R v Petroulias - [2005] NSWCCA 75

NEW SOUTH WALES COURT OF CRIMINAL APPEAL

CITATION: REGINA v Nikytas Nicholas PETROULIAS [2005] NSWCCA 75

FILE NUMBER(S):
2005/266

HEARING DATE(S): 1 March 2005

JUDGMENT DATE: 11/03/2005

PARTIES:
REGINA
Nikytas Nicholas PETROULIAS

JUDGMENT OF: Spigelman CJ Mason P Hunt AJA

LOWER COURT JURISDICTION: Supreme Court

LOWER COURT FILE NUMBER(S): 70076/02

LOWER COURT JUDICIAL OFFICER: Sully J

COUNSEL:
Appellant: P Hastings QC/ C Hoy
Respondent: I Temby QC/ N Clelland SC/
C Livermore

SOLICITORS:
Appellant: Commonwealth Director of Public Prosecutions
Respondent: Coadys Barristers and Solicitors
c/- Watsons Connaught Offices

CATCHWORDS:
Crimes Act 1914 (Cth) s29D - charge of defrauding the Commonwealth - taxation fraud - risk to revenue - Pt IVAA Taxation Administration Act 1953 (Cth) - Private Rulings - loss of ability to litigate - s170BB Income Tax Assessment Act 1936 (Cth) - mixed question of fact and law - directions to jury - arguability in law - permanent stay of criminal prosecution - fair trial (D)

LEGISLATION CITED:
Evidence Act 1995 s80
Crimes Act 1914 (Cth) s29D
Income Tax Assessment Act 1936 s170BB

DECISION:
Appeal upheld. Stay set aside.

JUDGMENT:

IN THE COURT OF
CRIMINAL APPEAL

CCA 2005/266
SPIGELMAN CJ
MASON P
HUNT AJA
Friday 11 March 2005

REGINA v Nikytas Nicholas PETROULIAS

Judgment

SPIGELMAN CJ & HUNT A-JA : We agree with Mason P, for the reasons he has given, that the decision of Sully J to grant a permanent stay in this case should be set aside. However, the view we have formed as to how the trial should be conducted in relation to the charge of defrauding the Commonwealth differs sufficiently from that proposed by the President that it is necessary to state our own reasons as to the nature of the issue posed and how it should be put to the jury.
Defrauding the Commonwealth
The first count in the indictment is currently in these terms:
Between 1 September 1997 and 27 February 1999, Nikytas Nicholas Petroulias did defraud the Commonwealth in that [,] while an officer of the Australian Taxation Office, he did put the revenue of the Commonwealth at risk by causing Private Binding Rulings and Advance Opinions to issue to taxpayers, by dishonest means.
The three categories of fraud identified by Toohey and Gaudron JJ in Peters v TheQueen (1998) 192 CLR 493 at [30], are the intentional creation of a situation in which the accused:
(1)deprives another person of money or property, or
(2)puts the money or property of that other person at risk, or
(3)prejudicially affects that person in relation to some lawful right, interest, opportunity or advantage,
knowing that he has no right to deprive that person of that money or property or to prejudice his or her rights. (See also McHugh J to the same effect, at [73].)
In previous proceedings in the Court of Appeal relating to the case against this accused — Wills v Petroulias (2003) 58 NSWLR 598 — these three categories were described, respectively, as an actual loss of revenue, an actual risk of loss of revenue and the lost opportunity of testing the views expressed in the private rulings, the Commissioner being bound in other cases (at least administratively) by the rulings made. The distinction between the second and third categories is that, in the second category, there must be the possibility that the Commissioner, were he not bound by the rulings made, would win a case by demonstrating that the rulings were wrong (Wills v Petroulias at [64]), whereas in the third category there is merely a loss of the opportunity to litigate whether the rulings were correct.
The Crown asserts that its first count was intended to reflect the second of those three categories. The relevant particulars of harm nominated by the Crown as encapsulating its case are particulars N(iv) and N(v):
(iv) In each case, the revenue of the Commonwealth was at risk because of the prejudice to the right or advantage of the Commissioner of Taxation to otherwise act without restriction to assess and recover any tax payable.
(v) The PBRs [private binding rulings] and AOs [advance opinions] issued at the direction of Petroulias were arguably wrong.
Particular N(iv), standing by itself, is no more than a third category (lost opportunity) case, the least serious of the three categories. However, particular N(v) is said by the Crown to be itself a particular of particular N(iv). It is only by reading the two together that the case can be elevated to the second category (an actual risk of loss of revenue case). In order to support the count, therefore, the Crown must establish that the private rulings were, as tentatively suggested in Wills v Petroulias, arguably wrong. The Crown cannot be permitted to put these two particulars in the alternative.
Without the particulars, the first count in the indictment simply states that the accused “did put the revenue … at risk”. Such a charge would be ambiguous, and thus impermissible, in that it does not specify with sufficient particularity the actus reus of the offence. It would, for example, permit the Crown to pursue either a lost opportunity case or an actual risk of loss case. However, particulars N(iv) and N(v) cure any such problem of ambiguity so long as they are read together. The Crown has expressly abjured any reliance on a lost opportunity case. Sully J was correct to determine that the Crown would be held to pursue only the case which it has now particularised.
There was some discussion in Wills v Petroulias as to what had to be established in a case alleging that the private rulings procured by the accused constituted an actual risk that the revenue of the Commonwealth was imperilled, but only tentative views were expressed. We agree with Mason P that the time has come for this Court to express a concluded view. If this Court adopts the need to prove that the private rulings were arguably wrong (or any variation of that formulation), a second issue must then be determined — whether that issue is one of law for the judge to decide or whether it is one of fact (or mixed law and fact) for the jury to determine.

The relevant issue

The three categories of fraud already adopted descend in order of seriousness — and consequently in the nature of the punishment to be imposed — from the actual loss (when it would be necessary in the present case to establish that the private rulings were wrong) to the loss of an opportunity to test the views expressed in those rulings (when the Crown would need to establish no more than the existence of a genuine dispute by the Commissioner concerning those rulings). What must be established for the second category must therefore be somewhere between those two categories.
In Wills v Petroulias, the Chief Justice expressed support for, but did not determine, the proposition that it would be necessary to establish that the rulings were arguably wrong, and that there must be the possibility that the Commissioner would win a case in which he was allowed to dispute those rulings: Wills v Petroulias (No 1) at [55], [59], [62] and [64] and Wills v Petroulias (No 2) at [67]. The views expressed in the earlier judgments were, however, concerned with the application of a particular statutory regime — particularly with ss 104 and 110 of the Justices Act 1902. The issues which arise in relation to the trial of the charges on which the accused has now been committed are quite distinct. The phrase “arguably wrong” was appropriate for the arguments which arose in those earlier proceedings, but it is not helpful in understanding what is really involved in the issue which arises.
In our opinion, the issue which arises in a second category (actual risk of loss of revenue) case would be expressed more clearly if it were made clear on its face just why the issue arises and if it were couched in a positive rather than a possibly negative way. In this way, the difficulties which Sully J saw in the “arguably wrong” formulation would largely disappear. In our view, to say that the private rulings are arguably wrong is really saying no more than that the Commissioner of Taxation has an arguable case to put that the rulings were wrong and, accordingly, that the risk to the revenue was such that the Commonwealth was in fact deprived of something of value. There is no difference in substance in stating the issue in this way. But stating the issue in this way leads immediately to a consideration of the approach to be taken in determining the issue, be it by the judge or the jury.
One issue of law for the judge
The arguable case to be put by the Commissioner must first demonstrate the possibility that the Commissioner would win a case in which he was allowed to dispute the private rulings. Is the case which the Commissioner wishes to put one which is capable of success? That is clearly an issue of law for the judge. The ruling would be given in the way that every such ruling is given as to whether the Crown case can go to the jury — before counsel address the jury, and after argument in the absence of the jury. In the event that the judge rules that the arguable case the Commissioner wishes to put forward is capable of success, no further mention of the ruling should be made to the jury, as jurors may well be misled as to its effect upon the issues which they have to determine.
The approach to be taken by the relevant tribunal
An arguable case is (by definition) a case which is supported by an argument which is plausible or apparently reasonable on its face. A plausible or apparently reasonable argument is one which can be reasoned logically from one step to the next step to its conclusion. It is not the function of the relevant tribunal to determine whether either the Commissioner’s argument or the taxpayer’s argument is correct, or even which is the more plausible or the more apparently reasonable argument. It may be that each of those parties has an arguable case. That is not unusual in litigation. Even though the taxpayer may have an arguable case, it does not mean that the relevant tribunal may not be satisfied beyond reasonable doubt that the Commissioner also has an arguable case. If so satisfied that the Commissioner has an arguable case, the relevant tribunal has resolved the issue posed by a second category fraud case, subject to the mens rea component identified in par 3, supra.
Which is the relevant tribunal for this issue?
Whatever conclusion is reached on this question, the primary issue which must ultimately go to the jury is whether the Commonwealth has suffered a detriment as a result of the private rulings given. That detriment is to be established in this case by demonstrating, according to our suggested formulation, that the Commissioner has an arguable case that the private rulings were wrong. The parties (for quite different purposes) agreed before Sully J and this Court that the predecessor of this issue (whether the private rulings were arguably wrong) was an issue of law for the judge. We have already accepted that the capacity of the Commissioner’s arguable case to succeed is an issue of law for the judge (par 12, supra). But we do not agree that the remaining issue is a question of law to be determined by the judge. In our opinion, that issue, despite its obvious legal character, is a fact in issue in the trial to be determined by the jury.
A trial judge must direct the jury as to, inter alia, the manner in which the jurors are to carry out their functions and the elements of the crime charged. The contents of such directions are accepted as being matters of law for the trial judge to decide. But the issue of whether or not the Commissioner has an arguable case that the private rulings given were wrong is not an element of the crime of defrauding the Commonwealth. It is what the Crown puts forward as establishing that the actual risk that the revenue of the Commonwealth was imperilled amounts to the deprivation necessary for a charge of defrauding the Commonwealth. The relevant element of the crime is deprivation. Whether the Commissioner has an arguable case that the private rulings were wrong is, therefore, a fact in issue relevant to the deprivation element of the crime charged.
The directions to be given on arguable case
The requirement that the trial judge direct the jury on the applicable law in relation to the elements of the crime charged does not require the judge to direct the jury as to what the law is in relation to Fringe Benefits Tax, the meaning of the term “associate” in s 26AAB(14)(a)(iv) of the Income Tax Assessment Act 1936, or whether the employer’s payments under the relevant scheme were deductible. Nor is it appropriate that the judge direct the jury that the arguments put forward in the Crown case in this regard are “arguable”.
The directions which the judge will be required to give in this case will be such as are required for the jury to determine for themselves whether the Crown has established an arguable case on either of the two issues that have been raised — fringe benefits tax and deductibility — so that the jury are satisfied beyond reasonable doubt that the risk to the revenue was such that the Commonwealth was in fact deprived of something of value. Subject to the course of the trial, the judge will probably be required to outline the relevant provisions of the statute and identify in the material admitted into evidence the arguments in favour of the liability to fringe benefits tax and those against deductibility, together with the evidence that supports or detracts from the strength of those arguments. The assessment of the evidence — whether in terms of “arguable case” or “arguably wrong” — is a matter for the jury.
The issue for the jury to determine is whether or not the arguments propounded on either of the two issues identified in the previous paragraph are of sufficient strength to justify the conclusion beyond reasonable doubt that the Commonwealth was in fact deprived of something of value by reason of the issue of the private rulings. It should be noted that the Commonwealth’s case in this regard could be made out if the members of the jury determined that the Crown case was of sufficient strength on either the fringe benefits tax issue or the deductibility issue. It would not be necessary for jurors to be unanimous as to which of the two were of such strength.
The approach to be taken, already identified in par 13, supra, will constitute the basis for the directions to be given to the jury. It would not be appropriate for the trial judge to direct the jury that, depending on the view they have formed of certain factual matters, the Australian Tax Office did lose something of value — or even that the rulings did put the revenue of the Commonwealth at risk. Such a direction may well be subject to various permutations and combinations of the factual matters, and the direction could lead to confusion. The more usual form of directions to a jury — and the infinitely safer one — is to define the issue which the jury must determine, then explain to them the arguments of both parties in relation to the relevant factual material, and finally conclude that, if the Crown has satisfied them that it has established that particular issue, they may move on to the next issue.
In order to produce directions in terms which can easily be assimilated by the jury, their contents should in our view be debated by the parties and tentatively determined by the judge at or before the trial, and reduced to writing for the jurors to use during the summing-up and in their deliberations. Those tentative directions can always of course be revisited before the summing-up in the light of what happens during the trial. Additional assistance by way of elaboration or comment may also be given by the judge during the course of giving the directions which are in writing.
The difficulties to which the Crown referred
The summing-up in this case will undoubtedly be a complex task, but it is not an impossible one. It will be by no means as difficult as the directions which must be given in a murder trial where there is a serious dispute between pathologists as to whether the injury causing the death of the deceased was consistent with murder or accident, or in a fraud case based on complicated accounting methods, or a prosecution under the Corporations Act 2001 (Cwth) based on alleged breaches of accounting standards, or (when civil juries were available in such cases) in a case based on medical negligence or a case based on legal negligence (where the negligence is alleged to have been a misinterpretation of the law).
The difficulty in directing the jury in this case was put forward by the Crown as a reason for the issue to be decided by the judge rather than the jury. The Crown’s argument was based on the decision of this Court in Regina v Cassell (1998) 45 NSWLR 325 at 334, where it was held that the issue of materiality, in a charge of giving evidence before the Independent Commission Against Corruption knowing that it was false or misleading in a material particular, was for the judge to decide as a question of law, thereby putting that charge on the same basis as perjury. Section 327 of the Crimes Act 1900 expressly provides that materiality in a perjury case is a matter of law, and thus for the judge to determine. Smart J, with whom the other judges agreed, pointed out that what would be material to the corruption investigation being carried out by ICAC could not be understood without a knowledge of ICAC’s statute, the objects and provisions of that statute and an appreciation of the particular investigation being carried out when the evidence is given, and that for that reason it was a question of law for the judge. Such difficulties are inherent not only in every case of perjury but also in every case of knowingly giving false evidence before tribunals other than courts. There was every reason for assigning the issue to the judge rather than the jury.
The High Court took a similar approach in Ahern v The Queen (1988) 165 CLR 87, where it was held in a joint judgment that it was for the judge, and not the jury, to decide whether there was a prima facie case of conspiracy made out by the Crown before statements made by co-conspirators not otherwise admissible against the accused became so admissible. It is true that the Court referred (at 104) to the need for directions which would be of unacceptable complexity if the issue were to be determined by the jury, but that was because of the nature of the issue itself, not because of the facts of that particular case. The true basis for the Court’s decision is stated (at 100) to have been that the existence of a prima facie case was only another way of saying that there was a reasonable basis in the evidence for admitting the evidence of such statements against the accused, a standard by which the admissibility of many different types of evidence may be admitted, and a decision which is always made by the judge.
Those two cases illustrate that the assignment of the relevant issue to the judge rather than the jury is to apply in every case in which that issue arises, and not just in those individual cases where the task of directing the jury as to its task in relation to that issue is a difficult one. We do not accept that the question as whether there is an arguable case in relation to a basic issue arising under the Income Tax Assessment Act could be a question of law for the judge in one case (because of the complexity of that particular case) but a question of fact for the jury in another case (because that complexity did not exist). We were informed by the Crown that, in Ahern (a “bottom of the harbour” tax scheme case of defrauding the Commonwealth), all of the issues under that statute relating to the tax scheme were left to the jury, but the rejection of the Crown’s argument in the present case does not depend upon the accuracy of that information.

