Cassaniti v Ball as liquidator of RCG CBD Pty Limited (in liq) and related matters; Khalil v Ball as liquidator of Diamondwish Pty Ltd (in liq) and related matters

1. [1]

   THE COURT: These reasons are organised under the following headings.

Introduction

1. [2]

   Fifteen appeals were heard together arising from the principal judgment and a supplementary judgment of Ward CJ in Eq concerning proceedings brought by five companies and their liquidator making claims of breach of fiduciary and statutory duty, accessorial liability in equity and under statute for knowing assistance or knowing involvement in those breaches of duty and also claims of unreasonable director-related transactions and uncommercial transactions under Pt 5.7B of the Corporations Act 2001 (Cth). The five proceedings were heard together in the Equity Division over 29 days in February and March 2020.

2. [3]

   Five of the appeals are brought by Gino Cassaniti, five by the Khalil parties and five by the liquidator. This judgment deals with the appeals by Gino Cassaniti and the Khalil parties (relevantly, Faouzi (Fred) Khalil, Peter Abboud and George Khalil) which raise a single issue in common with ground 1 of a notice of contention by the Borg parties in the liquidator’s appeal against the Borg parties. This judgment also deals with the two appeals by George Said which raise a different issue from that raised by the other Khalil parties.

3. [4]

   The liquidator’s five appeals on behalf of the companies against part of the judgment and orders of Ward CJ in Eq are the subject of separate reasons delivered contemporaneously with this judgment: see Bluemine Pty Ltd (in liq) v AKA (Civil) Pty Ltd [2022] NSWCA 160.

4. [5]

   The facts giving rise to the appeals can be summarised briefly. Earth Civil Australia Pty Ltd (in liq) (Earth Civil), Bluemine Pty Ltd (in liq) (Bluemine), RCG CBD Pty Ltd (in liq) (RCG), Diamondwish Pty Ltd (in liq) (Diamondwish) and Rackforce Pty Ltd (in liq) (Rackforce) were each wound up voluntarily by special resolution of its members in either June or August 2013. Mr Mitchell Ball was appointed liquidator of each company.

5. [6]

   The liquidator, together with each company, then brought separate proceedings against numerous defendants. The liquidator’s elaborate pleadings included an alleged conspiracy by an accounting firm, Banq Accountants and Advisors Pty Ltd (Banq), and principals of or persons employed by that firm, namely Gino Cassaniti, Fred Khalil, Peter Abboud and George Khalil, to promote and recommend a tax scheme to clients of the firm. The alleged conspiracy was only relied upon to establish that the nature of the alleged breaches of fiduciary duty by the relevant director of Earth Civil, Bluemine, RCG, Diamondwish and Rackforce were dishonest. No claim was made by the liquidator based on the tort of conspiracy, nor was it alleged that the defendants said to be liable as accessories to the breaches of duty were parties to a conspiracy.

6. [7]

   Stripped of their complexity, the pleaded claims alleged that various defendants who were referred to as scheme participants entered into and implemented a series of transactions, by arrangement with Gino Cassaniti, the director of Earth Civil, Bluemine and RCG, or Frank Criniti, the director of Diamondwish and Rackforce, and that those transactions constituted what are known as “carousel” frauds. It was alleged that the transactions involved each of Earth Civil, Bluemine, RCG, Diamondwish and Rackforce (a) without providing consideration, receiving money to which it had no legal entitlement, (b) without receiving consideration, paying money without legal obligation, and (c) being left with no money to satisfy its only legitimate creditor, the Commissioner of Taxation, upon being wound up in a creditors’ voluntary winding up within two years of incorporation.

7. [8]

   The alleged purpose of the carousel payments was for the relevant defendants to obtain improper tax benefits by claiming GST input credits and income tax expense deductions and thereby reduce its taxable income.

8. [9]

   The case against the scheme participants was based on accessorial liability under the second limb in Barnes v Addy (1874) LR 9 Ch App 244 and the provisions of the Corporations Act. The equitable claims alleged that the respective breaches of fiduciary duty by Gino Cassaniti and Frank Criniti involved a dishonest and fraudulent design, and that the scheme participants knowingly assisted or knowingly participated in the dishonest and fraudulent design of those directors.

9. [10]

   The statutory claims relevantly alleged that Gino Cassaniti and Frank Criniti breached ss 181 and 182 of the Corporations Act, in the case of Gino Cassaniti, as a director of Earth Civil, Bluemine and RCG, and, in the case of Frank Criniti, as a director of Diamondwish and Rackforce, by failing to exercise his powers and discharge his duties in good faith in the best interests of the company and for a proper purpose (s 181) and by improperly using his position to gain an advantage or to cause the company detriment (s 182).

10. [11]

    Sections 181 and 182 are civil penalty provisions: Corporations Act, s 1317E. By s 1317H(1), a court may order compensation to be paid for “damage suffered by the corporation” if the person has contravened a civil penalty provision and “damage resulted from the contravention”. The damage claimed to be suffered by Earth Civil, Bluemine, RCG, Diamondwish and Rackforce was the amount of taxation liabilities incurred by those companies by reason of payments received from the scheme participants, and the amounts which were paid away by Earth Civil, Bluemine, RCG, Diamondwish and Rackforce for no corporate purpose or benefit.

11. [12]

    A person who is involved in a contravention of s 181(1) and s 182(1) also contravenes those provisions: ss 181(2) and 182(2). By s 79 of the Corporations Act, a person is involved in a contravention if, and only if, the person was relevantly, in this case, “in any way, by an act or omission, directly or indirectly, knowingly concerned in or party to the contravention”: s 79(c). Section 79 is a definitional provision.

12. [13]

    In her principal judgment delivered on 6 August 2021 (In the matters of Earth Civil Australia Pty Ltd, RCG CBD Pty Ltd, Bluemine Pty Ltd, Diamondwish Pty Ltd and Rackforce Pty Ltd (all in liq) [2021] NSWSC 966), her Honour relevantly held that:

13. [14]

    Her Honour also found that the liquidator’s claims against several defendants failed, including the accessorial liability claims by RCG against the Borg parties (Borg Family Pty Ltd, Tanya Borg, Borg Civil Australia Pty Ltd and Michael Borg).

14. [15]

    In a supplementary judgment delivered on 15 September 2021, her Honour granted declaratory relief and entered judgments and made orders in favour of Earth Civil, Bluemine, RCG, Diamondwish and Rackforce against, relevantly, Gino Cassaniti and Fred Khalil, Peter Abboud and George Khalil, and in favour of RCG and Bluemine against, relevantly, George Said. Her Honour dismissed the claims against, relevantly, the Borg parties: In the matters of Earth Civil Australia Pty Ltd, RCG CBD Pty Ltd, Bluemine Pty Ltd, Diamondwish Pty Ltd and Rackforce Pty Ltd (all in liq) (No 2) [2021] NSWSC 1161. It is not necessary to refer to the detail of the relief granted against Gino Cassaniti and the Khalil parties. It is sufficient to observe that monetary judgments in significant sums were entered against those defendants.

15. [16]

    The common issue raised by the appeals by Gino Cassaniti and three of the Khalil parties (Fred Khalil, Peter Abboud and George Khalil) against RCG, Bluemine, Diamondwish and Rackforce, and by the Borg parties’ notice of contention (ground 1) in the liquidator’s appeal on behalf of RCG against the Borg parties is the “release” defence relied upon by these defendants at trial, which was rejected by her Honour. No other grounds were ultimately pressed by either Gino Cassaniti or the Khalil parties (except George Said). Nor were the appeals by Gino Cassaniti and the Khalil parties against Earth Civil pressed.

16. [17]

    The two appeals by George Said, as part of the Khalil appeals, raise a different issue. He challenges the finding that he had the requisite knowledge of Gino Cassaniti’s breaches of duty to amount to accessorial liability in equity and under statute.

The appeals by Gino Cassaniti, the Khalil parties (except George Said) and the Borg parties’ notice of contention

1. [18]

   It will suffice for present purposes to summarise the background insofar as it related to the defendant Ms Ivana Cassaniti, although she did not reagitate the “release” defence by way of notice of contention on the appeal to which she was a respondent.

2. [19]

   Ms Cassaniti was sued on the basis that she was knowingly involved in breaches of fiduciary and statutory duties owed by Mr Frank Criniti to Rackforce and Diamondwish. The primary judge found that Mr Criniti had caused Rackforce and Diamondwish to enter into carousel transactions for the purpose of incurring liabilities to the Commissioner of Taxation which they could never pay, and that this was a dishonest and fraudulent breach of his fiduciary duties as well as a breach of his statutory duties as sole director of both companies. The liquidator sued Ivana on the bases that she was liable in equity for knowingly assisting those breaches of fiduciary duty, and under s 79 of the Corporations Act as a knowing participant in the breaches of statutory duty. However, the liquidator had, in exchange for payment and a promise of assistance, granted a release to Mr Criniti.

3. [20]

   Ms Cassaniti’s senior counsel first raised the following point, towards the end of the trial (apparently, shortly after he was retained). Ms Cassaniti submitted that the liquidator’s release of Mr Criniti from liability for breaches of equitable and statutory duties owed by him meant that the liquidator could no longer pursue her for claims that she had knowingly assisted and knowingly participated in the same breaches of those duties

4. [21]

   Numerous other defendants availed themselves of the same point (the details are succinctly summarised at [2348]-[2351] of her Honour’s reasons and need not be reproduced here). This defence had not been pleaded. It needed to be pleaded, turning as it did on the effect of deeds between the liquidator and other parties. The primary judge granted leave to amend, despite the evidence having closed, and no appeal was brought from the decision to grant the amendment.

