It is unfortunate that a significant change in the criminal law results from a civil case where the issue was not relevant to the outcome of that case. With the arrival of Supreme Court Live and recordings of hearings, the Ivey the parties’ submissions can be watched.11 While the parties’ contentions were skilfully argued, there were no submissions directed at persuading the Court that Ghosh should remain the test of dishonesty in criminal cases. If one accepts the, presumably uncontroversial, proposition that submissions by counsel help judges reach good decisions, it is unfortunate there was no argument in defence of Ghosh. And none of the good arguments for Ghosh appear in Lord Hughes’ judgement.
In Ivey, Lord Hughes states that the possibility of looking again at Ghosh is ‘unlikely to occur in a criminal case whilst Ghosh remains binding on trial judges throughout the country.’12 While this is true, it does not follow that it was a good idea to revisit Ghosh in a case where it was not relevant. A useful contrast is R v G13 which overruled R v Caldwell14 after 20 years to replace a test of objective with subjective recklessness for the offence of criminal damage under the Criminal Damage Act 1971. R v G was a criminal appeal and the competing arguments for the change in the law were comprehensively addressed, including the fundamental issue of what Parliament intended when it created the offence. Ivey is striking in that it contains no analysis of what Parliament meant when it used the word ‘dishonestly’ in the Theft Act 1968 and, perhaps more significantly, when it used the word again in the Fraud Act 2006, enacted 25 years after Ghosh had become settled law.
3. The divergence in the test for dishonesty between criminal and civil proceedings
While the intention of Parliament as to dishonesty was not considered in Ivey, an aspect of the common law was significant. One of the problems identified with Ghosh by Lord Hughes was:
there can be no logical or principled basis for the meaning of dishonesty (as distinct from the standards of proof by which it must be established) to differ according to whether it arises in a civil action or a criminal prosecution. Dishonesty is a simple, if occasionally imprecise, English word. It would be an affront to the law if its meaning differed according to the kind of proceedings in which it arose.15
The civil cases cited by Lord Hughes in Ivey relate to breach of trust or fiduciary obligations by third parties – directors, bankers, etc – on a company insolvency. They are cases where claimants rely on equity to obtain relief from others where the company cannot pay because it is insolvent. The foundation case is Barnes v Addy where Lord Selborne LC stated accessory liability for breach of trust required knowing assistance ‘in a dishonest and fraudulent design on the part of the trustees.’16
It seems the primary conflict between the test for dishonesty in criminal and cvil proceedings arises from the use of the words ‘dishonestly’ in the Theft Act 1968 and ‘dishonest’ in an equitable case from 1874. There is no mention of Barnes in the pre-legislative material to the Theft Act 1968, nor in Hansard. It is most unlikely that Parliament intended to adopt a 19th century equitable case for a criminal offence and, in any event, it was only in Barlow Clowes that the Privy Council settled upon an objective test of dishonesty for Barnes cases.17 It is arguable that the primary conflict between the two tests of ‘dishonesty’ is a historical accident. Had Lord Selbourne settled on, say, ‘a deceitful and fraudulent design’ the problem would not have arisen. It is submitted that reinterpreting a 20th century criminal statute to resolve a conflict with a 21st century interpretation of a 19th century equitable case is neither a constitutionally nor logically sound basis for making law.
Once Parliament has ‘spoken’ in an Act, interpretation and application – giving the law its proper effect in actual cases – is for the courts. Since the Theft Act 1968, however, Parliament has spoken again by creating further offences which contain dishonesty as an element. Most significantly, there are the offences in the Fraud Act 2006. That Act was preceded by a Law Commission report, a bill with explanatory notes and debate in Parliament. The Law Commission’s Fraud report analyses dishonesty in detail.18 The report expressly approves the Ghosh test and notes the difficulty of adopting an alternative concept to dishonesty. Much the same could be said about adopting an alternative test for dishonesty. The report sates:
5.11 The second limb [of Ghosh] imposes an important brake on what might, despite its express terms, tend to be a subjective approach to the first limb decision. First it prevents naive or innocent defendants from being found to be dishonest when the jury is not satisfied that they must have recognised that their behaviour fell outside the norms of reasonable honest people. On the other hand it operates as a brake on the jury acquitting by virtue of the ‘Robin Hood’ defence.
5.18 … Many years after its adoption, the Ghosh test remains, in practice, unproblematic. We also recognise the fact that the concept of dishonesty is now required in a very large number of criminal cases, so to reject it at this stage would have a far-reaching effect on the criminal justice system.