The accused (the respondent to the appeal) did not present any further specific arguments as to why the issue of whether the private rulings were arguably wrong was for the judge rather than the jury. Rather, the accused’s argument was that, because it was for the judge and not the jury to decide, the law would be brought into public disrepute if it permitted the judge to decide such an important issue in the case, one which may effectively decide the whole case. We agree with Mason P that this argument should be rejected as untenable.
It is for these reasons that we are of the opinion that whether an arguable case exists that the private rulings were wrong is a fact in issue in the trial to be determined by the jury.
Evidence admissible in relation to this issue
Related to this issue as to whether the decision is one for the judge or the jury is the admissibility of evidence upon which the decision is to be based. In our opinion, but subject to the Crown being permitted to enlarge its particulars in the light of the decision now given by the Court of Criminal Appeal, there is no bar to the admission of relevant statements by courts or others as to the interpretation of these provisions merely because they may have come into existence after the time when the private rulings were made. What the law was at that time is an objective matter which the Crown must prove. That objective fact has nothing to do with any state of mind on the part of the accused which may be relevant to his mens rea. There is therefore no need to restrict the admissible material to material which was in existence at that relevant time. The issue as to what the law was at the relevant time may be proved by reference to what has since been held to be the law, provided that there has been no change in the relevant statutory provisions in the meantime. It is presumed to have been the law all along.
Although the Crown, on its particulars, does not propose to rely on expert evidence concerning this fact in issue, such would be admissible. The ultimate issue rule has been abolished by s 80 of the Evidence Act 1995, and the jury may be greatly assisted by such evidence. There is one caveat to the use to be made of opinion evidence on this issue. That evidence should, so far as is possible, be expressed as being that the Commissioner’s case on deductibility and/or fringe benefits tax was or was not correct, rather than that it was or was not arguable.
Mason P has outlined the evidence relied upon by the Crown to establish the proposition that the “arguably wrong” test could be satisfied. It is unnecessary to repeat the material here. It is sufficient to say that, for the purposes of an application for a permanent stay, the material referred to is sufficient to give rise to a real issue as to whether or not the risk to the revenue was such that the Commonwealth was in fact deprived of something of value. We do not, however, intend by these remarks to suggest that the trial judge — particularly in the light of such evidence to the contrary on which the accused may wish to rely in this regard — would not be entitled to form his own judgment on the issue of law we have identified in pars 12 and 17, supra, namely whether or not the case to be put by the Commissioner is sufficiently arguable to demonstrate the possibility that the Commissioner would win a case in which he was allowed to dispute the private ruling.
MASON P: The Director of Public Prosecutions for the Commonwealth appeals against the order of Sully J made on 21 February 2005 permanently staying proceedings on a particular count in a proposed indictment. The charge is brought pursuant to s 29D of the Crimes Act 1914(Cth) which provides:
A person who defrauds the Commonwealth or a public authority under the Commonwealth is guilty of an indictable offence.
The broad factual basis of the Crown case was described by the Chief Justice in earlier proceedings in the following terms (Wills v Petroulias (2003) 58 NSWLR 598 at 600):
3     Between 1997 and 1999 the Opponent held office in the Australian Taxation Office (“ATO”). In September 1998 he became First Assistant Commissioner. The Crown alleges that at the time of such employment the Opponent was also involved in the planning, promotion and implementation of tax schemes with certain other persons and that he used his position as an officer of the ATO to facilitate favourable Advance Opinions and Private Binding Rulings issued by the ATO.
4.     Private Rulings are issued under Pt IVAA of the Taxation Administration Act 1953 (Cth). The scheme of Private Binding Rulings is as follows:
•     A person may apply to the Commissioner for a ruling as to how a tax law would apply to the person in relation to an arrangement (s14ZAF).
•     The Commissioner is obliged to comply with an application for such a ruling, subject to certain exemptions (s14ZAL(1) and s14ZAN).
•     There are circumstances, not presently material, in which a Commissioner may decline to make a ruling (s14ZAQ).
•     A ruling given in this manner is binding by force of statute (s170BB of the Income Tax Assessment Act 1936 (Cth).
5     The Opponent and his partners established a business known as the Productivity Incentive Corporation (“PIC”) which, inter alia, applied for Private Binding Rulings with respect to tax schemes described as Employee Benefit Arrangements (“EBAs”). In general terms the EBAs involved the payment by an employer of monies into a trust fund. The trust would lend money to individual employees for the purpose of purchasing units in the trust. The income of the trust would enure to the benefit of nominated employees. The eligibility of the employees to receive returns on the units would depend on continued service with the employer and the attainment of productivity or performance standards. The corpus would eventually vest in favour of the employee. The purpose of these arrangements was said to be to encourage employees to remain with their employer and also to act as an incentive for such employees to become more productive.
6     As is usually the case, the taxation treatment of the cash flows was of critical significance to the commercial viability of the arrangements. It was always accepted that each employee would be assessable on his or her share of income of the trust. Four issues arose:
•     Would the employer’s contributions be deductible at the time they were made?
•     Would the employer be liable for fringe benefit tax under the Fringe Benefit Tax Assessment Act 1986 (Cth) with respect to its contributions?
•     Would the general anti-avoidance provisions found in Pt IVA of the Income Tax Assessment Act apply to the scheme?
•     Would the employee be assessable on the value of the units in the trust that each acquired?
7     The Crown case was that the Opponent used his position to ensure that officers under his direction issued favourable rulings with respect to applications made by the joint venture in which he had a financial interest. This was in contrast with the Opponent’s intervention with respect to applications by other promoters of similar schemes, in the case of which he allegedly intervened to ensure that favourable rulings were not given. Indeed, the Crown contended that the Opponent was involved in the preparation and publication of a Public Ruling in which the ATO indicated a view opposed to the efficacy of the EBAs.
The favourable rulings issued at the respondent’s instigation assisted the taxpayers who were “Rulees” (an ATO expression) on two matters of continuing relevance, the deductibility of the employer’s contributions, and the absence of liability to fringe benefits tax (FBT).
A brief chronology of earlier proceedings provides the background to the present appeal:

