Commonwealth of Australia v Sanofi

Date of hearing: 16-19, 22-24 February 2021

Counsel for the Appellant: Mr J Gleeson SC, Dr B Kremer and Ms F Roughley

Solicitor for the Appellant: Corrs Chambers Westgarth

Counsel for the Respondents: Mr J C Sheahan QC, Mr J J Hutton and Mr S Fitzpatrick

BETWEEN: COMMONWEALTH OF AUSTRALIA Appellant

AND: SANOFI (FORMERLY SANOFI-AVENTIS) First Respondent

BRISTOL-MYERS SQUIBB INVESTCO LLC Third Respondent

order made by: BESANKO, PERRAM AND YATES JJ DATE OF ORDER: 26 June 2023

1. The appeal be dismissed.
2. The Appellant pay the costs of the Respondent to the appeal as taxed, assessed or agreed. Note: Entry of orders is dealt with in Rule 39.32 of the Federal Court Rules 2011.

INTRODUCTION 1 Clopidogrel is a medication which inhibits the formation of blood clots and is usually prescribed to those who have suffered, or are at risk of suffering, a heart attack or stroke. Material before the trial judge indicated that revenue from worldwide annual sales of clopidogrel exceeded US$1 billion. In Australia at the times relevant to this litigation, clopidogrel was supplied in tablet form under the brand name 'Plavix' by Sanofi Australia Pty Ltd and under the name 'Iscover' by Bristol-Myers Squibb Australia Pty Ltd ('BMS'). As the parties did in their submissions, we will refer to the Respondents to this appeal together as 'Sanofi'. Clopidogrel is the non-proprietary name for the dextro-rotatory enantiomer of methyl alpha-5 (4,5,6,7-tetrahyrdo (3,2-c) thieno pyridyl) (2-chlorophenyl)-acetate. 2 Both Plavix and Iscover are listed on the Pharmaceutical Benefits Scheme ('PBS') and have been since 1999. The usual effect of being listed on the PBS is that a pharmacist who sells the drug receives a substantial subsidy from the Commonwealth. The listing on the PBS of a generic version of a drug results in a variety of reductions in the extent of that subsidy, including an automatic 12.5% reduction in the listed price of the drug. 3 In the present litigation, the Commonwealth seeks to recover from Sanofi the loss it claims to have suffered on the non-triggering of price reductions for clopidogrel after Sanofi successfully prevented the entry of a generic version of clopidogrel into the Australian market. This occurred on 25 September 2007 when Sanofi obtained an interlocutory injunction restraining Apotex Pty Ltd ('Apotex') from launching into the Australian market its generic form of clopidogrel. Sanofi obtained this interim relief on the basis of a patent later held to be wholly invalid. At the time Sanofi obtained the interlocutory injunction, it proffered an undertaking to the Court that it would compensate any person adversely affected by the grant of the injunction. 4 The Commonwealth's case at trial was that it was entitled to recover from Sanofi a range of price reductions that would have occurred in the pricing of clopidogrel starting on 1 April 2008. In essence, the Commonwealth seeks to recover the harm done to the PBS by Sanofi's thwarting of the entry of a generic competitor into the Australian market.

BACKGROUND AND ISSUES 5 In 2007 Apotex (which until 20 September of that year was known as GenRx Pty Ltd) wished to launch a generic version of clopidogrel into the Australian market. However, Sanofi held a patent, Australian Patent No 597784 ('the Patent') which was for clopidogrel, a process for its preparation and pharmaceutical compositions containing it. The active pharmaceutical ingredient in Plavix and Iscover is a salt, clopidogrel hydrogen sulfate. That salt was the compound specified in claim 3 of the Patent. The active pharmaceutical ingredient in Apotex's proposed generic clopidogrel product was also the hydrogen sulfate salt. 6 The Patent was therefore an obstacle that stood in the way of Apotex entering the Australian market with its generic product. On 16 August 2007, it took its first steps to clear that obstacle out of the way by commencing proceedings in this Court seeking to revoke the Patent on the basis that it was invalid. Sanofi, apprehending that Apotex was going to seek to invade the monopoly it held in the Australian clopidogrel market by virtue of the Patent, filed a defence denying that the Patent was invalid and filed a cross-claim against Apotex on 17 September 2007 which sought, inter alia, final injunctive relief restraining it from distributing its clopidogrel products in Australia. In the cross-claim Sanofi also sought a final injunction restraining Apotex from applying to have those products listed on the PBS. Also on 17 September 2007, Sanofi filed a notice of motion seeking equivalent interlocutory injunctions pending the trial of the substantive dispute between the parties as to the validity of the Patent. 7 Sanofi's application for an interlocutory injunction came before Gyles J on 18 and 21 September 2007 and again on 25 September 2007 when his Honour published reasons and made orders granting an interlocutory injunction restraining Apotex from distributing its clopidogrel products in Australia. The price Sanofi paid for this injunction was the proffering by it of an undertaking to the Court that if it should eventually transpire that it had not been entitled to enforce the Patent and therefore had not been entitled to the interlocutory injunction, it would compensate any person adversely affected by the injunction. The undertaking and the injunction it supported took this form: UPON the Respondent/Cross-Claimant undertaking to the Court to: (a) submit to such order (if any) as the Court may consider to be just for the payment of compensation, to be assessed by the Court or as it may direct, to any person whether or not a party, adversely affected by the operation of Order 1 set out below or any continuation (with or without variation); and (b) pay the compensation referred to in (a) to the person or persons there referred to. THE COURT: 1. ORDERS that, pending the determination of the proceedings or further order, the Applicant/Cross-Respondent whether by itself, its directors, officers, servants, agents or otherwise, be restrained from infringing Australian Letters Patent No 597784 (the Patent) and, in particular, from engaging in the following acts within the patent area (as that term is defined in Patents Act 1990 (Cth)), without the license or authority of the Respondent/Cross-Claimant: (a) making, selling or otherwise disposing of the products known as GenRx Clopidogrel, Apo-Clopidogrel, Chemmart Clopidogrel and Terry White Chemists Clopidogrel or any other pharmaceutical composition the active ingredient of which is clopidogrel bisulfate, a compound claimed in claims 1, 3, 10 and 11 of the Patent (collectively, the GenRx Clopidogrel Products); (b) offering to make, sell or otherwise dispose of the GenRx Clopidogrel Products; (c) using or importing the GenRx Clopidogrel Products; (d) keeping the GenRx Clopidogrel Products for the purpose of doing any of the acts described in sub-paragraphs (a) to (c) above; (e) authorising other people to engage in any of the acts described in sub-paragraphs (a) to (d) above. 