5. [22]

   Depending on whether ancillary liability is characterised as primary or secondary, the analysis might in principle be different depending upon whether (a) a knowing assistant or knowing participant sought to take advantage of a release of the person who owed the equitable or statutory duty to the company in liquidation, or (b) the person owing the equitable or statutory duty sought to take advantage of a release of a knowing assistant or knowing participant. Not least because of the use of language such as “accessorial” and “ancillary” and “secondary”, it has been thought from time to time that the liability of a knowing assistant or participant may be parasitic upon the liability of the fiduciary (for an account of the debate see J Dietrich and P Ridge, Accessories in Private Law (Cambridge University Press, 2016), pp 12-25). However, in Michael Wilson & Partners Ltd v Nicholls (2011) 244 CLR 427; [2011] HCA 48 at [106] it was said that “liability to account as a constructive trustee is imposed directly upon a person who knowingly assists in a breach of fiduciary duty”, and that the language of “accessorial” does “no more than recognise that the assistant’s liability depends upon establishing, among other things, that there has been a breach of fiduciary duty by another”. That authoritative recognition that the liability of a knowing assistant is not secondary or parasitic upon that of the fiduciary means that it is sufficient to consider the case where the fiduciary is released and the knowing participant is sued (for it is difficult to see how the position under statute could be more generous to defendants).

6. [23]

   The primary judge correctly observed that there was no binding authority and addressed a very wide range of judicial and academic writing bearing on the effect of a release upon one of a number of co-obligors in equity, before resolving it as a question of principle. Ultimately, her Honour rejected the submission on the bases that (a) s 95 of the Civil Procedure Act 2005 (NSW) abrogated the common law rule that a release of one joint and several wrongdoer releases all other joint and several wrongdoers, and (b) in any event, the rule did not apply in equity.

7. [24]

   The subject addressed by s 95 of the Civil Procedure Act is “joint liability”:

8. [25]

   Despite being found in the Civil Procedure Act s 95 is evidently substantive, and deals with the circumstances when joint liability is discharged or qualified It will be convenient to notice the following textual points concerning the application of the provision immediately, because they may not all be immediately apparent:

   1. (1)

      the provision is expressed to apply “in any proceedings”;

   2. (2)

      the provision is expressed to apply only where a judgment is given against some but not all of persons who have a joint liability;

   3. (3)

      “liability” is expressed to include liability in contract, tort and under statute;

   4. (4)

      however, the provision does not apply to judgments to which s 5(1)(a) of the Law Reform (Miscellaneous Provisions) Act 1946 (NSW) applies;

   5. (5)

      nor does the provision affect rights of contribution.

9. [26]

   Her Honour’s account of the parties’ submissions and reasons disposing of the issues presented are extensive. They amount to a valuable review of the entire area, including not merely Australian but also New Zealand, United Kingdom, Canadian and United States decisions and academic writings on the point. We have been considerably assisted by that account, which occupies in excess of 200 paragraphs ([2252]-[2485]). However, the point being one of law, it would add needlessly to the length of this judgment to summarise that reasoning.

10. [27]

    In this Court, the principal submissions challenging the conclusion reached by her Honour on s 95 were as follows:

    1. (1)

       The premise of s 95 was that a judgment had been given. Section 95 abrogated the former rules concerning the merger of single causes of action into judgment, but said nothing of the position here, where there had merely been a release. “[T]he legislature has selected one aspect of the principle, namely, the merger rule … The provision says nothing as to the joint release rule or that the whole principle should be held to have gone”. That was said to be supported by the extrinsic materials and the commentary upon the section and its predecessor, s 97 of the Supreme Court Act 1970 (NSW).

    2. (2)

       Section 97 was enacted in 1970 following the very substantial review by the NSW Law Reform Commission in its Report on Supreme Court Procedure (LRC 7), as part of the long delayed introduction of the judicature reforms in New South Wales. Section 97 was a related reform drafted by the Commission, of which it wrote at para 25:

    3. (3)

       The (current) commentary in Ritchie’s Supreme Court Procedure NSW includes the following:

    4. (4)

       The appellants also sought to rely upon the definition of “liability” being expressed in terms of “liability in contract, a liability in tort and a liability under a statute”, and not explicitly referring to liability in equity.

11. [28]

    The appellants recognised that it was necessary also to challenge the reasoning to the effect that even if s 95 was inapplicable, equity should not follow the common law joint release rule. The appellants emphasised what had been said in Yeshiva Properties No 1 Pty Ltd v Marshall [2005] NSWCA 23; 219 ALR 112 at [80] by Bryson JA with the agreement of Mason P and Beazley JA:

12. [29]

    The appellants also said that the joint release rule had been “too often affirmed to admit of doubt”, and “the correct approach is by legislative reform and not unilaterally declining to follow the common law rule”.

13. [30]

    In response, the liquidator submitted that in circumstances where “s 95 abrogates the merger into judgment rule, there is no point of principle or authority in support of the appellant’s proposition that the joint release rule itself remains unaffected, because the joint release rule is based on the cause of action merging into judgment”. The liquidator said that there could be no principled reason for excluding settlement, as opposed to judgment, from the beneficial effect of the legislation, and that “[s]trict adherence to the text of the legislation at the expense of purposive interpretation of legislation would defeat the beneficial purpose of s 95”.

14. [31]

    For the reasons which follow, we have concluded that the primary judge reached the correct conclusion on both points. Insofar as there was a rule which had the result that a release of a fiduciary brought about the release of a person who knowingly assisted in that fiduciary’s breach of duty, it cannot survive the enactment of s 95 of the Civil Procedure Act. Further, the better view is that there was no such rule in equity.

15. [32]

    Two observations may be made at the outset concerning the parties’ submissions and the way argument was presented on appeal. One derives from the analysis of the primary judge at [2477]-[2478] regarding the finding in Grimaldi v Chameleon Mining NL (No 2) (2012) 200 FCR 296; [2012] FCAFC 6 holding that the liability of a fiduciary and a third party assistant acting in concert was joint and several. The primary judge pointed to the incongruity of the result that a fiduciary and a third party who acted in concert were jointly and severally liable, such that the release rule applied, while a third party who did not act in concert with the fiduciary was not jointly and severally liable, and therefore would not be released by a release of the fiduciary if the release rule applied. The appellants said that the difficulty arose not from the joint release rule, but from the decision in Grimaldi itself. The liquidator’s written submissions challenged reliance on the obiter dicta in Grimaldi to the effect that where the principal and accessory acted in concert for their mutual benefit, their liability was joint and several to compensate for loss. The liquidator said that (a) that proposition was wrong in law and (b) no factual findings had been made sufficient to establish that there was the requisite acting in concert (this reflecting the fact that the point was only raised later in the trial).

16. [33]

    The second observation is that the liquidator added that there was no challenge to the breach of statutory duties nor the findings of knowing participation for the purposes of s 79 of the Corporations Act, and that it had been held that the statutory duties were owed severally by the directors. In Daniels v Anderson (1995) 37 NSWLR 438 at 505, Clarke and Sheller JJA had said, of the duty at common law owed by directors, that “[t]he duty is a common law duty to take reasonable care owed severally by persons who are fiduciary agents bound not to exercise the powers conferred upon them …” (emphasis added). The liquidator observed that s 185 confirmed that ss 180-184 were in addition to and did not derogate from other duties owed by directors. Hence it was said that there was “clear statutory indication that the statutory duties are stand-alone duties and their force and effect should be construed having regard to the statutory text, and that their effect must not be bound by equity”.

17. [34]

    The appellants did not address the second point, to the effect that even if everything that was said about liability in equity and s 95 of the Civil Procedure Act was accepted, that still left them liable for knowingly participating in breaches under the Corporations Act, despite this being drawn to their attention during the hearing, save in a brief supplementary note. For his part, the liquidator made only brief oral submissions. No oral submissions at all were made by the liquidator developing his challenge of Grimaldi. It should be made clear that Ivana, whose recently briefed senior counsel had introduced the issue, elected not to participate in the argument on this point on the appeal, but rather confined her submissions to defending the demeanour-based findings made by the primary judge to the effect that she was not knowingly involved in the breaches. It was perfectly appropriate for Ivana to adopt that approach. But the consequence was to diminish the assistance the Court had from argument on the point.

18. [35]

    Sometimes courts have to resolve questions of law without the benefit of argument, including oral argument. However, it is generally undesirable to do so, especially if it involves departing from the reasons of another intermediate appellate court, and the precedential value of a decision made in the absence of argument is diminished for that reason. These grounds can be resolved on the basis of the points which were fully developed in the parties’ submissions, and for that reason, should be so confined. The principle of judicial economy reflected in this approach is discussed in Massoud v Nationwide News Pty Ltd; Massoud v Fox Sports Australia Pty Ltd [2022] NSWCA 150 at [35]-[41].

19. [36]

    It is convenient, in an area which is far from straightforward, to start with statute. The appellants’ submissions focussed on the text of s 95, and the requirement for there to be a judgment in order for the provision to be engaged. No judgment having been entered against or in favour of Frank Criniti, the appellants submitted that the section could not override the rule that the release of a fiduciary must also release a third party knowingly involved in the fiduciary’s breach of duty.