24 March 2000	Respondent first charged. The original charge of conspiracy to defraud the Commonwealth was subsequently amended to include a s29D charge
November 2001-June 2002	36-day committal hearing
17 July 2002	Respondent committed for trial on three charges, including a s29D charge.
16 December 2002	Simpson J granted leave to appeal against the committal order on the basis of errors of law (see *Petroulias v Wills* [2002] NSWSC 1190, 51 ATR 564, 174 FLR 258)
3 March 2003	Simpson J allowed appeal and quashed the order for committal on s29D charge (see *Petroulias v Wills* [2003] NSWSC 106, 138 A Crim R 182)
3 October 2003	Court of Appeal set aside Simpson J's orders and dismissed appeal to Supreme Court (see *Wills v Petroulias* [2003] NSWCA 286, 58 NSWLR 598 ("*Petroulias (No 1)*"))
22 December 2003	Court of Appeal varied its orders, remitting the case to a magistrate to make the order of committal authorised by law and to amend the order [for committal] accordingly (see *Wills v Petroulias* [2003] NSWCA 390. 58 NSWLR 598 (“*Petroulias (No 2)*”) at 618)
24 March 2004	Order for committal made by Ms D Sweeney LCM
16 August 2004	Sperling J dismissed appeal against committal order (see *Petroulias v Wills* [2004) NSWSC 739)
7 February 2005	Date originally fixed for trial before Sully J

In Petroulias (No 1) and (No 2) the reasons of the Court of Appeal were given by Spigelman CJ. Handley JA and Santow JA agreed with the Chief Justice.
The changing forms of the s29D count
The s29D count has undergone several transformations, both as to its form and the case sought to be proved under it.
The history down to Petroulias (No 1) in the Court of Appeal was outlined by the Chief Justice (emphasis added):
12 The charge before the magistrate, which remained the basis of the proceedings at the time that the leave application was heard by Simpson J, was in the following terms:
“That between about 1 September 1997 and 17 February 1999 Nikytas Nicholas Petroulias, also known as Nick Petroulias, did defraud the Commonwealth, namely the Australian Taxation Office contrary to s29D of the Crimes Act 1914 in that, while an officer of the Australian Taxation Office he did, by dishonest means, assist taxpayers to avoid the payment of taxation.”
13 The two particulars originally given of this charge were as follows:
“While an officer of the Australian Taxation Office involved in the regulation of tax avoidance arrangements, in breach of his duty he
(a) was involved in devising, promoting and implementing tax avoidance arrangements known as employee benefit arrangements; and
(b) facilitated the issue by the Australian Taxation Office of favourable Advance Opinions and Private Binding Rulings to taxpayers entering into such arrangements.”
14 Further particulars of dishonesty were sought and supplied. No particulars of the element of deprivation were sought.
15 An issue arose in the course of the hearing before Simpson J as to whether or not the charge, as originally formulated, required the Crown to establish an actual loss of revenue. During the course of the hearing those appearing for the Crown, whilst rejecting the proposition that the charge was capable of such an interpretation, nevertheless put forward an alternative charge formulated in the following terms:
“Between about 1 September 1997 and 27 February 1999 Nikytas Nicholas Petroulias, did defraud the Commonwealth, namely the Australian Taxation Office, in that, while an officer of the Australian Taxation Office, he did, by dishonest means, assist taxpayers to endeavour to avoid the payment of tax.”
16 In the submissions to this Court, the Crown put its case in another formulation. It alleged that the Opponent had “put the interests of the Commonwealth at risk”. It may be that the concept of ‘risk’ or ‘imperilment’ in this formulation was what was sought to be conveyed by the word “endeavour” in the reformulated charge.
17 During the course of the hearing before Simpson J the Crown case was expressed in a third way, although this version was never reduced to the terms of a formal charge. Her Honour recorded the Crown position at a number of points in her first judgment of 16 December 2002 in terms of the loss of the “opportunity to litigate”.
18 For example, after referring to submissions that the element of defrauding could arise where the Commonwealth’s revenue was “put at risk, prejudicially affected, or ‘imperilled’”, her Honour said at [38]:
“The consequence … is that the Commissioner was deprived of the opportunity to test, by adjudication in a court, the correctness of the views expressed in the rulings. Once the ruling was given, the employer and employees were safe from the imposition of taxation of the kinds the subject of the rulings. This was so whether or not curial adjudication of the issues involved would have resulted in a favourable decision for the taxpayer. What the Crown proposes to prove was ‘imperilled’ by the applicant’s dishonest conduct was its opportunity to litigate the effectiveness of the schemes or plans.”
The Chief Justice added the following comment (id at [19]):
19 I note that the Crown did not at any time formulate the element of deprivation in terms of “deceiving public officers into committing a breach of duty”. This has considerable support in the authorities. (See e.g. R v Bassey (1931) 22 Cr App R 160 at 162; Board of Trade v Owen [1957] AC 602 at 622; R v Withers [1975] AC 842 at 877 (see below [1974] 1 QB 414 at 420); Wai Yu-tsang [1992] 1 AC 269 at 277; Peters v The Queen (1998) 192 CLR 493 at [74]). This effect was once expressed in terms of “deceiving a person responsible for a public duty into doing something that he would not have done but for the deceit” (R v Terry [1984] 1 AC 374 at 379). The Crown case in these proceedings did not rely on any such loss or disadvantage.
Later in his reasons in Petroulias (No 1) the Chief Justice said:
50 As indicated above, Simpson J formed the view that the charge as originally formulated, i.e. “… did, by dishonest means, assist taxpayers to avoid the payment of taxation” was such as to allege that the Commonwealth did, as a matter of fact, lose revenue. The Crown submission to her Honour, repeated in this Court, was that the use of the word “to” in the phrase “to avoid the payment of tax” did not carry any implication that tax was in fact avoided, but simply put tax payers, to repeat the submission exactly, “in a position to endeavour to minimise their taxation liabilities.” It was in order to make this proposition clear that the Crown introduced the words “to endeavour to” in the alternative formulation of the charge.
51 It was, of course, always open to the Commonwealth to particularise an actual loss of revenue as the relevant deprivation for the purposes of the offence of “defrauding”. In my opinion, the specific and unusual use of the word “to” in the sense of “to endeavour to”, was not the natural meaning of the original charge. At the very least, it is so ambiguous as to be inappropriate for the formulation of an indictable offence. In my view, her Honour was correct to conclude that the charge as originally formulated presented a Crown case to the effect that revenue had in fact been lost. There was no evidence of any such loss.
Alternative Formulations
52 The alternative charge contains difficulties of its own in the use of the word “endeavour” and the word “avoid”. Without commenting on the terminology, the Crown invokes a line of authority to the effect that it was sufficient for the purposes of the element of deprivation in a defrauding charge to establish a “risk” or some form of “imperilment” or of “prejudice” to an economic interest of the Commonwealth. It was this approach which, her Honour concluded, required the Crown to adduce some evidence that the Private Rulings were at least arguably wrong. She concluded that it had not done so.
53 The third way the Crown put its case was variously described in terms of a “right to litigate” or as the deprivation of “the opportunity to test the correctness of the views by adjudication”. Her Honour rejected this approach to the Crown case on the basis that any such ‘right’ or ‘opportunity’ must be shown to have been of some value in the same way as her Honour rejected the previous formulation, i.e. there had to be evidence of an arguable case that the Rulings were wrong.
For reasons contained in Petroulias (No 2), the Court of Appeal remitted the matter for further committal. The Court described the arguability of the FBT issue and the arguability of deductibility as matters yet to be resolved (see 626[49] and 629[67]. See also Petroulias (No 1) at 612[59], [62]).
At that stage further versions of the count were discussed, but the Court decided that it should not settle the form of the indictment (see Petroulias (No 2) at [69]).
In early 2004 the Crown Prosecutor reformulated the charge once more. It became (emphasis added):
18 Between about 1 September 1997 and 27 February 1999 Nikytas Nicholas Petroulias did defraud the Commonwealth namely the Australian Taxation Office, in that while an officer of the Australian Taxation Office he did by dishonest means cause Private Binding Rulings and Advance Opinions to issue to taxpayers whereby the Commissioner of Taxation was prejudiced in relation to his rights, interests, opportunity or advantages to determine the taxation consequences of the arrangements referred to in the Rulings and Opinions.
The respondent was committed for trial on such a charge on 24 March 2004. As indicated, his appeal against that order was dismissed by Sperling J on 16 August 2004.
The proposed indictment tendered by the Prosecutor to Sully J earlier this month contains a different formulation, as follows (emphasis added):
Between about 1 September 1997 and 27 February 1999, Nikytas Nicholas Petroulias did defraud the Commonwealth, in that while an officer of the Australian Taxation Office, he did put the revenue of the Commonwealth at risk by causing Private Binding Rulings and Advance Opinions to issue to taxpayers, by dishonest means.
This is the portion of the criminal proceedings that has been permanently stayed by the order under appeal.
Different species of deprivation
The focus of the proceedings in the Supreme Court, the Court of Appeal and now in this Court has at all times been the element of “loss” or “disadvantage” in the offence of “defrauding the Commonwealth”. As Spigelman CJ observed in Petroulias (No 1) at [8]:
It is not enough to establish a fraud. Someone must have been defrauded.
The “someone” has always been the Commonwealth and/or the Australian ATO. But of what has the Commonwealth or ATO been defrauded? The several manifestations of the s29D count have endeavoured to answer this question, thereby hopefully setting the scene for the factual contest at the trial.
In Petroulias (No 1), the Chief Justice pointed to the availability of different species of loss/imperilment/prejudice. One species or variant was the proposition that a charge of defrauding can be made out by putting the property or rights of a person at risk or imperilling them or prejudicing them in some way (see eg R v Kastratovic (1985) 42 SASR 59 at 62, Peters v The Queen (1998) 192 CLR 493 at 508[30], 525[73]-[74], 529[84], 552[138], MacLeod v The Queen (2003) 214 CLR 230 at 241[35], Petroulias (No 1) at 611[54]). This is the species invoked in the final version of the count, with “the revenue of the Commonwealth” being identified as the particular interest imperilled.
However, the charge on which the respondent was committed for trial on 24 March 2004 described the Commissioner of Taxation’s imperilment as prejudice in relation to his rights, interests, opportunity or advantages to determine the taxation consequences of the arrangements referred to in the Rulings or Opinions issued by the respondent. This different species of prejudice was discussed in Petroulias (No 1) at [63]-[79] under the heading Opportunity lost. The Chief Justice summarised that concept as follows:
63 The Crown also relied on an element of deprivation expressed in terms of a loss of an opportunity to test the views expressed in the rulings by adjudication or of an opportunity to litigate. The issue that arises in this respect is whether or not the element of detriment in a charge of defrauding can be made out on the basis of a loss of opportunity to determine a matter, which is an open question and about which a bona fide belief favourable to a person is held by that person. If that is sufficient, then it is not, in my opinion, necessary to establish in a positive way that the Commonwealth had an arguable case.
64 It was the loss of the ability to argue the case, by the mandatory operation of s170BB of the Income Tax Assessment Act 1936 (Cth) which gave the Private Rulings binding force, as distinct from the possibility of winning the case that was, on this approach, the relevant deprivation. It is only if reliance is place on the possibility of winning a case, that establishing an arguable case could be an essential part of the Crown case in proceedings under s29D.