2. NOTES that the Applicant/Cross-Respondent undertakes to the Court that, pending the determination of the proceedings or further order, it will not, whether by itself, its directors, officers, servants or agents or otherwise, take any steps to obtain listing of any of the GenRx Clopidogrel products under the pharmaceutical benefits scheme maintained by the Commonwealth under the National Health Act 1953 (Cth). 8 There are three matters to note about this. First, Sanofi's undertaking as to damages was expressed to extend to persons adversely affected by the injunction whether they were parties to the litigation or not and it is this aspect of it which the Commonwealth submits founds its entitlement to seek compensation. Secondly, the notation refers to an undertaking by Apotex not to seek PBS listing of its products. Thirdly, although the injunction restraining Apotex from distributing its clopidogrel products was supported by Sanofi's undertaking as to damages, Apotex's undertaking not to seek listing of its clopidogrel products on the PBS was not supported by that undertaking. 9 The Commonwealth alleges that if the interlocutory injunction had not been granted then Apotex would have attempted to list its generic product on the PBS on 1 April 2008 and would have succeeded in that endeavour. This would have triggered three kinds of price reductions. The amounts claimed by the Commonwealth are large. Clopidogrel was, for the financial year ending 30 June 2008, the third most heavily subsidised prescribed drug in Australia and the cost involved to the Commonwealth for that year alone was approximately $170 million. The Commonwealth submits that the mandatory 12.5% price reduction triggered by the launch of a PBS-listed generic clopidogrel product would have reduced the price per packet of Plavix from $64.94 to $56.82 from 1 April 2008 and there would have been a further price reduction of 2% on 1 August 2009. These would have saved it $50,718,312. It says a different system of price reductions known as 'price disclosure reductions' would have saved it a further $215,922,465 from 1 April 2010. In addition, it claims $58,307,301 for payments made by it in relation to certain combination products containing both clopidogrel and aspirin. The total sought by the Commonwealth exceeds $325 million together with interest. 10 Apotex was partially successful at trial in invalidating part but not all of the Patent. On appeal to the Full Court it was wholly successful and the Patent was declared invalid. An application to the High Court by Sanofi for special leave to appeal was refused on 12 March 2010. On 11 April 2013, the Commonwealth filed an interlocutory application seeking compensation from Sanofi pursuant to the undertaking as to damages it had proffered and which was reflected in the orders of Gyles J dated 25 September 2007. The trial judge dismissed the Commonwealth's claim on 28 April 2020: Commonwealth of Australia v Sanofi (formerly Sanofi-Aventis) (No 5) [2020] FCA 543 ('J'). It is from his Honour's orders that the Commonwealth now appeals. For its part Sanofi has filed a notice of contention seeking to uphold the trial judge's reasons on alternate grounds. The two principal issues which occupied most of the appeal are: (a) The Apotex Launch and Listing Issue. The trial judge found Apotex would not have sought to list its generic clopidogrel products on the PBS on 1 April 2008 even if it had not been restrained by the interlocutory injunction. The Commonwealth argues that the primary judge erred in making that finding in four ways: (a) he applied the wrong standard and should have asked whether the Commonwealth had made out a prima facie case that Apotex would have sought listing by 1 April 2008; (b) he overlooked critical evidence; (c) he drew an inference adverse to the Commonwealth because of its failure to call the chief executive officer ('CEO') of Apotex, Dr Barry Sherman, to give evidence about what Apotex would have done but for the injunction; and (d) his Honour delayed for a substantial period of time before delivering judgment and thereby lost any advantage he had in relation to the credit of witnesses. These issues are encompassed in Ground 2 of the Commonwealth's amended notice of appeal. (b) The Directness Issue. The trial judge concluded that even if he had found that Apotex would have successfully listed on the PBS on 1 April 2008, nevertheless, the loss claimed by the Commonwealth did not flow directly from the interlocutory injunction and therefore did not satisfy the requirements of Air Express Ltd v Ansett Transport Industries (Operations) Pty Ltd (1981) 146 CLR 249 ('Air Express') and European Bank Ltd v Evans [2010] HCA 6; 240 CLR 432 ('European Bank'). His Honour did accept that the claimed loss would not have occurred but for the grant of the interlocutory injunction and in that sense was caused by it. However, he thought that the loss claimed flowed directly from the undertaking not to seek PBS listing which Apotex had proffered. Whilst his Honour accepted that the undertaking not to list would not have been proffered if the interlocutory injunction had not been granted, nevertheless he concluded that its existence as an intermediate step between the interlocutory injunction and the loss claimed meant that the loss could not be said to flow directly from the interlocutory injunction. His Honour also thought that it was significant that the undertaking not to seek listing proffered by Apotex was not supported by the undertaking as to damages proffered by Sanofi. This issue is the subject of Ground 1 of the amended notice of appeal. 