20. [37]

    That submission has the attraction of simplicity, but it cannot suffice to resolve the point. There are two reasons for this. The first is that s 95 was directed to altering or abrogating a pre-existing rule, and its content can only be understood against the rule which was altered or abrogated by the provision. Secondly, it is to be borne in mind that the appellants’ argument is not that they fall within the terms of s 95. The appellants say that s 95 was not engaged, and therefore they can continue to take advantage of the pre-existing rule which, according to them, survives the enactment of s 95 and entitles them to a discharge. That is to say, an implicit but essential element of the appellants’ submission is that s 95 has no effect on any pre-existing rule of judge-made law applicable in circumstances which do not involve giving or entering a judgment. Thus, it is necessary not merely to analyse the explicit effect of s 95, but also how and to what extent it amends or abolishes the pre-existing rule.

21. [38]

    Those considerations make it essential to consider not merely the text of s 95, but how the section came to be enacted. The background is not without interest. The essential steps are:

    1. (1)

       Section 95 was a consolidation of s 97 of the Supreme Court Act 1970 (NSW);

    2. (2)

       Section 97 of the Supreme Court Act was introduced as one of the reforms effected following the delayed enactment of judicature legislation in New South Wales and was proposed following consideration by the Law Reform Commission in LRC 7. It derived from the Uniform Joint Obligations Act, model legislation drafted by Samuel Williston acting under the auspices of the Commissioners on Uniform State Laws, although the Law Reform Commission consciously departed from that proposed uniform legislation.

    3. (3)

       Section 97 of the Supreme Court Act (and its modern equivalent, s 95 of the Civil Procedure Act) are explicitly linked to the reforms made by s 5(1)(a) of the Law Reform (Miscellaneous Provisions) Act 1946 (NSW). As will be seen below, the language is similar and they were directed to the same end but operating in the area of joint liability which was not joint liability in tort. They should be construed in the same way.

22. [39]

    The United States ancestry of s 95 is unusual. However, it is neither controversial nor concealed. Indeed, it is apparent from the paragraph of LRC 7 reproduced above, on which the appellants relied, and from the legislative side-notes to s 95 of the Civil Procedure Act and s 97 of the Supreme Court Act 1970.

23. [40]

    Williston’s model legislation is reproduced in the Annual Report of the American Bar Association (48 Annu Rep ABA 560 (1925) at 595-6, accessed via Heinonline), which records the application for endorsement of the statute by the American Bar Association as part of the campaign for its enactment by the various States. The committee reported that:

24. [41]

    It is plain from the above that the mischief to which the reform was directed extended to the release of joint obligors. Something should be said of the injustice to which the committee referred.

25. [42]

    A joint obligation was regarded by the common law as a single cause of action. To use Glanville Williams’ example, if A and B jointly covenanted to pay £100 to C, there was a single promise, which each of A and B was liable to pay in full. That is quite distinct from several promises by each of A and B to pay £50 to C. Most significantly for present purposes, Glanville Williams explained in his classic work Joint Obligations (Butterworths, 1949) that:

26. [43]

    One of those “other rules” is the rule that an unqualified release of one joint promisee releases all joint promisees. The underlying theory was the “singleness” of the joint promise.

27. [44]

    Likewise, if two persons are joint tortfeasors (for example, the partners of a firm give negligent advice), then there is a single cause of action. It was stated in State of New South Wales v Loh Min Choo [2012] NSWCA 275 at [72] that:

28. [45]

    The difficulties associated with joint obligations led to the creation of joint and several obligations from a very early period. This was recognised as possible by the language in a deed or a bond, and in various United States jurisdictions statutes were enacted many years ago deeming all joint obligations to be joint and several or otherwise modifying the rule: many are summarised in S Williston, A Treatise on the Law of Contracts (rev ed 1936 by S Williston and G Thompson), vol II, §336. In days when there could only be a single cause of action in a writ, it was more important than it is now that there can be joinder of causes of action in the same proceeding (see UCPR r 6.18) and until such reform was made, the promisee with the benefit of a joint and several promise by A, B and C could sue A, B and C on the one writ, or each of them on a separate writ, but could not sue two of them on the one writ. See the anonymous 16th century case at 3 Dyer 310a; 73 ER 702 and the decisions there mentioned, and Glanville Williams, Joint Obligations, pp 60-61.

29. [46]

    Baron Parke’s analysis in King v Hoare (1844) 13 M & W 494; 153 ER 206 has proven to be influential. His Lordship explained that it had never squarely been decided whether a judgment recovered against one of two joint contractors was a bar to an action against the other. It was settled that a judgment obtained against a joint tortfeasor was a bar against an action against the other, on the basis that there was a single cause of action, which had merged into the earlier judgment. His Lordship considered that the same must be true where there were joint contractors. He then turned to joint and several contractual liability:

30. [47]

    Baron Parke then added:

31. [48]

    Those considerations led to the conclusion that where judgment had been obtained for a debt, no differently from the case of a tort, the right given by the record merged with the inferior remedy by action for the same debt or tort against another party.

32. [49]

    Glanville Williams wrote that King v Hoare was decided “in ignorance of earlier authority to the contrary”: at 94. Although doubted in Chancery by Knight-Bruce VC in Ex parte Waterfall (1851) 4 De G & SM 199 at 203; 64 ER 797 at 799, King v Hoare was upheld by the House of Lords in Kendall v Hamilton (1879) 4 App Cas 504. Glanville Williams regarded the analysis as unsatisfactory, stating that the analogy with tort could not provide strong support for the rule, because legislation had abrogated the rule in tort, while “[t]he theory of merger is pure technicality and should not be allowed to work injustice”: at 95-96.

33. [50]

    The reasoning in King v Hoare assumed heightened significance because, with a striking disregard for internal consistency, common law treated the release of a joint and several obligation as having the same effect as a release of a joint obligation. Once again, in Glanville Williams’ words:

34. [51]

    Even the ordinarily restrained authors of Halsbury’s Laws of England regarded this rule as “perhaps surprising”: 4th ed, vol 9 para 627. Glanville Williams explained that the source of the rule was a case of 1641 where no judgment was reported, after which “the rule became settled by a series of cases that followed each other uncritically, and prevails at the present day”.

35. [52]

    Likewise, Williston stated at §334 that not only were all joint debtors discharged by a release of one of them, but the same rule applied to the joint liability of joint and several debtors, and that it was also “early decided that the several as well as the joint liability of all the debtors was discharged”. With considerable understatement, Williston added “It is not easy to find a technically satisfactory reason for this latter rule”. Nonetheless, Williston wrote that “[t]he rule in the early English cases has been widely adopted”, noting what was said in In re EWA [1901] 2 KB 642 at 648: “It is too late now to question the law – that where the obligation is joint and several, the release of one of two joint debtors has the effect of releasing the others”. He noted that a commonly enacted provision to ameliorate the effect of the harshness of the common law rules about joint obligations, namely, deeming all joint obligations to be joint and several, could have no effect on a rule which applied indifferently to joint obligations and joint and several obligations.

36. [53]

    It is not necessary to trace the evolution of the rule for present purposes, because it was upheld in Walker v Bowry (1924) 35 CLR 48; [1924] HCA 28, decided at a time when the High Court was for most practical purposes bound by the House of Lords. At 56, Isaacs ACJ drew upon the justification given by Baron Parke:

37. [54]

    Starke J (at 57-58) was perhaps less enthusiastic as to the history and rationale of the rule, but accepted that it had been held that joint suretyship was the “essential condition of the liability” in such cases, even if the liability were joint and several. Irrespective of the persuasiveness of the explanation, there can be no doubt that this Court is bound (subject always to the effect of subsequent legislation) by what was held in Walker v Bowry. McLelland CJ in Eq proceeded on the same basis, despite considering the rule “impossible to justify” in principle, in Dorgal v Buckley (1996) 22 ACSR 164 at 167, as did this Court in Carr and Purves v Thomas [2009] NSWCA 208 at [15].

38. [55]

    The upshot is that for Australian purposes, it was accepted that the release of a joint and several liability was effective to release all of the obligees. This was the same position noted by the committee reporting to the American Bar Association in 1925 mentioned above.

39. [56]

    Williston’s Model Act was adopted in Nevada in 1927 and then, much more importantly, in New York in 1928: see the note by C Eberhart in 13 Cornell Law Quarterly (as the Cornell Law Review was then known) 640 (1928). A detailed analysis of the statute and the mischief to which it was directed may be found in Williston’s A Treatise on the Law of Contracts at §§336 and 336A. Former paragraphs 97(2)(a) and (d) of the Supreme Court Act 1970 derive broadly from ss 2 and 3 of Williston’s Model Act, which provide:

40. [57]

    Sections 4 and 5 dealt in terms with releases of joint, or several, or joint and several obligors. Those provisions were not picked up by the Law Reform Commission. No doubt an argument might be made based on the inference that the Law Reform Commission had consciously chosen not to recommend enacting those provisions of the Model Act concerning releases but confined the reforms to those based on the effect of a judgment. As noted above, no such submission was made in this Court, because the appellants did not direct themselves to the Model Act. We return to this below. It is sufficient for the moment to note that the submission would amount in substance to an application of the principles associated with the expressio unius maxim, upon which it is dangerous to rely: see for example Daniels Corporation International Pty Ltd v Australian Competition and Consumer Commission (2002) 213 CLR 543; [2002] HCA 49 at [34].