The explanatory note to section 2 of the Fraud Act 2006 states:
Section 2 makes it an offence to commit fraud by false representation. Subsection (1)(a) makes clear that the representation must be made dishonestly. This test applies also to sections 3 and 4. The current definition of dishonesty was established in R v Ghosh  QB 1053. That judgment sets a two-stage test. The first question is whether a defendant’s behaviour would be regarded as dishonest by the ordinary standards of reasonable and honest people. If answered positively, the second question is whether the defendant was aware that his conduct was dishonest and would be regarded as dishonest by reasonable and honest people.19
It is not clear whether the use of ‘current definition’ suggests either an acceptance that the test may change or that the Bill is adopting the ‘current definition’. As might be expected, Hansard contains a range of not wholly consistent statements made by the minister responsible for the bill, the Attorney General, Lord Goldsmith. At the committee stages of the bill in the House of Lords, however, Lord Goldsmith said:
Perhaps I may interpolate one word on the noble Lord’s observation about the emphasis being placed on dishonesty. I am placing emphasis on dishonesty, as the Government do and as the Law Commission also did, and applying the test which arose from the case of Ghosh. The noble and learned Lord, Lord Lloyd of Berwick [one of the three judges in Ghosh], referred to it at Second Reading, and it is supported strongly and applied throughout the courts in the country. We are entirely content that that test is workable in practice and can be determined by juries and other fact-finders. When the noble Lord studies Hansard, he will find that I have used both ‘juries’ and ‘fact-finders’ when I have been talking about this.20
As I have said, there is a shared understanding of what ‘dishonesty’ means, thanks to the case of Ghosh. I remind noble Lords that the Ghosh test requires some subjective awareness by the offender that what he did was not in accord with the standards of ordinary people. This is an essential test for criminal liability for fraud. To remove the requirement for dishonesty would result in a criminal offence that is too widely drawn.21
Ivey changes the meaning of dishonesty for all offences which include it as an element, including the offences in the Fraud Act 2006. The above material provides grounds for arguing that Parliament intended the test for dishonesty for the Fraud Act 2006 offences to be Ghosh. Those arguments were not made in Ivey. If they had been made and found favour, overruling Ghosh was not an option for the Supreme Court.
4. Ghosh allows defendants with ‘warped’ standards to be acquitted
It is important to be clear about the second limb of Ghosh. The question is: ‘Must the Defendant have realised that his conduct was dishonest by the standards of ordinary honest people?’ which is different to ‘Did the defendant think his conduct was dishonest?’ In Ivey, Lord Hughes finesses the gap between those two questions, concluding that defendants with a warped view of what is dishonest are likely to have a warped view about what reasonable people think is dishonest. He states:
… it is not in the least unusual for the accused not to share the standards which ordinary honest people set for society as a whole. The acquisitive offender may, it is true, be the cheerful character who frankly acknowledges that he is a crook, but very often he is not, but, rather, justifies his behaviour to himself. Just as convincing himself is frequently the stock in trade of the confidence trickster, so the capacity of all of us to persuade ourselves that what we do is excusable knows few bounds.22
While there is force in this, it does not fully recognise the effect of the second limb of Ghosh. It is one thing to persuade a jury you believe your view on morality is right, but a different proposition to persuade a jury you believe ordinary people believe your view on morality is right. A person might hold the view that contraception is morally wrong. No doubt a jury may accept that belief as completely honest and sincere. But persuading a jury that you did not realise that ordinary people do not share your view is not the same. A jury is likely to require a cogent explanation as to how you arrived at your mistaken view of contemporary morality. And, for a defendant seeking to rely on the second limb of Ghosh, that evidence must in practice come from the defendant and withstand cross-examination.
The second limb of Ghosh is actually self-limiting in practice: the greater the divergence between a defendant’s conduct and societal norms of honesty, the harder it becomes for the defendant to rely on the second limb of Ghosh. The more warped the conduct, the less likely it is that the defendant will be believed and the more likely it is he will be convicted. People, cheerful or otherwise, who steal purses from old ladies’ shopping bags having convinced themselves it is excusable get no change out of juries or magistrates.
In Ivey, Lord Hughes cites two cases to support the view that juries cope well with applying the objective limb of Ghosh: R v Hayes,23 a LIBOR rigging case; and R v Rostron,24 involving the removal of golf balls at night from the bottom of a lake on a golf course. Both cases could be cited equally as support for the proposition that juries do not acquit defendants with warped standards. In both cases the jury had to determine whether the defendants must have realised their conduct was dishonest. The defendants were convicted and there is nothing to suggest the juries were ‘taken in’ by the defendants’ claims that they did not realise their behaviour was dishonest.
In R v G, referred to above, Lord Bingham addresses the question of a defendant making an unlikely assertion in relation to his state of mind:
There is no reason to doubt the common sense which tribunals of fact bring to their task. … with recklessness: it is not to be supposed that the tribunal of fact will accept a defendant’s assertion that he never thought of a certain risk when all the circumstances and probabilities and evidence of what he did and said at the time show that he did or must have done.25
The same can be said for the assertion ‘I never thought ordinary people would think my conduct dishonest’ in the context of an allegation of theft or fraud. The common sense of juries and magistrates prevent defendants seeking to rely on ‘warped standards’ being acquitted.