11 There were other issues in the appeal none of which, however, arise if the Apotex Launch and Listing Issue is, as we believe it should be, resolved adversely to the Commonwealth. These issues related to further hypothetical causative obstacles sequentially secreted within each other and all of them within the overarching hypothetical situation in which Apotex did seek to list its clopidogrel products on the PBS from 1 April 2008 and to launch those products in Australia. 12 Thus, lest the Commonwealth succeed on appeal in demonstrating that Apotex would have sought that listing, Sanofi took issue with his Honour's alternative conclusion, had he so found, that the Minister or the Minister's delegate would most likely have approved that application (Notice of Contention Ground 1). The Commonwealth then disputed his Honour's further conclusion that if that had occurred then Apotex would, in any event, have ceased to supply its products following the grant by Gyles J of final injunctive relief in August 2008 (Notice of Appeal Ground 3(c)). In the hypothetical scenario where Apotex did continue after August 2008 to supply its products, the Commonwealth challenges the trial judge's conclusion that various price reductions which would then have occurred were not recoverable by it since, in fact, they did not occur (Notice of Appeal Ground 3(d)). In the event that the Commonwealth succeeds on that issue, Sanofi then submitted that the trial judge ought to have found that these losses did not flow directly from the grant of the interlocutory injunction (Notice of Contention Grounds 2 and 4). Sanofi also submitted that the trial judge ought to have found that all of the claimed losses were too remote (Notice of Contention Ground 3). More generally the Commonwealth took issue with his Honour's refusal to deal with its case as a loss of a chance case (Notice of Appeal Ground 4) and an argument based upon undertakings given by Sanofi in subsequent appeal litigation (Notice of Appeal Ground 3(a)). For its part, Sanofi submitted that the Commonwealth ought not be permitted to recover any compensation since, as a polity, it was not a person adversely affected by the interlocutory injunction (Notice of Contention Ground 5). Finally, Sanofi submitted that the trial judge had erred in rejecting a number of discretionary defences to the compensation claims such as delay (Notice of Contention Ground 6). Because we are of the view that the Commonwealth fails at the threshold in its challenge to his Honour's conclusion that Apotex would not have sought to list its clopidogrel products from 1 April 2008 we see no utility in resolving these issues. The Directness Issue also does not need to be resolved but, out of deference to the extensive submissions made about it and because it involves a question of principle, we will indicate our views on it. In the result the appeal will be dismissed with costs. 13 We begin with Ground 1 and the question of whether the claimed loss flowed directly from the injunction.

THE DIRECTNESS ISSUE: NOTICE OF APPEAL GROUND 1 14 The premise upon which the Commonwealth's claim for damages rests is that but for the injunction, Apotex would have successfully launched its clopidogrel products on 1 April 2008 including by applying for and successfully securing listing on the PBS by that date. The trial judge was not persuaded of this as a matter of fact and we reject the challenge to that conclusion below. 15 However, his Honour also concluded that even if he had accepted that Apotex would have launched its products by 1 April 2008 including by successfully achieving PBS listing by that date, he still would not have accepted the Commonwealth was entitled to damages. This was because whilst he accepted that the loss claimed by the Commonwealth would not have occurred if the interlocutory injunction had not been granted, nevertheless, he did not accept that the claimed loss flowed directly from it. This conclusion was material because of the High Court's decision in Air Express which contains a statement by Aickin J (sitting as a trial judge) that in a claim for compensation pursuant to an undertaking as to damages it is usually necessary that the loss should flow directly from the interlocutory injunction supported by the undertaking: Air Express at 266-267. 16 As mentioned above, the trial judge thought that the claimed loss flowed directly from the undertaking not to seek PBS listing which Apotex had proffered. In fact the trial judge described the loss as the 'direct consequence' of the Apotex undertaking - a formulation which his Honour understood to be interchangeable with that of Aickin J: J [445], [451]. That undertaking was therefore causally interposed between the interlocutory injunction and the loss claimed by the Commonwealth and, as such, the loss claimed could not be said to flow directly from the injunction. His Honour also thought this conclusion was supported by the fact that Apotex's undertaking not to seek listing was not supported by Sanofi's undertaking as to damages. 17 The resolution of the issues relating to this ground turn on the events surrounding the grant of the interlocutory injunction and the form of the undertakings proffered by Sanofi and Apotex. 18 Sanofi filed a notice of motion seeking two interlocutory injunctions on 17 September 2007. The first was an injunction restraining Apotex from supplying its generic products into the Australian market. The second was an injunction restraining Apotex from seeking to list those products on the PBS. In the notice of motion, Sanofi proffered an undertaking in respect of both injunctions that it would submit to such order as the Court may consider just for the payment of compensation to be assessed by the Court 'to any person whether or not a party, adversely affected by the operation' of the injunctions. 19 This application came before Gyles J on Tuesday, 18 September 2007. One issue which was the subject of debate was whether seeking PBS listing for its clopidogrel products was an act by Apotex which infringed Sanofi's Patent. It is apparent that Gyles J was sceptical that it could be. At T22.7-22 this exchange with senior counsel for Sanofi took place: HIS HONOUR: Mr Bannon, how can you possibly stop them getting the listing, or how could I possibly stop them getting a listing done? MR BANNON: Because that would be a step towards launching. HIS HONOUR: You can forget that. MR BANNON: Sorry? HIS HONOUR: You can forget that. I just don't - at the moment I just can't see a proper basis why I could intervene that process at all, based upon the patent. I mean, if there's something I've overlooked, well of course I'll review that, but at the moment, just as a matter of principle, I can't quite see it. You know, you can come back to that, you know, it's not a final view, but it's - - - MR BANNON: Yes. Well I will address that. 20 As it happens, his Honour's doubts about this were well founded for it has been subsequently established that a person who seeks PBS listing for a drug does not exploit a patented invention: Warner-Lambert Company LLC v Apotex Pty Ltd [2017] FCAFC 58; 249 FCR 17. 