As indicated in the final sentence of the two paragraphs just quoted, this lost opportunity to litigate approach did not require proof that the rulings were arguably wrong. It would suffice if they put to rest a genuine dispute (see Petroulias (No 1) esp at [71]-[72]). Adverting to the material placed before the magistrate at the original committal, the Chief Justice said:
56   The Crown pointed to the evidence, presented to her Honour and this Court in summary form, which indicated that strong views had been expressed within the ATO that the regime would apply; that the Opponent himself had actively promoted that opinion with respect to the consideration within the ATO of Private Rulings sought for EBA schemes, other than those in which the Opponent had a financial interest; that the Opponent actively promoted a Public Ruling to the effect that the schemes did not work in terms of their fringe benefits tax implications and that a Public Ruling to that effect was in fact issued.
…
78   In my opinion, there was evidence before the magistrate which established that there was a “genuine dispute”, to use the language of King CJ in Kastratovic, or that the Commissioner was “genuinely asserting” a position, to use the language of Toohey and Gaudron JJ in Peters. The element of “genuineness” was established on the basis of the contrary opinions expressed, the contrary rulings made and the circumstances and manner in which they were made.
79   On this basis, there was material before the magistrate which entitled her to conclude that there was “evidence … capable of satisfying a jury beyond reasonable doubt” (within s41(2) of the Justices Act 1902) and that there was “a reasonable prospect that a jury would convict the defendant” (within s41(6)).
When the matter returned to the magistrate after Petroulias (No 2), Sweeney LCM ordered committal on a lost opportunity to litigate charge because she regarded the Court of Appeal’s decision about the evidence binding on her. She added that it accorded with her own view of the evidence previously given (see extracts from her reasons set out in the judgment of Sperling J in Petroulias v Wills [2004] NSWSC 739 at [22]-[25]).
One might be excused for thinking that the stage was now set for the trial.
However, the Crown Prosecutor later recast the count in a materially significant manner.
The s29D charge now proffered and its supporting particulars and evidence
As indicated above, the charge now proffered reads (emphasis added):
Between about 1 September 1997 and 27 February 1999, Nikytas Nicholas Petroulias did defraud the Commonwealth, in that while an officer of the Australian Taxation Office, he did put the revenue of the Commonwealth at risk by causing Private Binding Rulings and Advance Opinions to issue to taxpayers, by dishonest means.
By departing from the “lost opportunity” to litigate formulation, this latest form of the count drew back down upon the prosecutor’s shoulders the need to establish that the revenue was actually put at risk. At various stages this has been referred to as the second of the three categories of deprivation mentioned by Toohey and Gaudron JJ in Peters v The Queen (1998) 192 CLR 493 when they said (at 508[30], citations omitted, numbers inserted):
Ordinarily, however, fraud involves the intentional creation of a situation in which one person [1] deprives another of money or property or [2] puts the money or property of that other person at risk or [3] prejudicially affects that person in relation to “some lawful right, interest, opportunity or advantage”, knowing that he or she has no right to deprive that person of that money or property or to prejudice his or her interests.
Senior counsel for the appellant confirmed that the present count is to be understood this way, both before Sully J (Tr 17/02/05 p321) and in this Court.
The Crown Case Statement that has been provided offers Particulars of the prosecution case with reference to the Evidence said to support each Particular. I imply no criticism when I observe that this information sometimes addresses the defrauding (actus reus) and the dishonesty (mens rea) arms simultaneously. This overlap reflects the reality of a Crown case of this nature, where dishonesty is sought to be inferred from the conduct proven.
Particular A(ii) states:
PBRs are issued pursuant to Part IVAA of the Taxation Administration Act 1953 and are binding on the Commissioner of Taxation, provided the factual circumstances specified in the application for the ruling are put into effect. AO’s are also issued by the ATO but are without statutory authority, however the Commissioner considers himself as administratively bound by the decision of the AO.
Section N of the Statement deals with Financial Harm. This item is summarised with the averment:
The action of Petroulias put the economic interests of the Commonwealth at risk – imperilment.
The detailed Particulars are:
N(i)               As a result of the favourable rulings issued the taxpayers who received them and implemented the arrangements in the manner described in the applications, were in a position to lodge returns declaring their income and deductions consistent with the rulings, and were thereby entitled to expect Notices of Assessment issued by the ATO in accordance with the rulings and the returns submitted. The Commissioner of Taxation lost the advantage of independently assessing the tax returns because of the rulings, and was precluded from taking appropriate action to ensure that any further tax which might otherwise have been assessed, was paid.
N(ii)              Similarly, as a result of the Advance Opinions being issued taxpayers were able to take advantage of the protocol applied by the Commissioner of Taxation regarding the AOs as administratively binding upon him, so that the Commissioner of Taxation again lost the advantage of independently assessing the tax returns, and proceeding to take appropriate action to ensure that any further tax assessed, was paid.
N(iii)             Petroulias endeavoured to achieve the same advantage for other taxpayers who entered into similar arrangements, but without the benefit of a PBR or AO, by endeavouring to create a situation where by virtue of the number of favourable rulings and AOs issued to taxpayers entering into the arrangements, the Commissioner of Taxation would also regard himself as administratively bound to follow the precedents for those taxpayers.
N(iv)             In each case, the revenue of the Commonwealth was at risk because of the prejudice to the right or advantage of the Commissioner of Taxation to otherwise act without restriction to assess and recover any tax payable.
N(v)               The PBRs and AOs issued at the direction of Petroulias were arguably wrong.
Senior Counsel for the appellant informed this Court that N(v) is a particular of N(iv) (CCA Tr p2). There is potential latent ambiguity in this observation, if not in the count thus understood. It is possible that the Crown wishes to prove N(v) is an aggravating element of the offence, if only for sentencing purposes. It is also possible that the Crown perceived that proof that the private rulings were “arguably wrong” was an essential element of the count as presently formed. So long as N(v) remains an element of the Crown case the charge in the count stands to be dismissed if it is not established.
The Evidence referable to Particular N(v) is said to be:
“See evidence from H(i) and H(v); Essenbourne Pty Ltd v Commissioner of Taxation; Ingersoll April 04.”
I discuss Essenbourne below. But it should be stated at the outset that (the reasons for) a Federal Court decision are not themselves evidence. Nothing to the contrary was stated in Petroulias (No 1) or (No 2), which went no further than pointing to that decision as a well from which to draw ideas that might become the basis of admissible evidence.
We were informed that “Ingersoll April 04” is an inapt reference for Particular N(v). It deals with the financial consequences of the rulings in question.
The critical evidence for Particular N(v) is therefore the cross reference “See evidence for H(i) and H(v)”.
Particular H avers generally:
Petroulias facilitated the issue of favourable AOs and PBRs to taxpayers using the PIC scheme knowing that the AOs and PBRs were contrary to his view of the ATO policy.
Sub-particulars H(i) and H(v) and the Evidence relating to them read are follows:

H(i) Early in March 1998 Lowman Chow advised Petroulias of his view that Fringe Benefits Tax (“FBT”) applied to the EBT schemes, and on 25 March 1998 Chow provided Petroulias with a paper entitled “Initial Opinion – Employee Benefit Trust (“EBT”)” to the effect that FBT applied to EBT schemes. Petroulias advised Chow that the view expressed by him would be the ATO view.	Chow (5/00) p.6.ann D, E; p.7; Chow (8/00) p.10, 11, 12, 13, ann 25,26,27;
H(v) While the favourable AOs and PBRs were issuing, Petroulias promoted a draft tax ruling (TR 98/D12) which issued on 28 October 1998 to the effect that FBT applied to EBT schemes contrary to the favourable AOs and PBRs issued.	D’Ascenzo (2) p.6-8, ann MDA-E1 to E4, MDA – C1 to C3

Particulars I, J and K should also be referred to. They state:
I.     Petroulias used his position in the ATO to disadvantage promoters of similar EBT schemes:
J.     Petroulias used his position in the ATO to detect threats to the continued marketing of the schemes and to devise new schemes to be promoted and implemented pursuant to the aforesaid arrangement:
K.    The Commissioner was deprived of the opportunity to evaluate the potential tax implications of these schemes.
The evidence said to support sub-particulars H(i) and H(v) was placed before Sully J and we were taken to it. It discloses the following (in summary):
(i)On 16 January 1998 a letter was sent to the Commissioner on behalf of Morgan H R Pty Ltd (MHR) requesting a private ruling in relation to the taxation implications associated with a Productivity Incentive Trust Plan which MHR proposed to establish. The letter outlined the structure of the Plan. It asserted the belief
(a)that MHR would be entitled to an immediate deduction for the amounts contributed by it to the unit trust in respect of participating employees; and
(b)no FBT liability to MHR would arise in respect of the issue of units in the trust.
This application is said to be typical of 74 applications all of which resulted in identical favourable treatment at the respondent’s hands.
(ii)MHR was a company associated with Mr Morgan. The prosecution will seek to prove that Mr Morgan was an associate of the respondent and that the respondent’s dishonesty was evidenced by the steps he took to obtain a favourable private ruling in favour of MHR and other associates willing to enter into dishonest arrangements with him. The prosecution will also seek to establish that “competitors” were given unfavourable rulings at the respondent’s instigation in circumstances where there was no legitimate basis for distinguishing between the schemes laid before the Commissioner by the respective applicants for the rulings (see Particulars I and J above).
(iii)On 25 March 1998 Mr Lowman Chow, who was an officer of the ATO, provided the respondent with a written opinion in relation to Employee Benefits Trusts (see Particular H(i) above). The opinion refers to the steps of “a typical EBT” (Employee Benefits Trust). The Crown will seek to prove that the matters summarised by Mr Chow are in fact indistinguishable from those placed before the Commissioner of Taxation in MHR’s application for a private ruling.
Mr Chow expressed the view that FBT liability would arise. This opinion was supported by detailed legal reasoning and reference to the opinion of a professor set out in detail in the memorandum.
(iv)An email from the respondent dated 15 April 1998 to another officer in the ATO expressed the respondent’s own views as to the tax effects of a scheme which was the subject of an application for a private ruling. The applicant in this case was not an associate of Morgan or the respondent, on the Crown case. However, the basic elements of the scheme were identical to those outlined in the MHR proposal, according to the prosecution. The respondent expressed himself as having reached a “conclusion” that fringe benefits tax was payable and that there were “also problems with deductibility”. The respondent’s reasons for each view are set out. They included express incorporation and adoption of the opinion of Mr Chow.
(v)The Crown also relies on an internal email from the respondent dated 1 June 1998. It is addressed to another officer within the ATO. Once again it is said to concern an application for a ruling on behalf of a “promoter of [a] similar EBT scheme”, not being one of the respondent’s associates. The Crown will seek to show that the circumstances of that “competitor” scheme are indistinguishable from those of the MHR scheme.
(vi)At various dates in 1998 the respondent is said to have assisted in the issuance of several rulings favourable to taxpayers on both the deductibility and FBT issues (Particulars H (iv) – (vii)).
(vii)This Court was informed that the Crown will seek to establish that 74 taxpayers were given favourable private rulings due to the respondent’s intervention. It is said that the information on which each ruling was given was identical in substance. In each instance the ruling was favourable to the taxpayer as regards the deductibility and FBT issues. The rulings enabled the taxpayers to claim the avoidance of tax, this in effect being the actual risk to which the Revenue was subjected by virtue of the respondent’s conduct.
(viii)There is also evidence to the effect that these particular rulings favourable to taxpayers had a certain “precedential” effect, at least in favour of persons associated with particular schemes for which favourable rulings emerged. If necessary, the tax advantages claimed or obtained by these persons are also part of the “actual risk to the revenue” case that the prosecution will seek to prove.
Sully J’s reasons