Further, the Law Commission’s Fraud report – based on extensive consultation within the criminal justice system – does not record any concern that defendants were getting away with criminal offences. It is submitted the problem of defendants taking advantage of the second limb of the Ghosh test to secure their acquittal is more theoretical than real.
Some questions of dishonesty will continue to be resolved without reference to Ivey. Section 2(1) of the Theft Act 1968 sets out three instances where an appropriation is not to be regarded as dishonest. These are where the defendant appropriates property with the belief: (a) he has the legal right to appropriate it (a claim of right); (b) he would have the other’s consent to the appropriation; or (c) the person to whom the property belongs cannot be discovered by taking reasonable steps. Ivey does not change the position that a belief under section 2(1) is subjective and does not have to be reasonable. Even if one accepts Lord Hughes’ point about defendants with warped standards, Ivey cannot, therefore, wholly solve the problem. The defence in Rostron could have been based on the belief that the person to whom each golf ball belonged could not be discovered and therefore the defendants were entitled to take them. With Ivey, section 2(1) now assumes greater significance and is likely to generate appeals as to its scope.
Indeed, Ivey opens up a potential lacuna between theft and fraud: section 2(1) of the Theft Act 1968 only applies to offences where there is an appropriation, so it has no application to fraud offences under the Fraud Act 2006. It seems clear from para 7.66 of the Law Commission’s Fraud report that it decided not to recommend the inclusion of a ‘claim of right’ for dishonesty in fraud because Ghosh was sufficient to ensure justice was done. So: while a ‘claim of right’ avoids dishonesty for theft, it does not have the same effect for fraud. An obvious illustration relates to a disreputable employer who sacks an employee for demanding the minimum wage. The 16 year old employee who leaves with £50 from the till honestly believing she has the right to take the money to bring her pay up to the minimum wage is not dishonest under section 2(1). But: the 16 year old employee who transfers herself £50 from her employer’s bank account believing the same has no claim of right and her honesty is to be determined by Ivey. The problem of a defendant’s knowledge or belief for the purposes of Ivey is addressed below. For present purposes, however, it is sufficient to note that if a claim of right is founded on a belief that ordinary people would not consider the conduct dishonest, it may be no defence under Ivey.
5. Ghosh was a departure from the pre-Theft Act 1968 law.
Lord Hughes’ argument seems to be that the replacement of ‘fraudulently’ in the Larceny Act 1916 by ‘dishonestly’ in the Theft Act 1968 was intended to shift responsibility for deciding whether the element was proven from the judge to the jury; but it was not intended to change how the element was determined. Lord Hughes states in Ivey:
Prior to the Theft Act 1968, the expression “dishonestly” had not appeared in the legal definition of acquisitive offences. The mental element was usually marked by the expression “fraudulently”. There is no doubt that that latter expression involved an objective evaluation of the defendant’s conduct, given his actual state of knowledge and belief as to the facts. The Criminal Law Revision Committee, in its eighth report, advised the substitution of the word “dishonestly”, on the grounds that “fraudulently” had become technical and its meaning had departed somewhat from the ordinary understanding of lay people. At para 39 the Committee advised that
“‘Dishonestly’ seems to us a better word than ‘fraudulently’. The question ‘Was this dishonest?’ is easier for a jury to answer than the question ‘Was this fraudulent?’. ‘Dishonesty’ is something which laymen can easily recognize when they see it, whereas ‘fraud’ may seem to involve technicalities which have to be explained by a lawyer.”
It was in accordance with this substitution that in Feely a five-judge Court of Appeal, Criminal Division, held that the question whether a defendant had behaved dishonestly was to be left to the jury and should not, as had been the case with “fraudulently”, be the subject of judicial ruling. But there is no hint in the Committee’s report of any contemplation that whether a man was or was not dishonest should henceforth depend on his own view of his behaviour. On the contrary, the report clearly assumed that the prior objective approach would continue, save that the question would be a jury matter rather than one of law.26
Lord Hughes again does not do justice to the second limb of the Ghosh test in stating that a defendant’s dishonesty depends on ‘his own view of his behaviour’. And while his point as to the replacement of ‘fraudulently’ by ‘dishonestly’ has some force, there is support for the view that a significant departure from the pre-Theft Act 1968 law is precisely what was intended, not least since the paragraph from eighth report of the Criminal Law Revision Committee cited by Lord Hughes contemplates a different test and not only a shift in decision-making responsibility. The report also states the draft bill is ‘based on a fundamental reconsideration of the principles underlying this branch of law embodied in a modern statute.’27 At the second reading in the House of Lords, the minister responsible for the Theft Bill, Lord Stonham stated:
When they came to examine the existing law on larceny and related offences, the Criminal Law Review [sic] Committee decided that it would be necessary to make far-reaching changes, not merely by amendments to the present law but by introducing a completely new code to cover the relevant offences. On all the most important of these they found that the law was so defective that they had to go back to first principles — to consider what were the essential elements of the offence, and reconstruct the law on that basis.28
And in R v Feely, Lord Justice Lawton stated:
“Theft” itself is a word known and used by all and is defined in what the marginal note to section 1 of the Act of 1968 describes as the basic definition as follows:
“A person is guilty of theft if he dishonestly appropriates property belonging to another with the intention of permanently depriving the other of it; …”
These words swept away all the learning which over the centuries had gathered around the common law concept of larceny and in more modern times around the statutory definition of that offence in section 1(1) of the Larceny Act 1916.29
The full definition of larceny was ‘A person steals who, without the consent of the owner, fraudulently and without a claim of right made in good faith, takes and carries away anything capable of being stolen with intent, at the time of such taking, permanently to deprive the owner thereof.’30 Clearly, there is some continuity of wording between that offence and theft, notably of the intent to permanently deprive. An obvious argument is that the Criminal Law Revision Committee and Parliament retained the same words where the intention was to retain the same concepts and, conversely, adopted different words where different legal concepts were intended. It is more than arguable that ‘dishonestly’ was intended as a departure from the pre-Theft Act 1968 law.