21 However, due to the way in which the hearing played out it became unnecessary for his Honour to determine this issue. This was because of two practical matters raised during argument at the hearing. The first was mentioned by senior counsel for Sanofi immediately after Gyles J had expressed his scepticism that the Court could restrain Apotex from seeking PBS listing. It was submitted that Apotex could not 'plausibly apply for listing without indicating they're going to be able to supply': T22.36-37. This was being put by Sanofi as a reason why the Court's power to grant ancillary relief might extend so far as to restrain PBS listing however the submission reflected both a practical and a legal reality. 22 The practical reality was that the Health Department imposed as an administrative requirement in the process of achieving PBS listing that an applicant provide a written assurance that sufficient stock of its brand would be available on the proposed date of listing to meet anticipated demand. 23 The legal reality was that from 1 August 2007 s 99AEB of the National Health Act 1953 (Cth) required that for a newly listed brand the supplier must supply the medicine for a period of 24 months beginning on the day of PBS listing. Failure to comply with this statutory obligation of supply could result in the delisting of the brand or any other brand held by that person: s 99AEH(2). It was also an offence. In practical terms, a pharmaceutical company which failed to meet its supply obligations during this two year period faced the potential risk of having all of its products removed from the PBS. Such an outcome would be commercially undesirable. 24 These two matters are related in the sense that both are directed at the need of an applicant for listing to be able to meet its supply obligations. The assurance of supply was prospective in nature and was required before listing, but had no legal effect. The guarantee of supply was enlivened on listing on the PBS and was attended with significant legal and commercial peril if not met. 25 The point senior counsel for Sanofi was making was that once Apotex was restrained from supplying its generic products it would not be able to give the written assurance of supply which was necessary, at least in a practical if not legal sense, in order to obtain listing. Further, assuming Apotex obtained listing, the injunction would also prevent it from complying with its supply obligations under s 99AEB, a failure which could lead to the whole of Apotex's listed pharmaceutical repertoire being removed from the PBS. Sanofi was making this point as part of an endeavour directed at persuading Gyles J that PBS listing was so bound up in questions of supply that the power of the Court to restrain the latter carried with it ancillary jurisdiction extending to the former. 26 The second practical reality was identified by senior counsel for Apotex. It related to the automatic price reductions. Once Apotex's products were listed on the PBS this would result in an automatic 12.5% reduction in the price of Sanofi's products to pharmacists. Although this had no direct impact on the pharmacists' profits it had an indirect effect because, as the primary judge explained at [65], a mark-up which pharmacists were entitled to charge (known as the dispenser mark-up) was a percentage of the price at which the pharmacist purchased the product from Sanofi. Thus a reduction in that price affected a reduction in the mark-up which could be charged by pharmacists and, therefore, a reduction in their profits. 27 Where Apotex, however, was restrained from supplying the generic product, PBS listing would have resulted only in the reduction of the pharmacists' profits, corresponding resentment by them towards Apotex and, at the same time, no benefit to Apotex apart from the negative one consisting of the risk of having all of its products taken off the PBS for not complying with its supply obligations. 28 This second matter played out before Gyles J at T79.32-80.29 where his Honour raised the issue he had raised with senior counsel for Sanofi as to the ability of the Court to restrain PBS listing: HIS HONOUR: Could I just - sorry, to interrupt you, but there's one matter which I really had raised this morning and I meant to take it up with Mr Bannon and you both, and that is this question of - really it is for Mr Bannon, how can he get a - how can he stop your registration? MR CATTERNS: Your Honour, our applying on 1 December. If your Honour - - - HIS HONOUR: Is that under this - have you got any certificates and all that sort of thing? MR CATTERNS: We have done those, your Honour. But for us to get PBS listing as an equivalent so they can be substituted at the pharmacy, we would, unless restrained from selling, apply in 1 December, list in 1 August. HIS HONOUR: April. MR CATTERNS: We would be prepared, if your Honour is against us and grants an injunction against us from selling, we will not apply to the PBS during the period of that restraint with liberty to apply. HIS HONOUR: Yes. All right. MR CATTERNS: Because, in short your Honour, it's pointless for us, the question of needing to restrain us from doing anything with the PBS or restraining the PBS doesn't arise. If your Honour were against us - - - HIS HONOUR: Why would that be, why would that be, because - - - MR CATTERNS: It's of no value to us, your Honour. We don't want to get listed, to be candid, we don t want to be listed on the PBS if we can't sell. It would damage our friends by bringing their price down twelve and a half per cent. HIS HONOUR: Yes, that's a risky thing anyway. MR CATTERNS: Yes. It will only damage our friends, it's of no benefit to us, and we will make ourselves enemies in the industry. HIS HONOUR: Yes. All right. I'm not inquiring really why, I just want to know what the situation is. MR CATTERNS: So, your Honour, we would be happy to - if your Honour, we would be unhappy, but if your Honour makes an injunction