The learned primary judge first addressed and rejected a stay application based on a concatenation of alleged and mainly procedural deficits. These were complaints about absence of particulars, duplicity within the s29D count, unmanageability of the Crown case, failure to identify evidence, decisions not to call key witnesses, delay and problems of admissibility of vital evidence.
This part of his Honour’s reasons has only peripheral relevance to this appeal. There is nothing in the nature of a notice of contention whereby the respondent contends that the order for a permanent stay should be upheld on bases other than those ultimately accepted by the primary judge.
But before I pass from this section of his Honour’s reasons I record that the judge observed that the Crown Prosecutor, Mr Hastings SC, effectively pinned the Crown case to the particulars and evidence detailed in the Crown Case Statement. His Honour rejected the submission that a trial in accordance with that case was unmanageable.
The second and ultimately dispositive part of the judge’s reasons began by referring to the caselaw about the concept of defrauding. Passages from Peters and Kastratovic were cited.
Sully J described the evolution of the s29D count as “a history of unfortunate vacillation”. However, he stated the true question in the application to be whether the charge as finally formulated was foredoomed to fail.
The judge construed the present charge as requiring proof of actual or real risk to the revenue, “being a risk unqualified by notions of lost opportunities of whatsoever kind”. In other words, this was not to be determined on the basis of the principles discussed in Petroulias (No 1) at [63]ff under the heading Opportunity lost. As indicated, this reading of the count correctly understood the way that the prosecution now intends to run its case.
His Honour said that the two judgments of the Court of Appeal required proof in the present situation that it was at least arguable that the relevant rulings were wrong. He continued:
What does “wrong” mean in such a context? I should have thought that it cannot sensibly mean something other than wrong in law. I think further, that the notion must be one of a ruling arguably wrong in law at the time at which the ruling issued.
What does that conclusion entail in terms of the doing of justice in the projected trial? I think it entails a number of things, leaving aside of course, the further but discrete questions of dishonest means and dishonest intent.
Firstly, and obviously, it entails that the Crown must be able to establish that at the time of the issue of the relevant rulings, there was on foot a real dispute as to the correctness in law of the substance of the rulings. That seems to me to raise a question of fact; and I accept for present purposes that the Crown is in a position to adduce evidenced which it would be proper to leave to the jury on the point.
But where, it must then be asked, is the jury to go from there? To prove no more than that the rulings were controversial within the Taxation Office does not seem to me, just as it did not seem to Simpson J, to prove that the rulings were arguably wrong in law. Such a conclusion could not be reached on any rational basis that did not entail a critical and informed look at the substance of the competing views about the correctness in law of the rulings; and, moreover, an objective critical and informed look.
The question thus posed is, obviously I should have thought, a question of law for the trial Judge, and not a question of fact for the jury. That consideration seems to me to raise a very troubling prospect in connection with the projected trial. That prospect can be put in the form of the question: what is the trial Judge to direct the jury upon the particular aspect of the proofs necessary to make out the Crown case?
It would be relatively straightforward to direct the jury as to their need to make the obvious findings of foundational facts: that is to say, are they satisfied that at the time when each of the relevant rulings was issued, there was within the Taxation Office genuine controversy as to the correctness in law of the rulings?
The only way in which I can see that the trial Judge might then move forward in the directions to the jury, would be for the Judge to instruct the jurors that they must proceed upon the footing that, if they should find those foundational facts in the affirmative and to the requisite standard, then they were directed as a matter of law that the arguments in favour of incorrectness were at least arguably correct in law.
That does not seem to me to be a sustainable approach, for the reason that it does not tell the jury, with the authority of the Court, what some relevant aspect of the law actually is. It is, of course, always possible that a trial Judge will misdirect a jury upon some relevant matter of law. When that happens, and the Court of Criminal Appeal is persuaded to intervene, the necessary correction is administered, broadly speaking, upon the basis either that the Judge has misconceived what the relevant law actually is; or that he has correctly discerned the law, but has misapplied it in its directed application to the possible jury findings of fact.
In the present case, the trial Judge is to be faced with a requirement, not to instruct the jury as to the actual state of the relevant law; but to form an opinion as to what the law might be supposed to have been at the relevant time, so as to proceed to a further conclusion that the relevant contra arguments were arguably correct; and then to instruct the jury that the opinion of the Judge upon that highly controversial question is to become their opinion.
I have considered whether the conundrum might be resolved by the calling of expert evidence. It seems to me, however, that the provisions of the Evidence Act (supra) are to the contrary, for those provisions are directed to the proof of questions of fact. There is a specific exception in the case of foreign law: s 174; but even that exception does not extend to the admission of allegedly expert evidence as to the correct application to the relevant facts of the particular case of the foreign law as proved by the permitted means.
I have kept in mind throughout this discussion the observation of the Chief Justice that important public interests are involved in the applicant’s prosecution. I have been similarly mindful, too, of his Honour’s reminder that it is not for a trial Judge to settle, or to seek to influence the formulation of, or otherwise to dictate the form of, a proposed count in an indictment.
Because of those considerations I am, to speak frankly, reluctant to stay the proposed proceedings on the projected first count. But the more I have thought about the practical problems as I have been discussing them, the more I have come to the conclusion this is a case in which a stay should be ordered.
It seems to me that it cannot be correct in principle to allow a trial to proceed, when something essential to the proof of the Crown case entails the making of a value judgment which raises neither a true issue of fact for the jury, nor a true question of law for the Judge. It is regrettable that the situation of the present matter has come to such a pass; but in my opinion that is the case, and, given the troublesome history to date of the formulation of the s 29D charge, it seems to me that it is time to say, with whatever feeling of dissatisfaction, that enough is enough.
I order therefore that the proceedings on the proposed count based on s 29D of the Commonwealth Crimes Act be stayed permanently.

It may be observed that this did not address the issue earlier stated by the judge as to whether the charge was foredoomed to fail. The nub of the dispositive reasoning was that a fair or just trial could not take place because of the inability to direct the jury that the “arguments in favour of incorrectness [of the private rulings] were at least arguably correct in law”.
Issues in the appeal
The appellant submits that the prosecution of the s29D count should not have been stayed permanently on the grounds applied by Sully J or at all. It is submitted that the judge failed to apply the “foredoomed to failure” test enunciated earlier in his reasons. It is further submitted that the passage set out above relating to the “conundrum” involved in proof that the private rulings were wrong or arguably wrong disclosed a number of particular errors. The issue did not, as his Honour stated, entail “the making of a value judgment which raises neither a true issue of fact for the jury, nor a true question of law for the judge”.

The Crown further submitted that there was ample material capable of establishing the “arguably wrong” element of the Crown case and/or informing the judicial mind to put it in a position to instruct the jury as to the true legal position. I have chosen the word “material” advisedly, so as not to pre-empt the question whether evidence is required on this issue.
In this Court, as below, the parties addressed the question whether the particular matter in issue is one of evidence, requiring proof and judicial instruction to the jury; or one of law, requiring judicial instruction simpliciter.
Each side put as an opening gambit the preferred submission that the issue as to the private rulings being “arguably wrong” was one of law. Here the paths diverged. According to the Crown, it followed that it was merely a question of coming up with the correct direction for the jury. This Court was invited to offer guidance to the trial judge. The appellant submitted that there was nothing peculiarly difficult about the problem: it certainly did not justify a permanent stay.
The respondent’s preferred position was also that the “arguably wrong” issue was one of law. But the ensuing problem, as identified by Mr Temby QC representing the respondent, was that the law would bring itself into disrepute if a judge told a jury that any legal matter was “arguable”. The authority of the court giving directions on other matters such as onus and standard of proof, jury unanimity etc would in some way be undermined by revealing that a state of “arguability” attended any legal matter.
I should say at once that this proposition strikes me as quite untenable. Jury directions usually relate to matters where the law is clear or where, if it is not, the lack of clarity has no relevance to the issues to be tried. In the latter situation the judge does his or her best and the directions will be subjected to appellate correction if legally erroneous. It does not follow that a trial judge cannot direct that a matter of fact or law is debatable, arguable or uncertain if it is relevant to give such an instruction. In any event, the idea that a proposition is capable of more than one answer is something frequently encountered in both life and the law. It is one that a properly instructed jury is well able to assess if it is necessary to do so.