41. [58]

    The enactment of s 97 of the Supreme Court Act 1970 and s 95 of the Civil Procedure Act 2005 must be seen as an extension of earlier statutory reforms of the effect of a judgment upon a joint obligation. Both sections provide that they do not apply to a judgment to which s 5(1)(a) of the Law Reform (Miscellaneous Provisions) Act 1946 (NSW) applies. Thus, the provisions were on their face supplementary to earlier, more narrowly confined, legislative reform.

42. [59]

    Section 5(1)(a) of the Law Reform (Miscellaneous Provisions) Act 1946 (NSW) provides:

43. [60]

    Section 5(1)(c) goes on to provide that a tort-feasor may recover contribution between tort-feasors liable in respect of the same damage. This paragraph overturned the rule that contribution was not available either at law or in equity between tortfeasors: see Hunt & Hunt Lawyers v Mitchell Morgan Nominees Pty Ltd (2013) 247 CLR 613; [2013] HCA 10 at [12] and [80]. The fact that these paragraphs address different problems was noted in Newcrest Mining Ltd v Thornton (2012) 248 CLR 555; [2012] HCA 60 at [45] (referring to the materially identical Western Australian provision, s 7(1) of the Law Reform (Contributory Negligence and Tortfeasors’ Contribution) Act 1947 (WA)):

44. [61]

    Section 5(1)(a) has not been amended since it was enacted in 1946. It was not derived from the United States proposals of the 1920s. Instead, it was derived from the reforms in the United Kingdom in the 1930s concerning joint obligations in tort and the rule that there could be no contribution between tortfeasors, summarised in Nau v Kemp & Associates (2010) 77 NSWLR 687; [2010] NSWCA 164 at [34]-[35]:

45. [62]

    In short, a 1946 statute abrogated the rule that judgment against a joint tort-feasor would discharge another joint tortfeasor, and a separate statute in 1970 abrogated the rule that judgment against a joint obligee would discharge any other joint obligee, but only in circumstances where the 1946 legislation did not apply. The 1946 legislation conferred a right of contribution, which was necessary because contribution was not available at law or in equity between tort-feasors. The 1970 legislation did not address contribution, it being unnecessary to do so because there already were rights of contribution between joint obligees which would continue so long as the liability was not discharged.

46. [63]

    Although the reforming legislation commenced with tort and then expanded to include joint liability in areas other than tort, the modern reason given by the law for the effect of a judgment given in relation to one joint obligee having the effect of discharging all other joint obligees was the same: the cause of action against the joint tortfeasors or joint obligees was “one and indivisible”. Thus, in Duck v Mayeu [1892] 2 QB 511 at 513, A L Smith LJ said:

47. [64]

    This was the approach taken in the Restatement of Contracts. The passage from Duck v Mayeu was reproduced by Morton LJ delivering the leading judgment of the Court of Appeal in Apley Estates Co Ltd v De Bernales [1947] Ch 217 at 220-221, by Taylor J (with whom Kitto J agreed) in Ramsay v Pigram (1968) 118 CLR 271 at 286 who added that “as far as I can see has never been questioned” and by this Court’s unanimous judgment in Carr and Purves v Thomas [2009] NSWCA 208 at [14].

48. [65]

    Section 5(1)(a) of the Law Reform (Miscellaneous Provisions) Act 1946 plays an important part in the analysis in this appeal because of what has been authoritatively decided of its Australian Capital Territory counterpart in Thompson v Australian Capital Television Pty Ltd (1996) 186 CLR 574; [1996] HCA 38. The decision has received a mixed reception in New Zealand (see Robinson v Tait [2002] 2 NZLR 30; [2001] NZCA 217) but the reasoning binds this Court. There the High Court considered the effect of s 11(2) of the Law Reform (Miscellaneous Provisions) Act 1955 (ACT) which provided:

49. [66]

    The 1955 ACT legislation was directed to the same end, and was materially identically worded, as s 5(1)(a) of the 1946 NSW statute.

50. [67]

    The joint judgment of Brennan CJ, Dawson and Toohey JJ said at 582 that “The effect of s 11(2) is that the cause of action against joint tortfeasors is no longer one and indivisible”. After dealing with Singaporean, English and Australian authority on counterpart provisions, their Honours explained at 584:

51. [68]

    Gummow J, with whom in this respect Gaudron J agreed, wrote to the same effect at 613-614:

52. [69]

    Of course, s 5(1)(a) of the 1946 NSW statute and s 11(2) of the ACT legislation were confined in their operation to tortfeasors. For that reason, the High Court limited the implied abrogation effected by the legislation to the release of joint tort-feasors.

53. [70]

    But why should not the same reasoning apply to the more broadly applicable s 97 of the Supreme Court Act 1970 and s 95 of the Civil Procedure Act 2005 directed to joint obligations other than those arising in tort? The provisions have materially identical wording and are directed to the same end. Both are explicitly tied to the 1946 NSW statute insofar as their application is delimited by reference to the area of operation of that statute.

54. [71]

    A submission to this effect was made to the primary judge and was recorded at [2301] and [2303]:

55. [72]

    When Thompson v Australian Capital Television Pty Ltd was decided, s 97 of the Supreme Court Act 1970 provided:

56. [73]

    Each of paragraphs (2)(a), (b), (c) and (d) is inconsistent with a judgment effecting a release of other joint obligees. Applying the reasoning which was dispositive in Thompson produces the result that the provision must impliedly abrogate the concept that the non-tortious cause of action was one and indivisible. That is to say, s 97(2)(a) of the Supreme Court Act 1970 provides that where a judgment is given against one of a number of persons who have a joint liability, then the liability of the others is not discharged by the judgment or any step taken for its enforcement. That is to the same substantive effect as s 5(1)(a) of the 1946 NSW statute and is contrary to the proposition that there could be a single cause of action which merged in the judgment obtained against one of the persons jointly liable. If s 5(1)(a) impliedly abolishes the rule that the release of one joint tortfeasor releases the others, then so too must s 97(2)(a) in the cases to which it applies, namely, co-obligors who are not joint tortfeasors. Paragraphs (b), (c) and (d) of s 97(2) confirm that that is so.

57. [74]

    That reasoning was applied in Harplex Pty Ltd v Konstandellos (2018) 54 VR 174; [2018] VSCA 67 at [60]:

58. [75]

    Section 24AA of the Wrongs Act 1958 (Vic) provides that “Judgment recovered against any person liable in respect of any debt or damage shall not be a bar to an action, or to the continuance of an action, against any other person who is (apart from any such bar) jointly liable with the first-mentioned person in respect of the same debt or damage”. This section is different from s 95 of the Civil Procedure Act in its scope. Section 23A of the Wrongs Act provides that for the purposes of provisions including s 24AA: “a person is liable in respect of any damage if the person who suffered that damage ... is entitled to recover compensation from the first-mentioned person in respect of that damage whatever the legal basis of liability, whether tort, breach of contract, breach of trust or otherwise”. Thus, rather than having two separate provisions (s 5(1)(a) applicable to joint tortfeasors, and s 95 applicable to joint obligees who fall outside of s 5(1)(a)), the Victorian legislation has a single provision, extending to recovering compensation “whatever the legal basis of liability” and explicitly including “breach of trust”. The Victorian Court of Appeal found no difficulty in applying the construction of the narrower provision considered in Thompson to s 24AA, to achieve the implied abrogation of the single and indivisible status of causes of action for debt in the same way as had been held for causes of action for tort in Thompson. We respectfully agree.

59. [76]

    That conclusion is reinforced by the fact that the New South Wales Legislature chose to re-enact s 97 as s 95 of the Civil Procedure Act. This was a sensible measure to take, for there is no reason for the entry of judgment in the Supreme Court against a co-obligor to have any different effect from the entry of judgment by the District Court, and yet that would arguably be the effect of the provision as originally enacted (a point observed in Woodgate v Davis (2002) 55 NSWLR 222; [2002] NSWSC 616 at [10]). Similar concerns led to the delayed commencement of the Supreme Court Act and the enactment of the Law Reform (Law and Equity) Act 1972 (NSW).

60. [77]

    When s 97 was re-enacted as s 95, the words “in proceedings in the Court” were replaced by “in any proceedings”, a point to which we shall return at [86] below. But the presently relevant significance of re-enacting s 97 as s 95 of the Civil Procedure Act is that it occurred after the reasoning in Thompson as to the implied abrogation of a broader rule of law. Indeed, it had been noted in the standard commentary. Ritchie’s Supreme Court Procedure (loose-leaf, service 197, which dates from 2003) is in materially identical terms as the current annotation to s 95 which is reproduced above.

61. [78]

    As noted above, the appellants relied on the opening words “The section only refers to the effect of a judgment”. This was put as follows:

62. [79]

    We cannot accept that submission, which misses the gravamen of the paragraph in the commentary. The authors after giving the background to the rule add (after “However”) that the High Court’s decision in 1996 treated the ACT counterpart of the provision as having a wider effect, namely, that it “impliedly abrogated the rule that release of one joint tortfeasor released all”. The appellants’ submission was that “s 95 of the Civil Procedure Act has not impliedly abrogated the joint release rule but rather, only the merger into judgment rule”. Fairly read, the authors of Ritchie’s commentary were cautioning readers in terms that s 95 was to be construed in light of how the High Court had construed a materially identical provision.