against us selling, we would agree on an appropriate undertaking that would fix that up. HIS HONOUR: Yes. All right. Thank you. 29 It was at this point that the question of the Court's power to restrain listing became moot. Because Apotex had indicated that it would proffer an undertaking not to seek listing if Gyles J otherwise granted an interlocutory injunction against supply, the issue no longer mattered, at least from the perspective of Gyles J. 30 However, the precise bargain which had been worked out between the parties was a little unclear. Sanofi had been seeking an injunction restraining Apotex from seeking PBS listing supported by an undertaking as to damages. Apotex had now indicated that if an injunction against supply was granted it would undertake not to seek PBS listing. On that basis Sanofi did not need to pursue its application. However, was Apotex's undertaking to be secured by Sanofi's undertaking as to damages or was it to be freestanding? This uncertainty marks the beginning of a conceptual fissure which runs all the way through to this appeal. 31 Gyles J reserved his decision on the Tuesday and gave judgment on the next Friday, 21 September 2007. He decided that Sanofi was entitled, in principle, to an interlocutory injunction restraining Apotex from supplying its generic products. It is apparent from the transcript of another hearing which took place on the following Tuesday, 25 September 2007, that his Honour's judgment went through two phases of development. The first was an oral judgment delivered on Friday, 21 September 2007. The second was a revised set of those oral reasons published in written form on Tuesday, 25 September 2007. The form of the published reasons show that no orders were made on Friday, 21 September 2007 and no undertakings were proffered. The train of events appears to have been that the parties considered the implications of his Honour's Friday reasons and formulated orders and undertakings to give effect to them for a directions hearing then held on Tuesday, 25 September 2007. It was at the commencement of that hearing that his Honour published the revised reasons. This exchange then took place at T1.8-40: HIS HONOUR: Mr Bannon. Yes, I'll publish the revised reasons that I delivered last Friday. MR BANNON: Your Honour, we've got some short minutes which are agreed, so they're for your Honour's consideration. There's two copies, if needed. HIS HONOUR: Yes, thank you. Yes, excuse me for a moment. Yes, I have got no difficulty with any of those matters. Now, the undertaking is given to the court, is that correct? MR BANNON: Yes. HIS HONOUR: That's the undertaking at the commencement of the - - - MR BANNON: Yes, I give on behalf of the respondent the first undertaking, the undertaking as to damages. HIS HONOUR: Yes. MR CATTERNS: And your Honour remembers your Honour raised the jurisdictional question relating to our applying to a listing under the PBS. HIS HONOUR: Yes. MR CATTERNS: And we've agreed - your Honour remembered in running we agreed that if your Honour would make the first injunction, to save the jurisdictional argument, we would make the second undertaking. HIS HONOUR: Yes, and you give that undertaking? MR CATTERNS: Yes, your Honour, I do. HIS HONOUR: Well, that's noted. That's the undertaking in paragraph 2, all right. 32 The form of the injunction against supply, Sanofi's undertaking as to damages, and Apotex's undertaking not to seek listing appear above at [7] and need not be set out again. 33 The trial judge found that the loss claimed by the Commonwealth did not flow directly from the interlocutory injunction because it flowed directly from the undertaking not to seek PBS listing which was therefore an interposed causal step. He also thought that the fact that the Apotex undertaking was not supported by Sanofi's undertaking as to damages provided strong contextual support for the view that Sanofi's undertaking should not be interpreted as extending to loss suffered by third parties such as the Commonwealth as a result of Apotex being prevented from applying to list on the PBS: J [446]. 34 In our opinion, his Honour erred in this conclusion. There are four reasons for this. 35 First, we do not agree that the loss claimed by the Commonwealth flowed directly from Apotex's undertaking not to seek to list on the PBS. The trial judge approached the matter this way: although the injunction did not in terms prevent Apotex from applying for PBS listing, it made it impossible for it to give the necessary assurance of supply or to comply with its supply obligations during the guaranteed period. The trial judge therefore accepted that the interlocutory injunction had the practical effect of preventing Apotex from applying for PBS listing from 1 April 2008 assuming it was otherwise willing and able to do so. (As will be seen the consequences of this finding are significant). But even so, his Honour did not accept that the interlocutory injunction directly affected the legal rights, obligations or interests of the Commonwealth: J [443]. In particular, his Honour noted that it would not have been a breach of the interlocutory injunction for Apotex to have taken steps to obtain PBS listing: J [445]. 36 This fact led his Honour to conclude that the Commonwealth's loss was a natural and direct consequence of the inability of Apotex to apply for PBS listing: J [445]. This was the conduct to which the Apotex undertaking was directed but it was not something required by the interlocutory injunction. This suggested to his Honour that the loss was an indirect consequence of the injunction: J [445]. Indeed, his Honour concluded that the direct cause of the loss was the undertaking Apotex gave not to seek PBS listing: J [451]. 37 As we have explained above, as the trial judge found at J [428], the interlocutory injunction put paid to any plan by Apotex to launch its generic product. Consequently it could not have complied with its obligation of supply for the two year period referred to in s 99AEB of the National Health Act 1953 (Cth) nor would it have been able to proffer to the Department a written statement that it could guarantee that supply. From the moment of the grant of the interlocutory injunction there was not the slightest prospect that Apotex would seek PBS listing. This was the very proposition which had been put to Gyles J by senior counsel for both Sanofi and Apotex and it was the reason that the trial judge accepted at J [428] that the interlocutory injunction had the practical effect of preventing Apotex from seeking PBS listing for its generic products. 