Finally, even if one accepts the view that a departure from the pre-Theft Act 1968 law was not intended, it is difficult to see how Ivey returns the law to the earlier course. The test in Ivey which takes account of a defendant’s knowledge and belief is not a return to a straightforward objective test. Nor is it a matter for the judge. And, after 35 years, Ghosh has as good a claim to pedigree as the pre-Theft Act 1968 case law.
6. It sets a test which jurors and others often find puzzling and difficult to apply
The basis of this conclusion is unclear. As noted above, the LIBOR rigging case referred to by Lord Hughes, Hayes, required the jury to engage with both limbs of the Ghosh test. The jury convicted the defendant after a trial of 47 days with complex evidence on the defendant’s conduct to manipulate the LIBOR rate to benefit his and his employer’s trading position. Hayes’ conduct seems to have been tolerated market practice. Despite this, there was no suggestion on appeal that the jury were puzzled or had difficulties with the question of whether Hayes must have realised his conduct was dishonest by the standard of ordinary people despite it being market practice.
The Law Commission’s Fraud report was based on extensive research and consultation. Para 5.18 is set out above and records that Ghosh remained unproblematic many years after its adoption. If juries do struggle with the test, one obvious risk is inconsistent verdicts. At para 5.8 of the report, the Law Commission’s view is:
On the other hand, we are not aware of any research or evidence to show that verdicts are in fact significantly inconsistent when dishonesty is a live issue in a case.
And para 5.14 records the response of the Magistrates’ Association that the requirement of Ghosh dishonesty was unproblematic. The Law Commission chose Ghosh for the fraud offences in the Fraud Act 2006 because it was not puzzling or difficult.
Even if one accepts Ghosh is puzzling and difficult, it is not clear that Ivey is less puzzling or difficult. The Ivey direction in the Crown Court Compendium is:
You must first consider all the circumstances in which the behaviour occurred, including what D himself knew or believed to be the factual situation. Have that in mind when you ask yourselves whether, in light of any understanding of the situation he had (or may have had), you are sure that D’s behaviour was dishonest by the standards of ordinary decent people.
If you are sure it was, the prosecution will have proved that D acted dishonestly and your verdict will therefore be ‘Guilty’, whether or not D personally thought his behaviour was dishonest.
But if you are not sure that D’s behaviour was dishonest by those standards, the prosecution will not have proved that D acted dishonestly, and your verdict will therefore be ‘Not Guilty’.31
It seems Ivey may require greater mental gymnastics than Ghosh in that the jury may have to apply the values of ordinary honest people to someone with irrational or unreasonable beliefs. The task is not identical, but there are parallels with the problematic provocation direction in that the jury has to conceptualise an ordinary reasonable person with a state of mind as to knowledge or belief as to facts which no reasonable person could have.
There is also the question as to what knowledge or belief is relevant to the Ivey test. In Ivey, Lord Hughes repeated from Ghosh the example of a man who comes to the UK from a country where public transport is free. On his first day he travels on a bus without paying. Lord Hughes concludes this conduct is not objectively dishonest because the man genuinely believes that public transport is free. The test of dishonesty is objective, but takes account of the defendant’s actual knowledge or belief.
That illustration is not altogether sound. The most suitable charge for this type of conduct is the offence of obtaining services by deception under section 11 of the Fraud Act 2006. The example does not necessarily engage dishonesty on the facts because the offence requires the defendant to know, at the point of obtaining the service, that the service is made available on the basis that payment has been, is being, or will be made. Plainly, the man does not know this. His knowledge (or lack of) is a defence irrespective of his honesty.