63. [80]

    The rule that the release of one or a number of persons subject to a joint and several obligation releases all of them has been disfavoured in all common law jurisdictions for decades. It may be accepted that the technical rules relating to joint obligations and releases – classic instances of “lawyers’ law” – may not have been at the forefront of the subjective intentions of the parliamentarians voting in favour of s 95. This was part of the basis upon which Dixon CJ deprecated this principle of statutory construction as “quite artificial” in R v Reynhoudt (1962) 107 CLR 381 at 388; [1962] HCA 23. However, parliamentary intention is imputed, rather than subjective, and in any event, reasoning of this nature has in more recent times been less disfavoured. Thus, the joint judgment in Vella v Commissioner of Police (NSW) (2019) 269 CLR 219; [2019] HCA 38 at [19] said:

64. [81]

    Applying these principles, we would reason as follows. Section 5(1)(a) is mentioned in terms in s 95(3). It is well known that s 5(1)(a) abrogated the harsh common law rules in relation to joint tortfeasors. It is clear that s 95 had a similar reforming effect, but did not overlap with s 5(1)(a). It is natural to impute to the Legislature an intention to apply the settled construction of s 5(1)(a) to the broader field now covered by s 95.

65. [82]

    Another way of considering the effect of re-enacting s 97 of the Supreme Court Act as s 95 of the Civil Procedure Act is as follows. In circumstances where s 97 is explicitly linked to s 5(1)(a), and where the latter has been held impliedly to abrogate a disfavoured rule of common law, after which s 97 has been re-enacted in materially identical terms, this Court would introduce disharmony in the law if it held that the rule impliedly abrogated by s 5(1)(a) in the subject area to which it applies was not impliedly abrogated by s 95 in the wider subject area to which it applies.

66. [83]

    Mr Gino Cassaniti, who took the lead on this issue, submitted that Thompson was “distinguishable on account of the materially different statutory text”. He submitted that s 95 was materially different from s 5(1)(a) (and s 11(2) of the ACT statute) because s 95 “is drafted in terms of the specific pre-requisite requirement of an existing judgment (which is reinforced by the inclusive definition of that expression in s 3(1)) in ‘proceedings’ and is, as above, primarily if not exclusively directed to the question of merger”. We do not agree. Both s 95 and s 5(1)(a) have as their premise that there is a judgment against one of the persons jointly liable. It is true that the legislation may be said to be directed to the question of merger effected by the entry of judgment. But the reasoning in Thompson was as to the implied abrogation of the common law rule. It is not an answer to the same implication being given effect to the comparably worded s 95 to observe that it does not expressly deal with the indivisibility of causes of action involving joint obligations.

67. [84]

    Reliance was placed on the explanatory note accompanying the Civil Procedure Bill 2005. But that made it clear that the legislation was of a consolidating nature. Hence if useful recourse were to be had to extrinsic materials, it would be to those accompanying the provision as originally enacted, namely, s 97 of the Supreme Court Act; and what the Law Reform Commission wrote about the provision it proposed has already been addressed.

68. [85]

    True it is that the Law Reform Commission in 1969 recommended implementing only part of Williston’s Model Act and refrained from recommending the enactment of the provisions dealing explicitly with releases. But that does not undercut the force of the reasoning in Thompson as to the effect of the provisions dealing with judgment, nor does it detract from the force of the considerations mentioned above when s 97 was re-enacted as s 95, after Thompson had held that the similarly worded counterpart to s 5(1)(a) had the effect of abrogating the one and indivisible rule for tortious causes of action.

69. [86]

    Submissions were also made based on the inclusive definition of “liability” in s 95(4). That term as used in s 95 extends to “liability in contract, liability in tort and liability under a statute”. Liability in equity is conspicuous by its absence. An ingenious submission was advanced on behalf of the Borg parties to the effect that the provision could not alter the position in equity. It relied on the words “jointly liable for money” in par 25 of Appendix A of LRC 7 (reproduced above) and the word “judgment” in the section. This language was said to suggest the provision did not apply to equitable liability. Damages are foreign to equity, and “judgment” was said to connote common law, rather than a “decree” which, under the pre-Judicature regime, would issue on finalising a suit in equity. We do not accept this submission. Section 97 of the Supreme Court Act was part of the judicature reforms which put in place a common procedure throughout the Supreme Court, one (minor) aspect of which was that decrees in equity were replaced by judgments in all proceedings in the Court. Further, the submission cannot be reconciled with the generality of “in proceedings in the Court” in s 97, still less with “in any proceeding” in s 95.

70. [87]

    For those reasons we would conclude that these grounds should be dismissed on the basis that the primary judge was correct to hold that s 95 impliedly abrogated any rule that the appellants were released by the liquidator’s release of other persons such as Mr Frank Criniti.

71. [88]

    If there were a rule in equity, then for the reasons given above it did not survive the enactment of s 97 of the Supreme Court Act and s 95 of the Civil Procedure Act. But was there such a rule? By way of example, Crennan and Kiefel JJ referred in Newcrest Mining Limited v Thornton (2012) 248 CLR 555; [2012] HCA 60 at [55] to the effect of s 5(1)(a) as abrogating “the common law defence of ‘release by judgment’”. Did any of this apply in equity? There is reason to think that it did not. This may be seen in a number of ways.

72. [89]

    First, the modern starting point for the effect of releases upon joint liability at law is surely Glanville Williams’ work Joint Obligations. The work is justly famed as a work of scholarship. It is directed to the very point in these appeals.

73. [90]

    Joint Obligations is a work dealing with the common law. The preface commences “This book discusses a difficult and seriously defective part of the common law” and urges legislative reform. Nonetheless, it addresses the liability of trustees in two places. Glanville Williams states in the outline at p 31:

74. [91]

    Glanville Williams also writes at p 159:

75. [92]

    Footnote 5 provides:

76. [93]

    It is tolerably clear that, as Glanville Williams states, trustees and third parties involved in a breach through being knowing assistants or knowing recipients may be jointly and severally liable to restore misdirected trust funds. See D Hayton et al, Underhill and Hayton, Law Relating to Trusts and Trustees (19th ed, LexisNexis 2016), p 1267. But it does not follow that the release of one effects a release of the rest, although as much is from time to time asserted. For example, it is said:

77. [94]

    But that presupposes that a release of a person who is jointly and severally liable in equity has the same consequences as the release of a person who is jointly and severally liable at law.

78. [95]

    Blackwood v Borrowes, the decision cited by Glanville Williams and the 9th edition of Underhill, is a decision of considerable persuasive weight. It is a decision of Sir Edward Sugden (as Lord St Leonards then was), sitting as Lord Chancellor of Ireland. The facts were complicated, but the relevant point amounted to this. One trustee had been released by the beneficiary. The other (Colonel Leslie) had died, and the pertinent part of the argument proceeded on the basis that the trustees were both liable. Counsel for the estate of the deceased trustee submitted “even assuming that a breach of trust was committed, the release to one of the trustees, Blackwood, executed with full knowledge of the circumstances, operates clearly to the release of the others” (at 466).

79. [96]

    The Lord Chancellor did not accept that argument. He said when dealing with this aspect of the case (at 475-476):

80. [97]

    We think that Glanville Williams and Underhill were correct to regard that decision as authority for the proposition that a general release of one trustee did not release the other, where both were liable for the same loss to the trust estate. Certainly, the Lord Chancellor chose not to apply the proposition which was advanced by the deceased trustee’s estate. The decision to leave for determination on another occasion the effect of a release which purported to leave in place the liability of the other trustee is inconsistent with the existence of a general rule equivalent to the release of a joint obligation in the law of contract. It is one thing for a beneficiary to release one of a number of trustees, and another thing entirely for the beneficiary to adopt, or ratify, or accept the conduct which was a breach of trust. It seems plain enough that a beneficiary who had accepted, and taken the benefit of, conduct which is in breach of fiduciary duty cannot thereafter sue an accessory for knowing assistance in the breach. That is not because of the release, but because the plaintiff would be approbating and reprobating, or adopting inconsistent stances in relation to the same wrongful conduct.

81. [98]

    A second area where equity regarded a release differently from common law may be seen in partnership debts. In Blyth v Fladgate [1891] 1 Ch 337 at 353, Stirling J rejected the submission that a suit for breach of trust against one partner, which had resulted in a judgment against him, amounted to a discharge of the liability of the other partners. Stirling J said:

82. [99]

    More recently, Barrett J drew attention to the different approach in equity to partnership debts in Woodgate v Davis at [10]:

83. [100]

    His Honour added at [12] reference to Dr Fletcher’s suggestion at p 173 of the eighth edition (2001) of Higgins and Fletcher, The Law of Partnership in Australia and New Zealand, that a partner’s liability for partnership debts is within “a rule which falls between ‘joint’ and ‘joint and several’ liability”.

84. [101]

    Thirdly, equity has long treated this area of the law differently from common law. Some differences are substantive. Consider the statutory discharge or release of a joint obligor effected by bankruptcy. Prior to the judicature legislation, both common law and equity permitted contribution between co-obligors, but in equity, the burden was divided only between those who were solvent: Lowe & Sons v Dixon & Sons (1885) 16 QBD 455 at 458. Such a difference suggests caution before concluding that a common law rule about releases applied in equity.