38 Once the interlocutory injunction was in place the proffering of the undertaking by Apotex was therefore no more than a promise not to do something which Apotex could never have done. The trial judge thought that the Apotex undertaking was the direct cause of the Commonwealth's loss. We respectfully disagree. If during the hearing on 18 September 2007 Apotex had refused to give the undertaking not to seek PBS listing Gyles J would still certainly have granted the interlocutory injunction restraining supply but on the issue of PBS listing there were only three possible outcomes which could have occurred: (a) Gyles J could have granted an interlocutory injunction restraining PBS listing; (b) Gyles J could have refused the application for that injunction; or (c) Sanofi could have abandoned the application for that injunction. 39 Regardless of which of these happened, the finding by the trial judge at J [428] was that the practical effect of the interlocutory injunction against supply was that Apotex was prevented from applying for a PBS listing of its clopidogrel products on 1 April 2008, assuming it was otherwise willing and able to do so. It follows that regardless of which of these three outcomes occurred Apotex would not have sought PBS listing in any realistic way and the Commonwealth's claimed loss would still have occurred. 40 That conclusion means that the Commonwealth's claimed losses cannot be seen as flowing directly from the Apotex undertaking not to seek PBS listing. If the undertaking is removed from the counterfactual (that is to say, if any of (a), (b) or (c) above occurred) then the Commonwealth's loss would still have been suffered. Neither party disputes that where an undertaking as to damages is concerned, a claimant for loss caused by an injunction must generally show that damages claimed would not have occurred but for the injunction: Air Express at 313 per Gibbs J, 316 and 320 per Stephen J, 325 per Mason J. We are unable to reconcile the trial judge's finding that the loss flowed directly from the undertaking not to seek PBS listing with this aspect of Air Express. Since the loss would still have occurred even if the undertaking had not been proffered it cannot be said that the undertaking is the sine qua non of the occurrence of the loss. 41 The matter may be tested another way: if Apotex's undertaking not to seek PBS listing had been supported by Sanofi's undertaking as to damages, it would not now be possible for the Commonwealth to recover under Sanofi's undertaking because the 'but for' test could not be satisfied. In such a situation the argument that the Apotex undertaking not to seek PBS listing was the direct cause of the Commonwealth's loss would be precluded by the holding in Air Express that the claimant must satisfy the 'but for' test. We do not think that whether the loss flowed directly from the Apotex undertaking not to seek PBS listing can be affected by whether that undertaking was itself supported by an undertaking as to damages. 42 Indeed, the consequence of the trial judge's finding that the Apotex undertaking not to seek PBS listing was the direct cause of the Commonwealth's loss would have the result, if it had been supported by Sanofi's undertaking as to damages, that the Commonwealth could not have recovered under any circumstance. It could not have succeeded in relation to the injunction because of his Honour's finding that the loss did not flow directly from that injunction; and, it could not have succeeded in relation to the Apotex undertaking not to seek PBS listing because the Commonwealth's claimed loss would have occurred even if that undertaking had not been proffered so that the general sine qua non requirement in Air Express could not be satisfied. 43 Consequently, in concluding that the Apotex undertaking not to seek PBS listing was the direct cause of the Commonwealth's loss the trial judge erred by failing to apply the 'but for' requirement in Air Express. Had he done so, his Honour would have been required to conclude that the Apotex undertaking not to seek PBS listing was not a cause of the Commonwealth's loss. 44 Once that conclusion is reached that leaves in place only his Honour's finding that the interlocutory injunction was an indirect cause of the Commonwealth's alleged loss. Shorn of the conclusion that the direct cause of the loss was the undertaking not to seek PBS listing, there is, in our opinion, no avoiding the conclusion that the interlocutory injunction was the direct cause of the Commonwealth's alleged loss and not an indirect cause as the trial judge held (for clarity, we are presently assuming, contrary to what we will find in relation to Ground 2 of the amended notice of appeal, that Apotex would have applied for and obtained PBS listing by 1 April 2008). 45 Secondly, even if Apotex's undertaking not to seek PBS listing was causally connected to the Commonwealth's loss (in the 'but for' sense) we would not accept that its presence in the causal chain implied inevitably that the loss could not also flow directly from the interlocutory injunction as we now explain. 46 His Honour's findings on directness were material to his Honour's reasoning because of a passage in the reasons of Aickin J in Air Express at 266-267: In a proceeding of an equitable nature it is generally proper to adopt a view which is just and equitable, or fair and reasonable, in all the circumstances rather than to apply a rigid rule. However the view that the damages should be those which flow directly from the injunction and which could have been foreseen when the injunction was granted, is one which will be just and equitable in the circumstances of most cases and certainly in the present case.' (emphasis added) 47 This passage was cited with approval by five Justices in European Bank at [18] and [29] in these terms: These considerations, bearing upon the interests of justice in the particular circumstances of the litigation, support the following statement by Aickin J in Air Express, made with respect to interlocutory injunctions, but applicable to the interlocutory order made by the Court of Appeal against European Bank. His Honour said: "In a proceeding of an equitable nature it is generally proper to adopt a view which is just and equitable, or fair and reasonable, in all the circumstances rather than to apply a rigid rule. However the view that the damages should be those which flow directly from the injunction and which could have been foreseen when the injunction was granted, is one which will be just and equitable in the circumstances of most cases and certainly in the present case." The phrase "could have been foreseen" should be noted. … On the inquiry before Gzell J the first question was "What is the loss that is now alleged?", the second "Did that loss flow directly from the order of 18 May 2004?" and the third "Could the loss sustained have been foreseen at the time of that order?" The inquiry presented by the third question is an inquiry as to whether a loss of the kind actually sustained could have been foreseen. Contrary to the submission by the respondent, Mr Evans, the inquiry is not as to whether the actual loss suffered was foreseen at the time the undertaking was given. 