R v Hayes – the LIBOR rigging case referred to above – illustrates the issues and how Ivey might apply to this type of case is addressed in the Crown Court Compendium. Hayes was convicted of eight counts of conspiracy to defraud by manipulating the Japanese Yen LIBOR Rate. The prosecution case was that the defendant, together with others, manipulated LIBOR to advance his trading interests and the profits of the bank for which he worked. His submissions to the LIBOR setting process moved the rates to his and his bank’s advantage.
The defendant’s case at trial engaged both limbs of the Ghosh test: he had not acted dishonestly by the ordinary standards of reasonable honest people; and he did not realise what he was doing would be considered to be dishonest by reasonable honest people.
The defence relied on evidence relating to: the ethos of the banking system at the time; the prevalence of ‘commercial’ LIBOR submissions in banking generally; the attitude of the British Banking Association (BBA), which operated LIBOR and which knew of the association between LIBOR submissions and the banks’ ‘commercial’ positions; and the attitude of the Bank of England and the Financial Services Authority towards LIBOR despite knowing that the benchmark suffered from flawed governance and the rate was not accurate. The nub of the defendant’s case seems to have been that his conduct was market practice, such practice being, at least tacitly, tolerated by his employer, other banks, the BBA and the market regulators.
At trial, Cooke J ruled as a question of law that the defence could not, in closing submissions, refer the jury to the defence evidence as relevant to the objective first limb of the Ghosh test. The defence evidence was not relevant to the standards of ordinary and reasonable people. The ruling was challenged on appeal and the clear and robust view of the Court of Appeal (Lord Thomas CJ – who also sat on Ivey, Sir Brian Leveson P and Gloster LJ) was that if objective standards of honesty were set by a market, that would gravely affect the proper conduct of business. From time to time markets adopt behaviours which are dishonest by the standards of honest and reasonable people. If standards of honesty were set by the market and not ordinary honest people, this would undermine the honesty and integrity essential to the conduct of business.32
The Court of Appeal’s decision recognises that if the jury take account of the knowledge and belief of the defendant in applying the objective first limb of the Ghosh test, it compromises it as an objective test of dishonesty. With Ivey, however, it seems the jury could take account of the defendant’s knowledge and belief as to market practice, and this opens the door to the problem recognised in Hayes.
The Crown Court Compendium seeks to resolve the problem with the following direction in a Hayes type case:
In order for you to be sure of D’s guilt, you need to be sure that he was acting dishonestly. You will consider the evidence about the culture and ethos of the market together with all other evidence, including that from D himself, which may have a bearing on what D knew or believed to be the factual situation in which he acted.
Taking account of D’s understanding of the factual situation in which the behaviour occurred, you must ask yourselves this question: Was what D agreed to do with others dishonest by the ordinary standards of reasonable and honest people? I will say that again: Was what D agreed to do with others dishonest by the ordinary standards of reasonable and honest people? Not by the standards of the market in which he operated, if different. Not by the standards of his employers or colleagues, if different. Not by the standards of bankers or brokers in that market, if different, even if many, or even all regarded it as acceptable; nor by the standards of the various market Regulators, but by the standards of reasonable, honest members of society.
There are no different standards [of honesty] which apply to any particular group of society, whether as a result of market ethos or practice. You must form your judgment as to what those standards are in the light of the arguments that have been put before you.33
The direction appears to ask the jury to reconcile the irreconcilable. In a Hayes type case, the jury may decide a defendant believed his conduct was entirely in accordance with accepted market practice and not dishonest. This is the defendant’s understanding of the factual situation of which the jury may ‘take account’, but then apply the standards of reasonable honest people without taking account of the defendant’s belief. It is submitted that a jury is likely to be puzzled by this direction. The fundamental problem is the conflict between Ivey being an objective test while seeking to take account of defendant’s knowledge and belief in applying the test. The obvious risk is that the jury will apply an objective standard irrespective of the defendant’s knowledge and belief and convict him.
As noted above, the illustration in Ivey and Ghosh of the visitor from a country where public transport is free is problematic. Perhaps a better example is the visitor who comes from a country where husbands have control over their wives’ finances. Shortly after arriving in the UK, the man takes his wife’s credit card without her consent to go shopping. He enters a shop, selects goods to the value of £30, presents them at the checkout and makes contactless payment. The payment is processed off-line later that day and refused.
The most relevant chargeable offence is fraud by false representation under section 2 of the Fraud Act 2006. The false representation is made to the person at the checkout (or the card machine) and is that the visitor is the person named on the card and entitled to use it. This type of conduct is squarely covered by the CPS legal guidance on prosecuting fraud. It also engages dishonesty in that the other elements of the offence are made out on the facts. As noted above, there is no ‘claim of right’ for fraud.
Applying Ivey again, the fact-finding tribunal must first ascertain subjectively the actual state of the individual’s knowledge or belief as to the facts. The man believed he had the right to use his wife’s card. With Ivey, there is no requirement that his belief must be reasonable, the question is whether it is genuinely held. The fact-finding tribunal must consider whether, in the context of what the defendant believed, his actions were dishonest by the standards of ordinary decent people. Presumably, if the man from the country where public transport is free is not dishonest because he believes bus travel is free, the man from the country where husbands have control over their wives finances is not dishonest because he believes he is free to use his wife’s credit card. But: if the jury takes account of the knowledge and belief of this defendant, it is not applying an objective standard.