85. [102]

    Some differences reflected the range of remedies available in equity. The clearest position is where a fiduciary and an accessory each make profits from a breach of trust. There is no reason for the profits for which the fiduciary must account to be equal to the profits for which the accessory must account. Alternatively, the beneficiary may choose to require the fiduciary to account for the property misapplied, and for the accessory to account for profits the accessory has derived. Those are examples of what the High Court contemplated when it said in Michael Wilson & Partners Ltd v Nicholls (2011) 244 CLR 427; [2011] HCA 48 at [106] that:

86. [103]

    In Grimaldi v Chameleon Mining NL (No 2) at [557] the Full Court said that:

87. [104]

    One case where a similar issue arose is United States Surgical Corporation v Hospital Products International [1982] 2 NSWLR 766 at 817-818, where the fiduciary HPI was under “the complete control and domination” of Mr Blackman. HPI sought to rely upon a release which had been given to Mr Blackman. It was held at 818 that the release did not as a matter of construction extend to the liability arising out of the activities giving rise to the litigation, but before reaching that conclusion, McLelland J said at 817c:

88. [105]

    His Honour also made it clear at 817f that, depending upon the way the account should proceed in relation to some of the artificial mark up (by which profits were derived by trusts controlled by him), “it may be that Mr Blackman should be held separately liable to account for any personal benefits gained through Morust Ltd or the Bellevue Trust”. This suggests a rather more nuanced approach to liability than at common law.

89. [106]

    However, although those authorities favour the conclusion that the approach in equity diverged from that at common law, it is not difficult to point to statements in the authorities and commentary to the opposite effect. These seem to derive from early editions of Lewin. For example, the 4th edition of 1861 states at p 598:

90. [107]

    It seems probable that the very similar statement in J Perry, A Treatise on the Law of Trusts and Trustees (Cambridge Mass, 1871), p 776 derives from this:

91. [108]

    Similarly, there is a very full account of releases in G Sweet, Bythewood’s and Jarman’s System of Conveyancing (3rd ed 1844). Volume IX contains some 120 pages of analysis on releases at pp 699-821, followed by 60 pages of precedents. For the most part, the work is directed to the position at law. However, the work also states (at p 812) that “A release of breaches of trust to one trustee will generally ensure to the benefit of his co-trustee”, citing Thompson v Harrison and Blackwood v Borrowes.

92. [109]

    Thompson v Harrison (1787) 2 Bro CC 164; 29 ER 94 and Blackwood v Borrowes were cited by both Lewin and Jarman. Blackwood v Borrowes in fact rejects the proposition, which is in substance equivalent to the “technical ground” upon which Sir Edward Sugden did not rely. Thompson v Harrison is a poorly reported dismissal of a bill against a brother who was released prior to trial, and examined by the plaintiff. The entirety of Lord Chancellor Thurlow’s reasons, as reported by Brown, is below:

93. [110]

    But that turns upon the old rule requiring joinder of all persons against whom the plaintiff had a joint and several demand. Lord Hardwicke had said in Madox v Jackson (1746) 3 Atk 405 at 406; 26 ER 1034 that “[t]he general rule of the Court is: where a debt is joint and several, the plaintiff must bring each of the debtors before the Court, because they are intitled to the assistance of each other in taking the account. Another reason is, that the debtors are intitled to a contribution, where one pays more than his share of the debt” (citation omitted). This was altered by the General Orders promulgated in August 1841, which provided:

94. [111]

    The extensive commentary at pp 234-237 in the 5th ed 1871 of Daniell’s Chancery makes it clear that the position was more nuanced, and sometimes the rule did not excuse the non-joinder. But all that matters for present purposes is that the reasoning in Thompson v Harrison turned on the necessity to join both defendants, and thus, caution is required in applying it after the abolition of that rule.

95. [112]

    Much more recently, in Edgewater Homes Pty Ltd v Donohoe [2019] NSWSC 44, where it was common ground that the liability of a fiduciary and a knowing recipient or a knowing assistant was joint and several (see at [23]), and where no reliance seems to have been placed upon s 95 of the Civil Procedure Act 2005 (NSW), it was held that the release of a knowing assistant had the effect of releasing the fiduciary. Reliance was placed on academic writing, including the 19th edition of Lewin on Trusts which was based on Thompson v Harrison, and upon what Bryson JA had said in Yeshiva Properties No 1 Pty Ltd v Marshall reproduced above.

96. [113]

    The current edition of Lewin (20th ed, Sweet & Maxwell, 2020) vol II, paragraph 41-110 states:

97. [114]

    The current edition cites Thompson v Harrison, Blackwood v Burrowes and notes that the paragraph was cited in Edgewater Homes Pty Ltd v Donohoe.

98. [115]

    It is helpful to return to principle. Upon the assumption that Ms Cassaniti would otherwise be liable to Rackforce and Diamondwish through knowingly participating in the breaches of fiduciary duty by its director Mr Criniti, is the effect of the companies’ liquidator’s release of him that she too is released? It seems improbable that the answer to that question will turn on the abstract question whether Ms Cassaniti’s liability and Mr Criniti’s liability is joint, or several, or joint and several, so as to engage an unsatisfactory rule whose origins lay at common law and which has at least in part been abrogated by statute. And it is far from clear that the liability of Mr Criniti and Ms Cassaniti will necessarily be regarded as either “joint” or “joint and several” or “several”.

99. [116]

    It is not difficult to see that the position in equity may be much less black and white than the all-or-nothing approach favoured by common law:

    1. (1)

       Relief in equity may be available on terms, and there is no reason why the terms would inexorably be the same as between making the fiduciary accountable and making the knowing assistant accountable.

    2. (2)

       There may be defences to which one but not all of the defendants may avail themselves. Consider for example a third-party professional adviser who procured an innocent trustee to make an unauthorised investment. The trustee may have a powerful defence of acting honestly and reasonably so as to engage s 85 of the Trustee Act 1925 (NSW) while the professional may be plainly liable for procuring the breach of trust. Why should the release of the trustee discharge the person who procured the breach of trust?

    3. (3)

       Further, a release is a consensual discharge, but equity has long regarded a discharge by bankruptcy differently (hence, contribution is not available as against a bankrupt or insolvent co-obligor).

    4. (4)

       Further, equity may give a plaintiff an election between requiring an account and restoring trust property. A plaintiff may require trust property to be restored, and also require a knowing participant to account for profits derived from the wrongful taking of trust property. One may without much difficulty contemplate a case where the trustee and the participant are required to account for the restoration of the trust property, while the participant may also be accountable for the profits derived from its use. It would seem passing strange that a release to the trustee would release the participant from being accountable for the loss of property, but leave in place liability to account for profits.

100. [117]

     A more general difficulty is this. A fiduciary or a third party involved in the breach may be called upon to account. That is quite clear in the case of an account of profits. It is fairly clear when the fiduciary or third party is required to account as a constructive trustee for property received. This was explained by Lord Millett in Dubai Aluminium Co Ltd v Salaam [2003] 2 AC 366; [2002] UKHL 48 at [141], who said that a person liable pursuant to either limb of Barnes v Addy is “not in fact a trustee at all, even though he may be liable to account as if he were”. The better view is that what is called “equitable compensation” is a shorthand for requiring a fiduciary to account; see M Conaglen, "Equitable Compensation for Breach of Trust: Off Target" (2016) 40 Melbourne University Law Review 126 esp at 146-150. This is quite different from the liability to pay common law damages.

101. [118]

     The views of Bryson JA on a point of this nature are of considerable persuasive force. The appellants placed weight on what his Honour said in Yeshiva reproduced above. His Honour’s concern was principally directed to the unfairness which could result by a plaintiff not joining a person who was liable in equity (with the result, for example, that the person was not liable to give discovery). This is readily addressed by the person being joined on a cross-claim. Further, the view that it is necessary to join both the fiduciary and the third parties involved in the breach will often be impracticable (the nature of such cases is that the fiduciary may have died, or have left the jurisdiction, or have been wound up) and in any event, is irreconcilable with the modern equivalent of the rules as to joinder, UCPR r 6.21(1) (“A person who is jointly and severally liable with some other person in relation to any act, matter or thing need not be a defendant in proceedings with respect to that act, matter or thing merely because the other person is a defendant in those proceedings”).

102. [119]

     There is a passage in this Court’s judgment in Lavin v Toppi (2014) 87 NSWLR 159; [2014] NSWCA 160 at [61] to the effect that there was no counterpart in New South Wales to s 3 of the Civil Liability (Contribution) Act 1978 (UK) or s 24AA of the Wrongs Act 1958 (Vic). Lavin v Toppi was a case about contribution, and some care has to be taken because United Kingdom and Victorian legislation extends statutory contribution to persons who are not joint tortfeasors. Lavin v Toppi is not authority for the effect of s 95 of the Civil Procedure Act on the release rule. A general statement, in circumstances where it is plain that there was no argument on the point, and where the Court was not directed to s 95, has no precedential value as to the effect of s 95. The point was correctly described by the Victorian Court of Appeal as “a passing reference on an issue which did not need to be decided in that case”: Harplex Pty Ltd v Konstandellos (2018) 54 VR 174; [2018] VSCA 67 at [51].

103. [120]

     It is idle to pretend that the decisions and commentators which have considered the effect of a release upon a joint or joint and several obligation in equity speak with the one voice. Hence the statement in Grimaldi at [553] that there are “subsisting uncertainties as to whether and/or when the liabilities of the knowing assistant or recipient are only several, or are joint and several, with those of the delinquent fiduciary or trustee”. However, it is fair to say that the most thoughtful analyses – those by Glanville Williams, by Sir Edward Sugden, by McLelland J – tend to favour the view which we would reach at the level of principle, which is that there is no reason for an unjust and disfavoured rule at common law to be followed in equity, where liability is conceptually different, is subject to discretionary defences and is apt to be measured differently as between defendants. In light of s 95 of the Civil Procedure Act the question is academic, but we nonetheless favour the view that the “release rule” whereby the release of one person under a joint or joint and several obligation releases all others did not apply to cases of fiduciaries and third parties who knowingly participate in dishonest and fraudulent breaches of fiduciary duty.