48 His Honour therefore posed for himself the question of whether the loss claimed by the Commonwealth had flowed directly from the interlocutory injunction. Because, on the trial judge's findings, Apotex's undertaking not to seek PBS listing was the direct cause of the claimed loss it followed that the loss had not flowed directly from the interlocutory injunction in that sense. Although his Honour did not use the language of an interposed causal step we are unable to construe his Honour's conclusion other than that the presence in the causal chain of another cause of the loss located between the interlocutory injunction and the loss claimed by the Commonwealth meant that the loss could not be said to flow directly from the injunction. 49 The Commonwealth submits that the statement in Air Express that the loss must flow directly from the interlocutory injunction does not entail that there can be no interposed causal step between an interlocutory injunction and the loss claimed under an undertaking as to damages. Sanofi, on the other hand, submits that the statement reflects a limitation on the scope of recovery which has been accepted in a number of authorities many of which were referred to by Aickin J in passages of his reasons which were immediately prior to the passage set out above. That existence of that limitation supported, so submitted Sanofi, the trial judge's conclusion that there should be no interposed causal step. 50 The Commonwealth's response to this was that the cases referred to by Aickin J were all concerned with remoteness of damage and largely not with causation. Further, to the very limited extent to which these cases did touch on principles of causation none provided support for the proposition that the requirement that the loss should flow directly from the interlocutory injunction necessitated that there should be no other causal step interposed between it and the loss claimed. 51 There is some uncertainty in our minds as to whether the statement made by Aickin J is a statement about causation, or remoteness or possibly both. Support for the view that it is a statement about causation may be garnered from the fact that the word 'flow' clearly implies some causative notion. Further, the subsequent reference by Aickin J in the same passage to an additional requirement of foreseeability may be apt to suggest that the 'flow directly' element was not a remoteness standard. 52 On the other hand, support for the view that the 'flow directly' requirement is concerned with remoteness of damage may be discerned from the fact that it appears in the reasons immediately after a survey of cases which are largely concerned with remoteness of damage. This interpretation is also consistent with the fact that the issue decided by Aickin J was actually one of causation - specifically the application of the 'but for' standard in this area - and that discussion occurred only later in his Honour's reasons. 53 On the other hand, it seems to us from the passage quoted above that in European Bank the High Court approached the 'flow directly' requirement as being distinct from any remoteness inquiry. 54 The uncertainty about the relationship between the 'flow directly' requirement and notions of causation is not unique to this area of the law. The requirement that loss should flow directly appears in a variety of contexts and much ink has been spilt on how it relates to remoteness of damage. In relation to the tort of deceit, it appears that a plaintiff is entitled to compensation for all losses directly flowing from the tortious act and that direct losses also include consequential losses: Smith New Court Securities Ltd v Citibank NA [1997] AC 254 ('Smith New Court') at 264-265 per Lord Browne-Wilkinson, 281-282 and 285 per Lord Steyn. The same approach was taken by the High Court to the tort of injurious falsehood in Palmer Bruyn & Parker v Parsons [2001] HCA 69; 208 CLR 388 at [54], [63]-[65], [76]-[79] per Gummow J, Gleeson CJ agreeing at [13]-[14], Hayne and Kirby JJ concurring in the result. In that case Gummow J observed (at [53]) that remoteness can sometimes be understood 'in terms of causation' rather than in terms of foreseeability, which has a certain resonance with the conceptual fluidity of the passage in Air Express. 55 Reference to losses which flow directly may also be found in the context of statutory misleading or deceptive conduct. A recent example is Wyzenbeek v Australasian Marine Imports Pty Ltd [2019] FCAFC 167; 272 FCR 373 ('Wyzenbeek') at [59], [73], [78]-[79], [93] and [111] per Rares, Burley and Anastassiou JJ, citing the High Court in HTW Valuers (Central Qld) Pty Ltd v Astonland Pty Ltd [2004] HCA 54; 217 CLR 640 at [65] for the proposition that s 82 of the Trade Practices Act 1974 (Cth) 'allowed a court to assess loss or damage on the basis of the loss flowing directly from the transaction without any reference to the date of the transaction or to any particular date'. More significantly the causation requirement in this area has not been thought to prevent the recovery of consequential losses. As Brennan J observed in Sellars v Adelaide Petroleum NL (1994) 179 CLR 332 ('Sellars') at 356-357, where the making of a false representation induces a person to act in a certain manner, loss may flow directly from the act and only indirectly from the making of the representation but in such cases the act 'is a link - not a break - in the chain of causation'. 