It was alternatively submitted on behalf of the appellant that the Crown could establish that the rulings were “arguably wrong” by relying on the evidence of the Chow memorandum and the other material referred to above. That material reveals the opinion of officers within the ATO that, with respect to schemes said to be similar to those addressed by the respondent, the respondent’s private rulings were wrong, or at least arguably so, as regards both the deductibility and FBT issues. These documents are also said to show that the respondent himself held that view, communicating it to more junior officers in the ATO and applying it to ensure that “competitor” schemes were given rulings adverse to the taxpayer on the two tax issues.
This submission led to debate as to whether the material was capable of supporting these propositions. The respondent submitted that the material rose no higher than evidencing discussion within the ATO, what Sully J described as “no more than that the rulings were controversial within the Taxation Office”.
The “arguably wrong” issues
Section 170BB of the Income Tax Assessment Act1936 provides:
170BB Effect of private rulings on tax other than withholding tax
(1) In this section:
"final tax" has the same meaning as in section 170BA.
(2) Expressions used in this section have the same meanings as in Part IVAA of the Taxation Administration Act 1953.
(3) Subject to sections 170BC, 170BG and 170BH, if:
(a) there is a private ruling on the way in which an income tax law applies to a person in respect of a year of income in relation to an arrangement ( ruled way ); and
(b) that law applies to that person in respect of that year in relation to that arrangement in a different way; and
(c) the amount of final tax under an assessment in relation to that person would (apart from this section and section 170BC) exceed what it would have been if that law applied in the ruled way;
the assessment and amount of final tax must be what they would be if that law applied in the ruled way.
(4) Subsection (3) applies to an assessment whether or not in respect of the year of income in paragraphs (3)(a) and (b).
The “binding” force of a private ruling and the manner in which it can act to the detriment of the Revenue was described in the following terms by Simpson J in her first judgment:
15   Although there appear to be no specific provisions to this effect, I was told, without dissent, that a private ruling made in accordance with Part IVAA is regarded as binding on the Commissioner, and, indeed, in the evidence rulings were at times referred to as “private binding rulings” or “PBR’s”. The Commissioner abides by a convention to give effect to a private ruling made under the Part. The effect of that, as I understand it, is that, provided the factual circumstances are as specified in the application, the Commissioner is precluded, or regards himself as precluded, from challenging a person who acts in accordance with a ruling. Where the ruling is favourable to the applicant, the Commissioner is bound to treat any tax in question as not payable.
16   As I understand Part IVAA, it allows for an assessment on behalf of the Commissioner of the applicability of taxation law; it is an interpretation of the operations of those laws, given in advance, to enable the person seeking the ruling to enter into arrangements with a degree of confidence about his/her/its taxation exposure. It is not conclusive in the sense that a curial adjudication is conclusive, but is an expression of the Commissioner’s view of the effect of taxation law in relation to the particular set of facts and circumstances outlined in the application. It is not intended to confer on the taxpayer any benefit except certainty, and it certainly is not intended to operate as an exemption from tax properly payable. A ruling does not alter the effect of taxation laws, but rather acts as a predictor of the outcome of a hypothetical adjudication.
17   A ruling is individual: that is, it applies to the particular taxpayer who makes the application, in relation to the particular set of circumstances set out in the application and in the year of taxation specified. Rulings do, however, have some precedent effect. The ATO maintains records of the rulings given, and officers asked to give a ruling to one taxpayer will have regard to previous rulings in relation to comparable cases. But a ruling in one case does not bind the Commissioner to make a similar ruling in a similar, or even identical, case, and a taxpayer making an application can have no legitimate expectation that an earlier ruling in a different case will be followed.
Particulars A(ii), N(i) and (ii) (set out above) endeavour to capture these aspects of private rulings. Private rulings themselves point to the legislative provisions that explain the extent to which the ruling is “legally binding on the Commissioner” (see explanatory notes to Notice of Private Ruling issued on 5 May 1998 (AB 207)).
A ruling that is incontrovertibly correct in its statement and application of the tax law could not be the instrument of defrauding the revenue, regardless of the dishonesty of the ATO officer involved. As Simpson J put it ([2002] NSWSC 1190 at [43]):
It cannot be said that a ruling which correctly acknowledges that, in a particular set of circumstances, a taxpayer is not liable for the imposition of a particular tax, confers any benefit upon that taxpayer, or any corresponding disadvantage upon the Commonwealth.
For the purpose of the s29D charge now proffered, the search is for a way of determining whether there was anything of value in the Commissioner’s rights to assess and recover tax as affected by the rulings and the way rulings operate in law. If the true legal position at the time was that nothing touching the revenue was lost by issuing the rulings then the charge is not capable of being made out and should be withdrawn from the jury. This would be so whether or not Particular N(v) were included.
When the proceedings were first before Simpson J the s29D count spoke in terms of “assisting taxpayers to [endeavour to] avoid the payment of tax”.
The Particulars at the time averred (in par H) that the accused “facilitated the issue of favourable (rulings) to taxpayers using the …scheme knowing the rulings were contrary to the view and policy of the ATO” (see [2002] NSWSC 1190 at [24]). [Contrast Particular H in its current form, which speaks of the respondent “knowing that the (rulings) were contrary to his view of the ATO policy”.]
Simpson J recorded that senior counsel for the Crown resisted any suggestion that it would be necessary also for the Crown to establish that the rulings were at least arguably wrong. Counsel nevertheless contended before her Honour that evidence to that effect was available if required (id at [39]).
At that stage, the count was framed in a way that make it necessary that the Crown establish that tax was in fact payable and was in fact avoided by the accused’s dishonest actions. This deficit, as pointed out by Simpson J and the Court of Appeal, led to the first of the count’s reformulations.
By the time the matter had proceeded to the Court of Appeal the count had been redrafted. Indeed, for a while it seemed that it would be further recast as an Opportunity lost variant, sometimes referred to as within the third category of the passage in the judgment of Toohey and Gaudron JJ set out above. Much of the reasoning in Petroulias (No 1) was directed at that variant, which was then the focus of Crown attention.
The Court of Appeal was at some pains to explain why something other than proof that the rulings were “arguably wrong” would suffice if the count charged imperilment of the Opportunity lost variety (see esp Petroulias (No 1) at [59], [71] – [74]). It is however fallacious to read this as adoption of an “arguably wrong” test as regards a charge that was never before the Court of Appeal. No such test was endorsed (see at [62], [64]). In Petroulias(No 2) the Court was primarily addressing an opportunity lost case. The Court was not engaged in redrafting the indictment nor in the task of considering the adequacy of the evidence to commit for trial. A fortiori, it was not involved in the exercise that confronted Sully J.
Now that the count has reached its “final” form it is useful to remind oneself that the relevant matter for ultimate proof is whether the respondent’s conduct in causing the private rulings to issue “put the revenue of the Commonwealth at risk”. This allegation of the actus reus is itself a gloss on s29D (albeit well supported by authority). It is potentially misleading to gloss the gloss by focussing on an “arguably wrong” test.
However, the “arguably wrong” issue must be addressed, for two reasons. The first stems from the proposition I stated earlier, that an incontrovertibly correct ruling could not be an event that puts the revenue at risk. To say that only arguably incorrect rulings have this consequence is an attempt to capture this idea. I perceive that this may be the thinking behind Particular N(v).
The second and possibly different way in which some kind of arguability issue intrudes is that the Crown has elected to make it part of the case, through Particular N(v). This may have been unnecessary (see par [58] above). But so long as the particular remains as an aggravating feature of the conduct charged it will require attention at trial, not necessarily by way of express jury direction.
It may be dangerous or misleading to use the same label (“arguably correct”) in these two senses, at least without confirming the prosecution’s intention in averring Particular N(v) as an aggravating feature and also confirming the meaning sought to be advanced by the Crown in this particular.

100 If all that is involved is arguability of the first type then it seems to me that one is only dealing with whether the count may go to the jury. If the ruling is incontrovertibly correct in law in both its tax aspects there is simply nothing capable of making out the Crown case on the s29D charge. The count should therefore be withdrawn from the jury. But, absent this situation appearing to the trial judge, the count should simply go to the jury according to its terms, without concern about directing the jury to find the rulings were “arguably wrong” (itself a mere particular).

101 If, however, an allegation that the rulings were arguably wrong is intended as an aggravating feature of the Crown case (ie if the Crown wants to use the concept otherwise than just in the first sense I have proposed) then an additional issue will be thrown up at the trial. How (if at all) is the jury to be instructed in this particular matter? Is it really another way of looking at the first type of arguability issue?

102 I return to these latter questions below.

103 It is pertinent to observe that the trial is yet to commence. This appeal does not exist to give preliminary advice as to evidentiary rulings. It is the defence that has assumed the high burden of seeking to establish irremediable abuse of process. The relevant category, on the reasoning of Sully J, is abuse of the nature of irremediable prejudice precluding a fair trial (Jago v District Court of New South Wales (1989) 168 CLR 23).

104 The High Court has frequently cautioned against too ready a disposition to grant permanent stays in criminal matters. Recently in Subramaniam v The Queen (2004) 79 ALJR 116 the Court said at 122[25]:

In Jago v District Court (NSW) Brennan J cautioned against too heavy a disposition to grant stays:
“The reasons for granting stay orders, which are as good as certificates of immunity, would be difficult of explanation for they would be largely discretionary. If permanent stay orders were to become commonplace, it would not be long before courts would forfeit public confidence. The granting of orders for permanent stays would inspire cynicism, if not suspicion, in the public mind.”

105 Sully J did not have to decide what directions to give the jury. Nor does this Court. Naturally, any directions will have to adapt to the case as it develops in the evidence and submissions of counsel. This said, I think it is incumbent that this Court come off the fence on which the primary judge rested when posing his either/or “conundrum”.

106 It is productive first to go to the evidence to consider whether it is capable of supporting the particular charge. If it is, then one major plank of the stay conundrum is removed.

107 That evidence shows that there was, to say the least, uncertainty about the tax effects of employee benefit trust schemes throughout the time of the private rulings issued through the respondent. The application of the law relating to deductibility and FBT to these schemes was a matter of discussion within the ATO. On a clearly available view of the evidence, the ATO (to the respondent’s knowledge) had certainly not given up the ghost on these matters. Taxpayers were also apparently uncertain about the issues, evidenced by the many applications for private rulings on these very matters.

108 Key uncertainties were whether this type of scheme had each of the two benefits of tax deductibility of employer contributions and freedom from FBT. Mr Chow and the respondent thought not, on available readings of the respondent’s email of 15 April 1998 and of the proposals in his emails of 15 April and 1 June 1998 to give taxpayer-adverse rulings to “competitor” schemes. The legal reasons for such outcomes were canvassed in the memoranda, with citation of statute and caselaw.

109 This was more than an internal ATO debate about the response to give applicants for private rulings. The arguments adverse to the taxpayers’ interests that were canvassed in the internal memoranda were about the then current state of the law, its application to the facts hypothesised, and whether the ATO should accept the taxpayer-favourable views advanced by prospective “Rulees”. In my view, the documents are also evidence (if evidence is required) that the rulings issued at the respondent’s instigation had a tax impact in the sense contended for in Particulars A(ii), N(i) and N(ii).

110 If it is established that the respondent caused the issuance of the private rulings favourable to associate taxpayers on the deductibility and FBT matters, he was arming the “Rulees” with the practical and legal advantages conferred by s170BB and described by Simpson J in the passage set out above. For favoured Rulees, matters that were debatable or doubtful achieved the status of being “legally binding on the Commissioner”, with favourable tax consequences that the jury could easily conclude likely to happen. For those Rulees, the ATO was henceforth prejudiced in its capacity to otherwise act without restriction to assess and recover any tax payable (cf Particular N(iv)).