104. [121]

     For those further reasons, these grounds must be dismissed. The Borg parties’ notice of contention must also be dismissed.

105. [122]

     The appeals by Gino Cassaniti and the Khalil parties (except George Said) must be dismissed.

The appeals by George Said

1. [123]

   Involved Recruitment Pty Ltd (Involved Recruitment) was incorporated on 16 May 2012 with George Said, George Khalil and Gino Cassaniti as its directors. Although her Honour observed that George Said and George Khalil both purportedly ceased as directors on that same day (at [292]), George Said’s evidence was that he understood that he was a director of that company throughout its time of operation. The registered office and principal place of business of Involved Recruitment was at the offices of Banq, where Gino Cassaniti, George Said’s cousin, Fred Khalil, his brother, George Khalil, and their brother-in-law, Peter Abboud, were principals or persons employed by that firm: at [152]-[153]. Banq was also its tax agent: at [197]. Involved Recruitment was wound up in a creditors’ voluntary winding up on 28 June 2013 and Mr Ball was appointed as liquidator: at [197].

2. [124]

   The liquidator’s pleaded claims relevantly alleged that George Said was a director of Involved Recruitment appointed and instructed by Gino Cassaniti, that RCG and Bluemine’s director, Gino Cassaniti, caused those companies to enter into a series of transactions with Involved Recruitment, and in carrying out Gino Cassaniti’s instructions, George Said delivered large sums of cash to Gino Cassaniti from the bank account of Involved Recruitment and caused Involved Recruitment to:

3. [125]

   It was alleged that in carrying out Gino Cassaniti’s instructions, George Said knew that he was assisting Gino Cassaniti in his breaches of duty and that what Gino Cassaniti was doing was dishonest, or that he had knowledge of circumstances that would indicate to an honest and reasonable person that Gino Cassaniti was acting dishonestly and fraudulently in instructing him to deliver cash to Gino Cassaniti, transfer $934,399 to RCG and receive $55,590 from Bluemine.

4. [126]

   There is no challenge to the primary judge’s findings that:

5. [127]

   In upholding the liquidator’s claim, the primary judge found that George Said’s knowledge of Gino Cassaniti’s breaches of duty was such that he was liable in equity and under statute for his assistance and involvement in those breaches.

6. [128]

   In proceedings 2021/242648 and 2021/242667, George Said appeals against the finding that he had the requisite knowledge of Gino Cassaniti’s breaches of duty. He says that this finding should be set aside because her Honour wrongly applied a test of constructive knowledge in upholding the claim in equity. It is common ground that it is not necessary to consider the liquidator’s claim under statute; it was accepted that his claim under statute rose no higher than the claim in equity.

7. [129]

   The liquidator seeks to uphold her Honour’s finding against George Said on the basis that the finding should not be read as a finding of constructive knowledge. Alternatively, the liquidator relied upon a notice of contention in each appeal that if her Honour erred in determining George Said’s liability by reference to the fifth category in Baden v Société Générale pour Favoriser le Développement du Commerce et de l’Industrie en France SA [1993] 1 WLR 509 at 575-576, then her Honour should have determined that the conduct of George Said amounted to assistance of Gino Cassaniti’s breaches of fiduciary duty within categories one to four in Baden.

8. [130]

   There is no dispute as to the applicable legal principles for accessorial liability claims in equity. Farah Constructions Pty Ltd v Say-Dee Pty Ltd (2003) 230 CLR 89; [2007] HCA 22 establishes (at [174]-[179]), accepting the view of Stephen J (Barwick CJ agreeing) in Consul Development Pty Ltd v DPC Estates Pty Ltd (1975) 132 CLR 373 at 412; [1975] HCA 8, that knowledge within the categories which were later identified as (i) to (iv) of Baden is sufficient for the purposes of the second limb in Barnes v Addy (1874) LR 9 Ch App 244, but constructive knowledge within category (v) of Baden is insufficient, that is, knowledge of circumstances which would put an honest and reasonable person on inquiry.

9. [131]

   Thus, for the purposes of the liquidator’s second limb claim against George Said, the relevant categories of knowledge are: (i) actual knowledge; (ii) wilfully shutting one's eyes to the obvious; (iii) wilfully and recklessly failing to make such inquiries as an honest and reasonable person would make; and (iv) knowledge of circumstances which would indicate the facts to an honest and reasonable person.

10. [132]

    It is convenient first to address the evidence concerning George Said’s state of mind in the period September 2012 to May 2013.

11. [133]

    At trial, George Said gave evidence under compulsion of a subpoena issued by the liquidator. He was examined by senior counsel for the liquidator and, in part, cross-examined with leave granted over the objection of his counsel.

12. [134]

    Her Honour addressed this evidence in some detail at [824]-[856]. As to the circumstances of his appointment as a director of Involved Recruitment, George Said became a director in 2012 at the request of Gino Cassaniti, who asked him if he wanted to “partake” in a business and told him that it might be a bit of administrative work and he could work via his laptop. At the time, George Said was working fulltime as a real estate agent. A couple of weeks later, George Said began to do some work at the direction of Gino Cassaniti: at [825]. George Said described this work as “just pretty much paying whoever they asked via email. Whoever Gino asked via email”. He said that he would receive an email with a timesheet to invoice and he would then get the payment and would pay whoever it needed to be directed to via email: at [827].

13. [135]

    In performing this role, George Said had access to the bank account of Involved Recruitment, although according to his evidence, he was not sure if he was a signatory on that account. He had the bank account login details and was able to do (internet) transfers on that account. He was “sure” that Gino Cassaniti had access to the bank account because sometimes he logged in and saw that there were other transactions that he had not done: at [828]. Her Honour summarised the effect of George Said’s evidence at [829]:

14. [136]

    As to the cash withdrawals, George Said’s evidence was that he would go to the bank and withdraw money from the account of Involved Recruitment. He said that sometimes Gino Cassaniti would tell him to take money from that account or give him cash for his work. He accepted that he would take out large sums and give the money to Gino Cassaniti: at [839].

15. [137]

    In cross-examination, George Said was taken to his out-of-court statements made during his interview with officers of the Australian Taxation Office on 3 June 2019. Portions of the transcript of the ATO interview were provisionally admitted as Exhibit U, subject to objection. In a subsequent ruling given by her Honour on 26 March 2020, identified portions of the transcript in Exhibit U were admitted against George Said as admissions against interest, or as evidence of prior inconsistent statements, and the balance of Exhibit U was admitted only as evidence of what was put to the witness in examination or cross-examination, as the case may be: In the matters of Earth Civil Australia Pty Ltd v RCG CBD Pty Ltd, Bluemine Pty Ltd, Diamondwish Pty Ltd and Rackforce Pty Ltd (All in liq) [2020] NSWSC 293.

16. [138]

    There is no challenge to these evidentiary rulings. To the extent that both parties referred on appeal to portions of Exhibit U that were not admitted into evidence against George Said, those submissions were withdrawn after the Court drew the parties’ attention to the evidentiary rulings of the primary judge.

17. [139]

    The statements made by George Said during his ATO interview which were admitted as admissions against interest include those which are extracted below. The first concerned the circumstances in which he was asked by Gino Cassaniti to provide assistance in relation to invoicing and making payments:

18. [140]

    As her Honour noted at [859], George Said acknowledged in cross-examination that Gino Cassaniti was the person he was referring to at Banq, who he did not want to identify to the ATO officers, as appears from the following extract:

19. [141]

    As to his role in preparing invoices and making payments as instructed, George Said described his understanding of what he was doing in these terms:

20. [142]

    As to his role in the withdrawal of cash from the bank account of Involved Recruitment and cash payments to himself, George Said described what occurred as follows:

21. [143]

    George Said gave inconsistent evidence at trial concerning his role in issuing invoices for Involved Recruitment. He initially agreed that his evidence in his ATO interview was truthful when he said that he changed the letterhead for invoices for Involved Recruitment. He said that Gino Cassaniti instructed him on how to use the laptop to enable the change on the letterheads to be made, and someone at Banq’s office programmed his laptop for this work, that he left his laptop in Gino Cassaniti’s office for it to be set up, and Gino Cassaniti also instructed him as to the withdrawal of money to take for himself and to give to Gino Cassaniti: at [830].

22. [144]

    He later claimed in his oral evidence that his previous answers during the ATO interview about changing the logo (on invoices) was incorrect. As her Honour recorded at [831]:

23. [145]

    Nonetheless, her Honour found that there was nothing unfair in the nature of the questioning in the ATO interview on 3 June 2019, as George Said had inferred in his oral evidence: at [833].

24. [146]

    He gave further evidence on the voir dire (which her Honour recorded at [834]-[842]), as to his involvement in changing the letterhead or logos on invoices, which was later admitted without qualification: at [856]. This evidence was that he would put the “Involved Recruitment” invoice header or logo on the invoices, send the invoices over to whomever was requested to pay it, and, if it was a timesheet, he would normally have instructions (from the person who sent the timesheet or from Gino Cassaniti) as to the person to whom to send the invoice: at [836]. As her Honour noted, this evidence was inconsistent with George Said’s evidence that he was not changing logos (on invoices) and that his earlier evidence to the ATO to that effect was incorrect and that maybe he did not understand what they had been asking him: at [837].