There also appear to be difficulties in relation to what knowledge or belief can be taken account of in applying Ivey. A banker’s knowledge and belief may be ‘It is accepted practice for bankers to agree to make LIBOR submissions to benefit their trading interests.’ A man from abroad’s knowledge and belief may be ‘It is accepted practice for a husband to spend his wife’s money without her consent.’ Both of these ‘accepted’ practices are dishonest by the standards of ordinary honest people. If Lord Hughes’ illustration in Ivey is right, the jury should take account of the belief of the man from abroad and conclude he is not dishonest. But, on the basis of the Crown Court Compendium, the jury should not take account of the belief of the banker in deciding if he is dishonest. The basis on which a jury may take account of certain knowledge and belief, but not other knowledge and belief, is unclear.
None of this is problematic with Ghosh: under the objective first limb, the defendant’s evidence is relevant to what happened or his conduct, but his own view as to whether his conduct was dishonest by the standard of ordinary honest people is irrelevant. And, with the second limb, the jury take account of the defendant’s knowledge and belief as to facts relevant to his dishonesty – whether right or wrong, reasonable or unreasonable – if they believe him. The question is whether the jury was sure the defendant must have realised that his conduct was dishonest by the standards of reasonable and honest people.
7. Ghosh was not compelled by authority
In Ivey, Lord Hughes engages in a comprehensive review of the pre-Ghosh Theft Act 1968 authorities. The central thread of his argument relates to his understanding of R v Feely: that the Court in that case settled on an objective test of dishonesty, albeit one applied by the jury rather than the judge. Lord Hughes states in relation to Feely:
The decision of the court was that it is for the jury, not the judge, to say whether the conduct established was dishonest or not. … It is therefore inherent in that case that what the jury has to do is to apply its own (objective) standards to whether the conduct was dishonest.34
The basis of Lord Hughes’ understanding of Feely appears to be Lawton LJ’s conclusion:
We do not agree that judges should define what “dishonestly” means. … Jurors, when deciding whether an appropriation is was dishonest can be reasonably expected to, and should apply the current standard of ordinary decent people.35
Clearly, dishonesty is a question of fact for the jury. But there is definitely a recognition in Feely that dishonestly must relate to the defendant’s state of mind. Lawton LJ states:
In section 1(1) of the [Theft] Act of 1968, the word “dishonestly” can only relate to the state of mind of the person who does the act which amounts to appropriation. Whether an accused person has a particular state of mind is a question of fact which has to be decided by the jury when there is a trial on indictment, and by the magistrates when there are summary proceedings.36
It can be seen that Feely gives rise to, but does not resolve, the issue of how the defendant’s actual state of mind is to be resolved by reference to an objective standard. That was, of course, the very issue resolved by Ghosh. If one accepts that Feely does not resolve the question, it effectively breaks the first link in Lord Hughes’ chain of analysis that the weight of authority was in favour of an objective test. in Ghosh, Lord Lane states:
The case [Feely] is often treated as having laid down an objective test of dishonesty for the purpose of section 1 of the Theft Act. But what it actually decided was (i) that it is for the jury to determine whether the defendant acted dishonestly and not for the Judge, (ii) that the word “dishonestly” can only relate to the defendant’s own state of mind, and (iii) that it is unnecessary and undesirable for Judges to define what is meant by “dishonestly”.37
That remains a fair summary of Feely. Lord Lane also states in Ghosh that brought the Court to the heart of the problem:
Is “dishonestly” in section 1 of the Theft Act intended to characterise a course of conduct? Or is it intended to describe a state of mind? If the former, then we can well understand that it could be established independently of the knowledge or belief of the accused. But if, as we think, it is the latter, then the knowledge and belief of the accused are at the root of the problem.38
If Ghosh was not compelled by authority, neither is Ivey. Further, Ghosh identified and solved the conflict in Feely between (a) dishonesty must be an objective standard; (b) dishonesty must also relate to the defendant’s actual state of mind. And did so in a such a way that the test or direction was easily understood by a jury.
8. Ghosh is not necessary to preserve the principle that criminal responsibility should depend on a defendant’s actual state of mind
Lord Hughes’ argument here is that the Court of Appeal in Ghosh was wrong to conclude that the second limb of the Ghosh direction was necessary in order to preserve the principle that, save for strict liability offences, criminal liability requires a defendant to have culpable state of mind.