111 This material or evidence is capable of supporting the offence charged unless the rulings were incontrovertibly correct when issued. It is material capable of showing that the opinion was held within the ATO at the relevant time (ie, throughout 1998) that schemes arguably similar to those in question did not carry income tax deductibility for employer contributions and did carry FBT liabilities. The legal chapter and verse on FBT is set out in the email of 25 March 1998, the contents of which the respondent adopted in his email of 15 April 1998 (cf Lustre Hosiery Ltd v York(1935) 54 CLR 314). Given that perception or belief, there was in my view ample evidence or material that the ATO was giving up something of value in issuing taxpaper-favourable rulings.

112 If a jury accepts the evidence that the “competitor” schemes were indistinguishable as regards either or both of their deductibility or FBT aspects, the respondent’s stated opinion that these had adverse tax consequences in those two aspects is also available as evidence or material out of the respondent’s own mouth that the rulings he was procuring in support of “favoured” taxpayers did not represent the true tax position at the time.

113 It will ultimately be a matter for the jury to decide if there was the necessary imperilment of the revenue. In my view, the material that has been identified indicates that it would be open for a properly instructed jury to be satisfied of this. For this reason the charge is not foredoomed to failure. In other words, the rulings were “arguably wrong” in at least the first of the two ways that this issue intrudes into the case. I prefer to state the matter negatively: since the rulings were not incontrovertibly correct (on this evidence and the legal principles to which it points), the count is not foredoomed.

114 If, however, Particular N(v) states an additional part of the positive Crown case it will throw up the question whether and how the jury are to be directed on an allegation that something was “arguably wrong” (in law). This particular matter vexed the primary judge. It is not easy of resolution.

Was a permanent stay justified?

115 Sully J in the lengthy passage explaining his reasons for a permanent stay posed a conundrum. He adverted (somewhat tentatively) to evidentiary difficulties on the one hand and difficulties or uncertainties touching possible legal directions on the other. He appeared to conclude that expert evidence might not be admissible as to whether the private rulings were “arguably wrong”. (No such evidence was led at committal, so far as I am aware. None is foreshadowed in the Crown Case Statement.) Sully J also concluded that the problem was not “a true question of law for the Judge”. The ultimate issue was described as “a value judgment”, with the further implication that the difficulties were insoluble and that the trial could not take place in a manner that was fair and just.

116 The appellant submits that this “conundrum” overstated the extent of the difficulty, which cannot at this point of time be characterised as an irremediable impediment to a fair trial. The judge erred in describing the ultimate issue as a “value judgment”, as if that were a legally impermissible category of reference. Many legal decisions could be so described (cf Evidence Act, ss130, 137, 138). So too would jury decisions about want of reasonable care, for example.

117 The appellant further submits that the problems confronting the trial judge were not of such a degree that the extraordinary remedy of a permanent stay was called for. The evidence was capable of establishing the requisite species of imperilment of revenue. It would be incumbent on the trial judge to assist the jury in addressing the ultimate actus reus issue about the actual or potential impact of the private rulings issued at the respondent’s instigation to the allegedly favoured taxpayers.

118 I accept the submissions summarised in the previous two paragraphs. The complexity of a trial cannot be the touchstone of whether the trial takes place at all. Civil and criminal jury trials may involve complex scientific, medical or accounting issues. Procedures may require adaptation to assist in proof and understanding of the true underlying issues. Recent developments in the law of expert evidence have emphasised the need for demystification and the importance of ensuring that the tribunal of fact is not baffled by science. Opinion evidence must demonstrate its assumptions in a manner that makes the opinion ultimately contestable (Makita (Australia) Pty Ltd v Sprowles (2001) 52 NSWLR 705).

How should the jury be directed in the circumstances?

119 The need to consider the broad scope of appropriate directions remains, since the primary judge expressed concern on the matter in the context of a Crown case that includes particular N(v).

120 The parties were agreed before us that it was open to the trial judge to direct the jury that the private rulings favourable to taxpayers were (in their words) “arguably wrong” when issued. I have already explained that this formulation may cover two separate areas and have concluded that the count is not to be withheld from the jury because the rulings were incontrovertibly correct.

121 What remains as regards directions needing to be given stemming from Particular N(v)? Ex hypothesi the trial judge will have satisfied himself that the rulings were not incontrovertibly correct in law when issued, this satisfaction stemming from considering the evidence and the legal materials (see further below). Merely because N(v) is a particular of the Crown case does not however oblige the judge to direct the jury on the specific matter in the sense of inviting them to find arguability in law. If that issue is itself one of law, then the judge should direct the jury appropriately, not necessarily in those terms.

122 [As the Chief Justice and Hunt AJA point out], such a direction may well be subject to various permutations and combinations of the factual matters. The more usual form of directions to a jury – and the infinitely safer one – is to define the issue which the jury must determine, then explain to them the arguments of both parties in relation to the relevant factual material, and then conclude that, if the Crown has satisfied them that it has established that particular issue, they may move on to the next issue. None of this, it seems to me, requires the jury having to be instructed that they have to decide whether the rulings were “arguably wrong” as distinct from having to be instructed as to how they imperilled the revenue through the particular private rulings and the legal and practical effects of such rulings.

123 I do not accept Mr Temby’s submission that the matter is answered in a favour of a direction in law by pointing to the complexity of the issues involved. There are pockets of caselaw where this factor may have contributed to the recognition of a specific universal rule (for example, as regards the “materiality” factor in perjury and false swearing: see R v Davies (1993) 7 SASR 373 at 388-394, R v Cassell (1998) 45 NSWLR 325). No authority supports the creation of a new category for taxation matters. Tax fraud trials have been part of the legal landscape for many years. A properly instructed jury is able to understand the legal tax principle said to have been dishonestly violated and to form its own opinion as to whether the requisite fraudulent conduct has occurred.

124 But the present situation is different, in my view. The matter to be determined is whether the private rulings were, in the language of Particular N(v), “arguably wrong” (in law) when issued. In the specific context of FBT this matter can be stated thus: was it open in law at the time for the ATO to argue that employees were “associates” within the legal provisions and principles stated by Mr Chow in his Initial Opinion of 25 March 1998 as adopted and reinforced by the respondent’s views stated in his email of 15 April 1998?

125 This concept of being “legally arguable” or open is, in my view, a matter of law. It does not matter that the question is asked about an identified time in the past (here, relevantly, before Essenbourne was decided). Accordingly, it is something that the trial judge (if satisfied as to such “arguability”) should direct the jury with his authority as the arbiter of matters legal. In the final analysis, it is no different from telling the jury that if facts A, B and C are established it is open for the jury to find X. Here “X” is that the revenue was put as risk by the rulings (if they were made as alleged).

126 It seems to me that a mixed question of fact and law is involved (see also Petroulias (No 2) at [46]).

127 Whether or not a particular transaction has particular tax consequences depends on locating the applicable law (statute and general law) and correctly applying it to the essential elements of the transaction. Minds may differ as to the legal consequences, but that is an inevitable aspect of a legal system administered by human agents. This does not relieve a judge from deciding, identifying and applying the relevant law to legal questions. It is not the jury’s competence or task. If the judge errs, correction is at hand through the agency of the Court of Criminal Appeal and the High Court. Subject to the proviso, legal error in admitting or excluding evidence or in directing a jury will be corrected by appellate processes.

128 The judge’s role is to give necessary and helpful legal directions to the jury. The distinction between matters of fact and matters of law is not always clear. It is, however, well established that the judicial duty encompasses directing the jury about relevant statutory and common law principles. That duty extends necessarily to relating those directions to the facts available to be found by the jury.

129 The law understands that some legal propositions can be stated with certainty whilst others cannot. Matters may change over time, such that what is certain in one era may be problematical in another. To give an example: when Sir Matthew Hale’s History of the Pleas of the Crown was published in 1736, the common law was quite clear that a husband could not be guilty of raping his wife. The situation was arguable in the 1980s, but became clear again (to the opposite effect) after 1991 in light of The Queen v L (1991) 174 CLR 379 and Regina v R [1992] 1 AC 599.

130 An expert witness or a judge who is called upon to form a view about the certain or uncertain state of the law at a particular time in history, or its application in law to particular facts, may need to inform himself or herself by reading caselaw, legal digests, textbooks, legal encyclopaedias and law journals. Some may have sufficient legal experience in the matter at hand to trust their own “judgment” on the matter. Either way, the legal expert will often have to apply that knowledge to a particular set of proven or hypothetical facts. Such application is itself a legal function, because it is law that answers issues of relevance and capacity to reach certain conclusions.

131 A judge can therefore direct a jury that a statute or contract means X. If relevant, the judge can also direct the jury that the relevant law on the subject is either A or B, with the consequence that when applied to a situation or contract the result is X (or X or Y as the case may be). An analogy is the situation where the jury is told that it is “legally open” for them to decide A or B, but not C.

132 Not every application of a legal standard to a particular problem involves a question of law. But the issue whether a particular tax outcome was available, a fortiori arguable, is a matter of law (see generally Vetter v Lake Macquarie City Council (2000) 202 CLR 439 at 450[24]). Mr Chow’s discussion about whether, on the assumed facts, an employee is arguably an “associate”, with FBT consequences, appears to me to entail a sharp legal issue. It is not quite so clear with the deductibility issue because factual inferences of intent may be open, although I perceive that identical fact scenarios were provided on behalf of all applicant “Rulees”.

133 I have already indicated that the putative vice of the respondent’s private tax rulings was that they restricted the Commissioner’s capacity to act against the taxpayers concerned otherwise than as represented in the rulings (cf Petroulias (No 1) at [64]). Viewed this way, the risk or imperilment was the tying of the Commissioner’s hands in 1998, restricting his capacity to contend against favoured “Rulees” that the situation was other than as represented in their rulings.

134 It follows from what I have already written that it would be open to the trial judge to inform himself of the state of tax law as it stood when the rulings were issued. If that state of law permitted the Commissioner genuinely to advance a tax outcome in either or both of two tax issues otherwise than as embodied in the taxpayer-favoured rulings promoted by the respondent, then the judge could so direct the jury. The jury would then be directed that, if the factual elements of the Crown case were established, it would be open for them to find the necessary imperilment of the revenue of the Commonwealth. This is the ultimate fact in issue (along with dishonesty).

135 This approach avoids using “arguably wrong” as part of the direction to the jury. But it does mean that the “aggravated” Crown case goes to the jury if the judge has the requisite satisfaction that an aggravated case as alleged is open in law and on the facts. If it is, the judge may craft a direction addressing the elements of the count itself.