25. [147]

    Her Honour found that George Said’s primary role in relation to Involved Recruitment was to undertake minor clerical work in the preparation of invoices, which, on his evidence, was at the direction of Gino Cassaniti, doing this on his laptop from electronic instructions received by email and using software or templates which, on his evidence, were provided to him by Gino Cassaniti; that George Said’s evidence as to whether this involved changing letterheads on invoices was more than a little confused; that George Said withdrew money from the bank which, on his evidence, was at the direction of Gino Cassaniti and for the payment of employees, and that he also withdrew large sums and gave them to Gino Cassaniti: at [857]. Her Honour observed that it was abundantly clear that George Said never made any decision himself (as to how much was to be invoiced or paid), and that he did what he was instructed to do when issuing invoices: at [858].

26. [148]

    Addressing George Said’s credit, her Honour found that his evidence must be approached with some caution given the inconsistency between his ATO interview and his evidence in Court and that his identification of Gino Cassaniti as the person whom he was reluctant to identify to the ATO officers (during the ATO interview) was “telling”: at [859].

27. [149]

    Her Honour’s dispositive reasons for finding that George Said’s involvement in the scheme rendered him liable in equity for knowing assistance are contained at [1655]-[1656]:

28. [150]

    Farah establishes that it is sufficient for George Said to be shown to have actual knowledge of Gino Cassaniti’s breach, or to have wilfully shut his eyes to the obvious, or to have wilfully and recklessly failed to make such inquiries as an honest and reasonable person would make, or knowledge of circumstances which would indicate the facts to an honest and reasonable person.

29. [151]

    Although her Honour correctly stated the test of knowledge at [987], she did not apply that test in terms. George Said submitted that her Honour erred by applying a test of constructive knowledge, pointing to the italicised sentence in [1655] which is extracted at [149] above. The liquidator said that the italicised sentence is no more than one factual finding leading to the ultimate finding of accessorial liability in the last sentence of [1655], which is to be read as referring to the second, third or fourth categories in Baden.

30. [152]

    It is arguable that when [1655] is read as a whole, her Honour should be taken to have found that although the issuing of invoices viewed alone might not have been sufficient to amount to knowledge such that George Said was liable in equity and under statute for Gino Cassaniti’s breaches of duty, viewed in the context of the withdrawal of large sums of cash and the arbitrary amounts Gino Cassaniti instructed George Said to take for himself, George Said’s conduct in issuing the invoices, making the payments and withdrawing large sums and delivering them to Gino Cassaniti was sufficient to find that his knowledge was such that he was liable in equity and under statute.

31. [153]

    The difficulty with reading the judgment this way is the absence of an express finding that George Said’s knowledge of Gino Cassaniti’s breaches of duty answered one of categories (i) to (iv) in Baden, which the High Court accepted in Farah is sufficient. Given the forensic significance of the subject matter of the constructive knowledge in the italicised sentence in [1655] (the withdrawal of large sums of cash), we are inclined to the view that the finding of liability against George Said was affected by error.

32. [154]

    Accepting that error has been established, it is necessary to address the liquidator’s notice of contention in each appeal seeking a finding of knowledge within categories (i) to (iv) in Baden. It was not submitted on behalf of George Said that this Court was not in a position on a rehearing to determine that issue on the evidence and unchallenged findings below, including the adverse credit finding.

33. [155]

    It is convenient to address the issue of knowledge by reference to the fourth category in Baden: knowledge of circumstances which would indicate the facts to an honest and reasonable person. It is to be kept in mind that the “morally obtuse” cannot escape by failing to recognise an impropriety that would have been apparent to an ordinary person applying the standards of such persons: Farah at [178].

34. [156]

    George Said submitted that “neither the findings made, nor the evidence of his actions relied upon in support of the [liquidator’s] notice of contention, rise above putting an honest person on their enquiry” (supplementary written submissions, par 3). In oral argument, counsel said that whilst George Said may have been negligent or careless as a director or deemed director of Involved Recruitment, including in assisting Gino Cassiniti in his breaches of duty, he did not have knowledge of the circumstances that would indicate the facts to an honest and reasonable person.

35. [157]

    These submissions were not particularly helpful as they were expressed in conclusionary terms and, with one exception relating to George Said’s issuing invoices, the submissions did not engage with the evidence and unchallenged findings below.

36. [158]

    Addressing George Said’s state of mind, his counsel submitted that:

    1. (1)

       George Said was only performing clerical work at the direction of Gino Cassaniti and there was nothing suspect with preparing invoices from timesheets for labour hire in a labour hire business;

    2. (2)

       changing the letterhead from that of a subcontractor to the letterhead of Involved Recruitment before being sent out to a “client”, particularly on instructions, is unlikely to have raised any serious concerns; and

    3. (3)

       the liquidator’s evidentiary case, taken at its highest, established no more than that “[George] Said’s preparation of the invoices should have put him on his enquiry” (supplementary written submissions, pars 7-10).

37. [159]

    The premise of these submissions is that Involved Recruitment was engaged in a labour hire business supplying labour to its “clients”, which had been provided by subcontractors who invoiced Involved Recruitment, which in turn invoiced its clients. However, there is no finding by her Honour to this effect and this premise is not otherwise established on the evidence.

38. [160]

    Insofar as the submission should be understood as suggesting that George Said honestly believed that the payments by Involved Recruitment to RCG and from Bluemine, and the withdrawal of large sums of cash, were for a genuine commercial purpose, there are several difficulties with the submission.

39. [161]

    First, as her Honour found, George Said was unable to give a cogent and consistent explanation of the events in which he was involved: at [859]. Second, consistent with him not having an honest belief that the invoices he was preparing were genuine, George Said acknowledged in his ATO interview that he did not check the details on the invoices he received before changing the letterhead to Involved Recruitment and he never paid attention to what the invoices were for. Third, George Said did not give evidence that he held any belief at the time that the payments to RCG, the cash withdrawals he made to Gino Cassaniti, or the payment RCG received from Bluemine were for any genuine commercial purpose.

40. [162]

    In our view, the facts as known to George Said would have indicated to an honest and reasonable person that the invoices he was creating were contrived and that the payments and cash withdrawals he was making to RCG and Gino Cassaniti lacked any genuine commercial purpose. That was in circumstances where it can be inferred that George Said knew that the amounts of cash which Gino Cassaniti instructed him to take for himself were arbitrary; and he knew that he had no arrangement with Involved Recruitment for the payment of director’s fees, or salary or wages. As George Said acknowledged in cross-examination, the cash payments which he received were “never a system”.

41. [163]

    It is not in dispute that George Said’s conduct assisted Gino Cassaniti in his breaches of duty; on his evidence he was doing what Gino Cassaniti asked him to do. George Said accepted, despite the ASIC records, that he regarded himself to be a director of Involved Recruitment: at [1635]. Whilst he was content to take instructions regarding that company primarily, or entirely, from Gino Cassaniti alone (at [1636]), he knew from those instructions and the bank statements for Involved Recruitment which he received addressed to “The Director” at his home address in Belmore over a 21-month period (at [1639]), that Gino Cassaniti controlled the payments from Involved Recruitment to RCG, including the large sums of cash which he withdrew.

42. [164]

    We conclude that in the context of withdrawing large sums of cash and delivering those amounts to Gino Cassaniti for no genuine commercial purpose, and also being instructed to take arbitrary amounts of cash for himself, the facts as known to George Said would have indicated to an honest and reasonable person the impropriety of Gino Cassaniti instructing him to issue invoices for which there was no genuine commercial transaction, and causing Involved Recruitment to pay $934,399 to RCG by 80 transactions and receive in one transaction $55,590 from Bluemine and to withdraw large sums of cash and deliver those amounts to Gino Cassaniti. George Said’s knowledge of Gino Cassaniti’s impropriety was such that he was liable in equity and under statute for Gino Cassaniti’s breaches of duty.

43. [165]

    For these reasons, the finding of liability against George Said should be upheld.

Conclusion and Orders

1. [166]

   The appeals by Gino Cassaniti and the specified Khalil parties against RCG, Bluemine, Diamondwish and Rackforce and ground 1 of the notice of contention by the Borg parties in the liquidator’s appeal have failed on the common issue that was pressed at the hearing. (The balance of the issues raised by the liquidator’s appeal against the Borg parties is the subject of the separate judgment referred to at [4] above.) The appeals by Gino Cassaniti and the Khalil parties (except George Said) against Earth Civil were not pressed. All the appeals will be dismissed. There is no reason why costs should not follow the event: UCPR r 42.1.

2. [167]

   The appeals by George Said against RCG and Bluemine have failed. These appeals will also be dismissed with costs.

3. [168]

   The liquidator’s notices of motion filed 7 April 2022 seeking dismissal of the appeals by the Khalil parties on the ground that the appeals were incompetent as being out of time, should be dismissed. However, the Khalil parties should pay the respondents’ costs of those motions, given that the notice of appeal in each matter was filed out of time without leave.

4. [169]

   The Court makes the following orders in each appeal:

   1. (1)

      Appeal dismissed.

   2. (2)

      The appellant to pay the respondents’ costs in this Court.

5. [170]

   The Court makes the following orders in each appeal:

   1. (1)

      Extend time for the filing of the notice of appeal to the date on which the document was filed.

   2. (2)

      Appeal dismissed.

   3. (3)

      Dismiss the respondents’ notice of motion filed 7 April 2022.

   4. (4)

      The appellants to pay the respondents’ costs in this Court, including the costs of the respondents’ notice of motion filed 7 April 2022.

6. [171]

   In the liquidator's appeal on behalf of RCG against the Borg parties, dismiss ground 1 of the notice of contention.