The term ‘white-collar crime’ was first coined in 1939 by American sociologist Edwin Sutherland who described it as “a violation of criminal law by the person of upper-socioeconomic class in the course of [their] occupational activities”. It continues to be a popular descriptor for financially motivated non-violent crime that in Australia includes fraud offences under the Schedule to the Criminal Code Act 1995 (Cth) (Criminal Code) and the Corporations Act 2001 (Cth).
‘Dishonesty’ is the common element between white-collar fraud offences, but there are two tests for dishonesty applicable to white-collar offences in each of the Criminal Code and Corporations Act: the ‘Ghosh test’ and the ‘Peters test’.
This article reviews the differences between these two tests and identifies which test applies to which offences.
Dishonesty and the Ghosh test
For the purposes of the Criminal Code, s 130.3 states that “dishonest” means:
- dishonest according to the standards of ordinary people; and
- known by the defendant to be dishonest according to the standards of ordinary people.
As Crispin J observed in Anderson v Bigmore (2006) 202 FLR 468 both objective and subjective tests are required to satisfy this section:
The concept of dishonesty in s 130.3 involves not only the objective test in subparagraph (a) but also the subjective test in paragraph (b). The latter test does not relate to any idiosyncratic views of dishonesty held by the alleged offender but depends rather upon whether they knew that the act in question was dishonest according to the standards of ordinary people.
In 2000 the Revised Explanatory Memorandum (EM) to the Criminal Code established that in relation to s 130.3 the test for ‘dishonesty’ is “a straight-forward definition … known as the Ghosh test”. The Ghosh test was conceived in R v Ghosh [1982] QB 1053 in which the UK Court of Appeal sought to reconcile conflicting authorities against the context of “Parliament cannot have intended to catch dishonest conduct … to which no moral obloquy could possibly attach”.
The Ghosh test has been subject to considerable academic and professional (especially perhaps by prosecutors) criticism for reasons including difficulties with proving dishonesty. On the other hand, the theory of the criminal law means that it should only be used to censure substantial wrongdoing. Also, maximum sentences should be proportionate to the seriousness of the wrongdoing.
What is the Peters test?
So then we come to the Peters test. In Peters v R (1998) 192 CLR 493 Toohey and Gaudron JJ stated:
In a case in which it is necessary for a jury to decide whether an act is dishonest, the proper course is for the trial judge to identify the knowledge, belief or intent which is said to render that act dishonest and to instruct the jury to decide whether the accused had that knowledge, belief or intent and, if so, to determine whether, on that account, the act was dishonest.
Unfortunately, the EM misunderstands the Peters test especially as it relates to the issue of subjectivity. As Stephen Odgers puts it in Principles of Federal Criminal Law ((4th ed) Thomson Reuters, citing Williams CR, ‘The Shifting Meaning of Dishonesty’ (1999) 23 Crim LJ 275):
This test does not mean that dishonesty under Peters could be established without a “guilty mind”. Rather, it held that proof of the knowledge, belief or intention alleged by the prosecution to have existed would be sufficient (given that such a state of mind would usually render the conduct dishonest), without any need to also prove knowledge that the conduct was dishonest by the standards of ordinary people.
In other words, the Peters test requires the trial judge to:
- identify the knowledge, belief, or intent which is said to render the relevant conduct dishonest;
- instruct the jury to decide whether the accused had that knowledge, belief or intent and, if so, to determine whether, on that account, the act was dishonest; and
- direct the jury that, in determining whether the conduct of the accused was dishonest, the standard is that of ordinary, decent people (Lusty D, ‘The meaning of dishonesty in Australia the Rejection and resurrection of the discredited Ghosh test’ (2012) 36 Criminal Law Journal 282).
How is dishonesty identified and/or proven?
Unlike their state and territory counterparts, federal white-collar fraud prosecutions must be tried by a judge and jury: s 80 Constitution. In Peters v R, McHugh J said dishonest means may be proven by evidence that the defendants intended to prejudice another person’s rights or interest or performance of public duty by:
- making or taking advantage of representations or promises which they knew were false or would not be carried out;
- concealing facts which they had a duty to disclose; or
- engaging in conduct which they had no right to engage in.
Which test applies when?
As was shown above, Chapter 2 of the Criminal Code codified the Ghosh test rather than the High Court’s reformulation in R v Peters. At common law however (that is, common law of criminal responsibility), the Peters test continues to apply. Accordingly, in relation to fraud offences in the Criminal Code (eg, general dishonesty, s 135.1(1)) as well as in statutes which expressly import Chapter 2 of the Criminal Code, it is the Ghosh test that is relevant. Conversely, in relation to offences where the Criminal Code has no application, R v Peters will continue to apply. Therefore, in relation to white-collar fraud offences under the Corporations Act (eg, Good faith, use of position and use of information, s 184(1)(b)) the Peters test applies: SAJ v The Queen (2012) 269 FLR 390.
Take away
White-collar criminal offences in Australia may share the common element of dishonesty. This common element does not equate to a common test and offences under the Criminal Code and Corporations Act are subject to the distinct Ghosh and Peters tests respectively. Contrary to some interpretations, both tests have subjective components. The Ghosh test may link legal responsibility more closely to moral culpability although be practically more difficult to satisfy/prove than the Peters test.