136 As indicated above, the ATO internal emails and memoranda summarised above are capable of being read as containing admissions by the respondent that are at least arguably consistent in 1998 with “the law” applied to the facts hypothesised as the judge may be persuaded to find it. Indeed, the judge may be assisted by the “chapter and verse” about the critical issue of the meaning of “associate” in s 26AAB(14)(a)(iv) of the Income Tax Assessment Act in the Chow memorandum, apparently adopted as correct by the respondent.

137 The judge may also draw assistance from case law both before and after the rulings in question (see further below as to Essenbourne). Of course, everyone knows that it is easy to be wise after the event, even in matters legal. Some cases apply settled law, some others settle the law, others unsettle the law. In the present case the critical point of time is when the rulings were issued, because they are the events said to have put revenue at risk. It also needs to be borne constantly in mind that the relevant imperilment under the present formulation of the count is the prejudice to the capacity of the ATO to act contrary to the position of the rulings against “Rulees” who received the favourable rulings. That prejudice stemmed from the law and practice touching private rulings. It existed unless the rulings were incontrovertibly correct when given.

138 I have come down on the legal side of the fence in the sense that I consider it to be a matter for the trial judge to make up his or her mind on the legal issue as I have sought to explain it. But the legal issue is a very narrow one: there are critical and associated factual issues that must be left to the jury, along with the ultimate issue of finding the matters averred in the count beyond reasonable doubt.

139 This does not mean that the judge can necessarily avoid explaining to the jury the legal issues canvassed in the ATO emails, especially the Chow opinion. An understanding of these matters may be important in evaluating the respondent’s role in procuring the rulings as well as assessing the separate dishonesty factor. This will not however entail asking the jury to form their own view about the correspondence between the rulings and the true tax position at the time.

Essenbourne and later decisions of the Federal Court

140 EssenbournePty Ltd v Commissioner of Taxation [2002] FCA 1577, 51 ATR 629 and later decisions including Walstern v Commissioner of Taxation [2003] FCA 1428, 54 ATR 423, Spotlight Stores Pty Ltd v Commissioner of Taxation [2004] FCA 650, 55 ATR 745 and PrideCraft Pty Ltd v Commissioner of Taxation [2004] FCAFC 339, (2005) ATC 4001 might put an end to the “arguability” of certain propositions for the future. It may be going too far to say that these cases show that the position advanced by the ATO to the Federal Court in those cases was “arguable”, whether or not it was accepted. But it would be fallacious to infer that an ATO-adverse decision in the Court established that the propositions there advanced by the Commissioner were untenable and unarguable then or at an earlier point of time.

141 Essenbourne is a decision of Kiefel J in the Federal Court of Australia. Coincidentally it was handed down the day after Simpson J made the order granting leave to appeal with respect to the original order to commit for trial (Essenbourne Pty Ltd v Commissioner of Taxation [2002] FCA 1577, 51 ATR 629). The decision was not taken on appeal. The present parties have always accepted that it must be taken to be incontrovertibly correct as to what it decides, but they have disagreed as to its application to the particular matter at issue.

142 In her second judgment Simpson J referred to the impact of Essenbourne in the following terms ([2003] NSWSC 106 at [7]):

[U]nless and until another taxpayer seeks to litigate another relevantly similar case, the judgment of Kiefel J states the law on the efficacy of the schemes and the liabilities to taxation of taxpayers implementing arrangements of a relevantly similar type. That, of course, does not have the same degree of finality as a decision of the High Court at the end of the available appellate process. Nevertheless, as the law presently stands, as stated by Kiefel J, the arrangements the subject of the private rulings in question being essentially and relevantly similar to the arrangements in Essenbourne, no tax is payable by the taxpayers in question, and the Commonwealth has been deprived of nothing of value. The most that could be said is that the Commonwealth does not have the opportunity to litigate the efficacy of the arrangements. But what does that amount to in the light of the decision of Kiefel J? In order further to litigate such a scheme, it would be necessary for the Commissioner to levy tax on a taxpayer (either by disallowing a claimed income tax deduction, or by requiring the payment of fringe benefits tax); for the taxpayer to commence proceedings; for the Commissioner to persuade the judge (or an appellate court) that Essenbourne was wrongly decided and that, in the circumstances, tax was payable. Although it is complicated by the fact that Essenbourne was partly settled by negotiation, given that the Commissioner opted not to challenge that decision, it seems unlikely that the opportunity to litigate a scheme or arrangement of the same kind is of any real value.

143 This conclusion strongly influenced Simpson J in her decision to quash the decision to commit for trial on the s29D charge as formulated at the time (see [2003] NSWSC 106 at [9]).

144 The Court of Appeal held that Simpson J had overstated Essenbourne’s impact. The Court nevertheless had regard to Essenbourne in the exercise of its discretion as to the remedy flowing from a conclusion that the committing magistrate had committed an error of law (cf Justices Act, ss 109-110). But it did not treat Essenbourne as having direct evidentiary impact.

145 The matter was complicated by an erroneous submission by the claimant Wills (the informant) in the hearing of Petroulias (No 1). This in turn led to the second hearing in the Court of Appeal (see (2003) 58 NSWLR 598 at 609[42]-[47], 624[33]). The upshot was that the Chief Justice ultimately addressed the Essenbourne issue as follows (Petroulias (No 2) at 624[40]-[49]:

40 It was the Opponent who presented Essenbourne as a matter to which her Honour should have regard on the exercise of the discretion. This Court is also entitled to do so. As I said in par [84] of the previous judgment, no submission to the contrary was made. I do not resile from the conclusion that Essenbourne indicates that there was an arguable case on the fringe benefits tax issue. I also concluded at [84] that Essenbourne establishes that the deductibility issue was “strongly arguable”. In view of the subsequent submissions, I would withdraw the word “strongly” but reaffirm the basic proposition.
41 Mr Clelland contends, as Mr Richter contended before Simpson J, that the issue of deductibility was always open for determination by a court on the evidence in a particular case. The submissions in this Court focus on that part of the exchange with Simpson J quoted above, which refers back to earlier consideration that issues of deductibility may depend on the facts of the particular case. It was submitted that par [1] of the Private Ruling, tendered in the proceedings as representative of other rulings, has a conditional quality about it. The application of the ruling depends upon the facts of a particular case.
42 I do not regard the submissions of Mr Hastings before Simpson J, with respect to the effect of different factual situations, as constituting a proposition from which he sought to resile on the appeal. It can be readily accepted that a claim for deductibility may be rejected on different grounds depending on the facts.
43 For example, one of the matters that may affect the outcome of a claim for deductibility in a particular case is identified in par [3] of the Private Ruling, quoted in par [21] of this Court’s previous decision, requiring any contributions to be ‘reasonable and not excessive’. This and other such matters can vary from case to case.

44 The written submissions of the Opponent referred to pars [24]-[29] and [33] of Essenbourne which were described as “findings as to the subjective intent of the taxpayer”. In those paragraphs Kiefel J was concerned with subjective elements on the issue of deductibility. These matters do turn on facts as the Opponent submitted. Mr Clelland referred the Court to the subsequent confirmation of the factual basis of this part of the decision in Essenbourne by Hill J in Walstern v Commissioner of Taxation [2003] FCA 1428 at [70]-[72].
45 However, that was not the only basis on which her Honour found for the Commissioner on deductibility. In pars [34]-[36] of her judgment, Kiefel J went on to deal with an alternative contention on the part of the Commissioner that the payment was of a capital nature. Her Honour concluded that the expenditure was of a capital nature. Hill J in Walstern also noted that this raised different considerations. (See at [79]-[80].) Factual issues may still arise. Mr Clelland drew particular attention to the possibility of annual contributions under the PIC scheme in contrast with the one-off nature of payments in Essenbourne and Walstern.
46 The proposition that the outgoing was of a capital nature remains distinctly arguable under the PIC scheme. It is a matter which par [1] of the Private Ruling could prevent the Commissioner contesting. The deductibility question is generally a question of mixed fact and law. The rulings determined that the facts submitted and assumed in the application for a ruling entitled the taxpayer, as a matter of law, to the tax treatment set out in the ruling. The Commissioner was deprived by the ruling of the opportunity to argue otherwise.
47 Accordingly, the Commissioner in the schemes presently before the Court, has been deprived of the opportunity of arguing that a scheme complying with the application – which was handed up to this Court – for the purpose of both ‘retaining key employees’ and ‘increasing employee productivity and profitability’ with the possibility of annual contributions, was expenditure of a capital nature. Accordingly, par [1] of the Ruling was capable of having an impact constituting the element of defrauding.
48 I would reach no different conclusion to that of the first judgment. Even if s110 did not require the Court to do so, the discretion under s109 should not be exercised to quash the order for committal.
49 The issues not resolved on the evidence before the magistrate, e.g. the arguability of the fringe benefits tax issue or both arguability and ‘genuine dispute’ or ‘genuine assertion of a position’ on deductibility are real issues which may still arise. They could arise on a reopening of the committal before the magistrate or on a Basha inquiry before a trial. These are choices which the Crown has yet to make. Important public interests are involved in these prosecutions. They should not be compromised by the exercise of a discretion to quash the order for committal in the circumstances of the proceedings as a whole

146 This extract from Petroulias (No 2) established that:

1.Essenbourne does not preclude the Crown in limine from proving that the respondent’s private rulings put the revenue at risk; and

2.Factual issues needing to be explored at trial may distinguish the scheme considered in Essenbourne and the schemes covered by the respondent’s private rulings.

147 The appellant submits that the Court of Appeal concluded that Essenbourne “established that the issues were arguable and that the primary judge was in error in reaching a different conclusion”. In my opinion, this misstates the reasoning of the Court of Appeal in Petroulias (No 1) and Petroulias (No 2).

148 In Petroulias (No 2) the Chief Justice said that Essenbourne did not justify the conclusion that the Commissioner had lost the deductibility issue (see at [34]-[47]). But this did not entail the contrary conclusion that Essenbourne overtook the evidence at committal, removed the need to consider the impact of that evidence or prevented the trial judge from considering “arguability” at an earlier point of time. This becomes crystal clear in pars [49] and [67] of Petroulias (No 2).

149 It is true that Essenbourne was not the knock-out blow that Simpson J had perceived. But it is equally true that the Court of Appeal was flagging that the Crown had to lay the groundwork at committal to show that the rulings had the requisite character as regards either or both of the deductibility or FBT arms. I would therefore reject the Crown submission that Sully J erred in failing to treat Essenbourne as “relevant and sufficient [evidence]” to warrant the accused being put on his trial (Crown Outline at par 8).

Disposition

150 For these reasons I would uphold the appeal and set aside the permanent stay ordered by Sully J on 21 February 2005.

**********

LAST UPDATED: 16/03/2005