It is plain that in Ghosh the court did conclude that its compromise second leg test was necessary in order to preserve the principle that criminal responsibility for dishonesty must depend on the actual state of mind of the defendant. It asked the question whether “dishonestly”, where that word appears in the Theft Act, was intended to characterise a course of conduct or to describe a state of mind. Lord Hughes’ view in Ivey is:
… “dishonestly”, where it appears, is indeed intended to characterise what the defendant did, but in characterising it one must first ascertain his actual state of mind as to the facts in which he did it. It was not correct to postulate that the conventional objective test of dishonesty involves judging only the actions and not the state of knowledge or belief as to the facts in which they were performed. What is objectively judged is the standard of behaviour, given any known actual state of mind of the actor as to the facts.39
Lord Hughes’ conclusion is that the principle is preserved because Ivey requires the fact-finding tribunal to ascertain the actual state of the individual’s knowledge or belief as to the facts before applying the objective standards of ordinary decent people. It is not clear this conclusion is justified.
Ivey appears to, and certainly the Crown Court Compendium direction does, preclude the fact-finding tribunal taking account of the belief of the defendant where that belief is a mistaken one as to contemporary standards of honesty. In R v G referred to above, the defendants were children who set fire to a wheelie bin without appreciating the risk that the fire might spread to an adjacent building. It is easy to foresee cases of where defendants do not appreciate – either through immaturity or lack of intelligence – that their conduct would be judged dishonest by ordinary honest people. There is the illustration above of the sacked employee who transfers £50 to her bank account as fair compensation for not being paid the minimum wage. There is no claim of right and her evidence may be ‘I did not realise ordinary honest people would consider what I did as dishonest.’ This is, apparently, no defence under Ivey. It becomes unfair where, say, the defendant is 16 years old or has learning difficulties. And, before settling on a standard based on ordinary reasonable adults for a mental element of any criminal offence, it should be recalled that the age of criminal responsibility in England and Wales is 10 years of age. Baldly put: naive defendants may be held to a standard of insight and understanding which they cannot meet. Ivey has the potential to repeat the problem for fraud and theft that R v G resolved for criminal damage.
Lord Hughes refers to Smith’s Law of Theft and the criticism of the Ghosh test that ‘… the second limb allows the accused to escape liability where he has made a mistake of fact as to the contemporary standards of honesty. But why should that be an excuse?’40 Notwithstanding the principle that criminal liability should require a culpable state of mind, that criticism is valid if one prefers that the criminal law should hold defendants to objective standards. Ivey, however, falls between the stools. The defendant who makes a mistake as to contemporary standards of dishonesty now has no defence. But the defendant who makes a mistake as to some other fact may have a defence even though that mistake of fact may be less excusable given the defendants’ personal characteristics.
Clearly, the Supreme Court revisiting the Ghosh test of dishonesty in Ivey was a matter of choice rather than necessity. It seems unlikely Lord Hughes took responsibility for drafting the judgment without at least some consensus that Ghosh was wrong. Lord Hughes’ conclusion is that ‘These several considerations provide convincing grounds for holding that the second leg of the test propounded in Ghosh does not correctly represent the law’.41 While Lord Hughes refers to the problems collectively, it is suggested that, for the Court, the most compelling will have been ‘the unprincipled divergence between the test for dishonesty in criminal proceedings and the test of the same concept when it arises in the context of a civil action’.42 That was the rock on which Ghosh foundered.
Lord Hughes formulation ‘… Ghosh does not correctly represent the law…’ is unfortunate. It seeks to maintain the fiction that law is being ‘discovered’ rather than made. While there is nothing in the judgement to indicate the Court actually tried to ‘discover’ what Parliament intended by dishonesty in the Fraud Act 2006. Stephen Sedley, former Lord Justice of Appeal, has expressed the view that a great advantage of a barrister-fed judiciary is that the arts of advocacy – including reasoning from a given conclusion – are deployed in the formation of judgements.43 It is submitted that Lord Hughes’ judgment in Ivey is such an example: the arguments against Ghosh are marshalled so as to make the conclusion that it was wrongly decided seem unassailable. It is, however, submitted that it was unfortunate there was no defence of Ghosh in Ivey. It is also submitted that if the practical problems with the Ivey test identified above are correct, there are no easy solutions. It would be unfortunate if, having had a respectable existence for 35 years, ‘dishonesty’ in criminal cases now requires extensive judicial tinkering to achieve justice in difficult cases.
Statutes and secondary legislation
Fraud Act 2006 (link is to legislation.gov.uk)
Gambling Act 2005 (link is to legislation.gov.uk)
Gaming Act 1845 (link is to legislation.gov.uk)
Larceny Act 1916 (link is to legislation.gov.uk)
Theft Act 1968 (link is to legislation.gov.uk)
Fraud Bill HL Bill 2005-06 7 (link is to parliament.uk)
Barlow Clowes v Eurotrust  1 WLR 1476 (link is to bailli.org)
Boggeln v Williams  1 WLR 873
R v Cornelius  EWCA Crim 500 (link is to bailli.org)
R v Feely  QB 530
R v Gilks  1 WLR 1341
R v Governor of Brixton Prison  3 KB 568
R v Greenstein  1 WLR 1353
R v G  UKHL 50 (link is to bailli.org)
R v Ghosh  EWCA Crim 2 (link is to bailli.org)
R v Hayes  EWCA Crim 1944 (link is to bailli.org)
Ivey v Genting Casinos (UK) Ltd t/a Crockfords  UKSC 67 (link is to supremecourt.uk)
R v McIvor  1 WLR 409
R v Rostron  EWCA Crim 2206 (link is to bailli.org)
R v Royle  1 WLR 1764
Scott v Metropolitan Police Commissioner  AC 819 (link is to bailli.org)
Twinsectra v Yardley  2 AC 164 (link is to bailli.org)
R v Waterfall  1 QB 148
R v Williams  1 QB 660
Welham v DPP  AC 103
Lords Hansard, HL Deb, 15 February 1968, vol 289, col 212 ” target=”_blank” rel=”noopener”>http://hansard.millbanksystems.com/lords/1968/feb/15/theft-bill-hl (link is to publications.parliament.uk)
Lords Hansard, HL Deb, 19 July 2005, vol 673, cols 1423 and 1424 ” target=”_blank” rel=”noopener”>https://publications.parliament.uk/pa/ld200506/ldhansrd/vo050719/text/50719-20.htm
(link is to publications.parliament.uk)
Law Commission, Fraud (Law Com No 276, 2002) (link is to law.com.gov.uk)
The hearing of Ivey can be watched here: https://www.supremecourt.uk/cases/uksc-2016-0213.html
Crown Court Compendium, ‘Part I: Jury and Trial Management and Summing Up’ (Judicial College, May 2016) <https://www.judiciary.gov.uk/wp-content/uploads/2016/05/crown-court-compendium-part-i-jury-and-trial-management-and-summing-up.pdf> accessed 28 February 2018 (link is to judiciary.gov.uk)
Crown Court Compendium, ‘Part I: Jury and Trial Management and Summing Up’ (Judicial College, November 2016) <https://www.judiciary.gov.uk/wp-content/uploads/2016/06/crown-court-compendium-pt1-jury-and-trial-management-and-summing-up-nov2017-v3.pdf> accessed 28 February 2018 (link is to judiciary.gov.uk)
Victoria Coren Mitchell, ‘Casinos gamble on their credibility’ The Guardian (London, 29 October 2017) <https://www.theguardian.com/commentisfree/2017/oct/28/casinos-gamble-on-their-credibility> accessed 14 November 2017
-  UKSC 67.
-  QB 1053.
- Judicial College, ‘Crown Court Compendium’ (May 2016) at [8-6] <https://www.judiciary.gov.uk/wp-content/uploads/2016/05/crown-court-compendium-part-i-jury-and-trial-management-and-summing-up.pdf> accessed 21 February 2018.
- Ivey at 
-  UKPC 37 at 
- Ivey at .
- Ivey at .
-  EWHC Admin 2820.
- Paterson (n 8) at .
- Judicial College, ‘Crown Court Compendium’ (November 2017) at [8-6] <https://www.judiciary.gov.uk/publications/crown-court-bench-book-directing-the-jury-2/> accessed 21 February 2018.
- The hearing took place on 13 July 2017 and the hearing can be viewed here: <https://www.supremecourt.uk/cases/uksc-2016-0213.html> accessed 21 February 2018.
- Ivey at .
-  UKHL 50.
-  AC 341.
- Ivey at .
- (1874) LR 9 Ch App 251-52.
- Barlow Clowes (n 5).
- Law Com No 276, 2002 <https://www.lawcom.gov.uk/project/fraud-and-deception> accessed 21 February 2018.
- Explanatory Note to the Fraud Act 2006, para 10 <http://www.legislation.gov.uk/ukpga/2006/35/notes/division/5/2> accessed 21 February 2018.
- HL Deb 19 Jul 2005, vol 673, col 1423.
- HL Deb 19 Jul 2005, vol 673, col 1424.
- Ivey at .
-  EWCA Crim 1944.
-  EWCA Crim 2206.
- R v G (n 14) at .
- Ivey at .
- Criminal Law Revision Committee, Theft and Related Offences (Cmnd 2977, 1966).
- HL Deb 15 February 1968, vol 289, col 212.
-  1 QB 530 at 537E.
- Larceny Act 1916, s 1.
- Crown Court Compendium (n 10) at [8-6].
- Hayes (n 24) at .
- Crown Court Compendium (n 10) at [8-6].
- Ivey at .
- Feely (n 29) at 537H.
- Feely (n 29) at 537F.
- Ghosh (n 2) at 1063B.
- Ghosh (n 2) at 1063F.
- Ivey at .
- Ivey at .
- Ivey at .
- Ivey at .
- S Sedley, Ashes and Sparks: Essays on Law and Justice (CUP 2011) 156.