56 Similarly, it would appear that in a claim for equitable compensation for breach of trust, notions of foreseeability of loss are not generally relevant, but the loss must 'be caused by the breach of trust, in the sense that it must flow directly from it': AIB Group (UK) plc v Mark Redler & Co Solicitors [2015] AC 1503 at [135] per Lord Reed. 57 Although none of these statements answers the question which is before the Court, they do underscore that the distinction between causation and remoteness may not be as clear cut as the arguments before this Court implied. They also suggest that the concept does not exclude claims for consequential losses. 58 It is not necessary to resolve this conundrum definitively. Regardless of where the requirement that loss should flow directly is to be accommodated as a matter of jurisprudential taxonomy, there is no dispute that it is indeed a requirement. That observation must, however, be understood in light of the equitable nature of the remedy at hand and the High Court's insistence that the principles in this area are always flexible: Air Express at 266-267 per Aickin J, 324 per Mason J; European Bank at [16]-[18]. It was not suggested in this case, however, that the admitted requirements of flexibility militated against the application of the directly flow requirement. 59 The real question is not where the 'flow directly' requirement fits into the structure of the topic but rather the ascertainment of its content. We do not think that there can be any hard and fast rule that an interposed causal step between an injunction and the loss claimed results in the injunction not being the direct cause of the loss. It depends on the nature of the interposed causal step. 60 For completeness, it is worth noting Sanofi's submission that the 'flow directly' requirement must be applied with 'particular care' in the case of claims on the undertaking made by non-parties such as the Commonwealth, in order to avoid a situation in which actual or potential losses are 'practically indeterminate': RS [21]. We accept that, as in other areas of the law, there is a need to ensure that some restraint is imposed on the recovery of damages - both as to the number and identity of claimants, and as to quantum. Assuming, in Sanofi's favour that the requirement that losses should flow directly from the injunction is intended to perform a limiting function complementary to that performed by the requirement that losses must have been of a kind that was reasonably foreseeable, it does not follow that, in every case, the mere existence of an interposed causal step between the injunction and the loss claimed will have the consequence that the loss did not flow directly from the injunction. Again, it depends on the nature of the interposed step and the circumstances more broadly, assessed having regard to the equitable rationale for the award of damages in this area. Indeed, as Sanofi recognised in its submissions, that task is 'one of evaluative judgment': T157.1-2. 61 We do not think therefore that any relevant guidance is to be obtained from the various cases referred to by Aickin J in the passages immediately before his statement that the loss must flow directly. These were: Smith v Day (1882) 21 Ch D 421; Ex parte Hall, in re Wood (1883) 23 Ch D 644; Schlesinger v Bedford (1893) 9 TLR 370 (CA); Douglass v Bullen (1913) 12 DLR 652 ('Douglass v Bullen'); and Re an Arbitration between Pemberton and Cooper (1912) 107 LT 716 ('Pemberton'). Sanofi seeks to derive support for its argument from various expressions deployed in these cases, for example Britton J's statement in Douglass v Bullen at [24] that 'the damages ought to be confined to the immediate natural consequences of the injunction', and Bankes J's reference in Pemberton at 718 to 'damage which necessarily and naturally flowed'. Properly understood, these cases do not support a definitive view one way or the other on the question of whether the existence of an interposed causal step negates a finding that damage flowed directly from an injunction. And indeed, a closer inspection of passages relied upon by Sanofi suggest that they may in fact hinder rather than aid its submission. Two examples suffice: 62 In Douglass v Bullen, the plaintiff obtained an interim injunction, supported by an undertaking as to damages, restraining the defendant from undertaking certain building work on the defendant's land. It was later held that the plaintiff had not been entitled to the injunction. The defendant claimed damages including for alleged loss of rent. Britton J held that, in his opinion, the alleged loss of rent was 'too remote' because the prospect that the injunction would cause such loss was not something within the plaintiff's knowledge or which he ought reasonably to have contemplated: [24]. The effect of Britton J's reasoning would seem to be that if the plaintiff did contemplate the loss of rent, or ought reasonably to have done so, such damages may have been available notwithstanding, as is obvious, that they were the result of a series of causal steps. 63 In Pemberton, a tenant of a farm proposed to switch from raising sheep to growing corn. The landlord obtained an interim injunction restraining this and proffered an undertaking as to damages. It transpired that the period of the restraint corresponded with a drought which had the result that the sheep depreciated substantially in value whereas a corn crop would likely have succeeded. The tenant obtained damages from the landlord both for the depreciation of the sheep and the loss of the prospective corn crop: 718. Sanofi submits that Bankes J held that the tenant was entitled to these damages because they 'necessarily and naturally flowed from the interlocutory injunction' (RS [16]) but in fact what Bankes J said is that the damages 'necessarily and naturally flowed from the course which the landlord compelled [the tenant] to adopt': 718 (emphasis added), quoted by Aickin J in Air Express at 264-265. Thus, consistently with Brennan J's statement in Sellars to which we have referred above, Bankes J seemed to recognise that a course of conduct entered into by the claimant may amount to a series of links (rather than a break) in the chain of causation. 64 As we have explained, however, in truth the decisions referred to by Aickin J were simply not concerned with the effect of an interposed causal step on the 'flow directly' analysis and for that reason do not provide definitive instruction on that issue. 65 On the other hand, the reasoning in Air Express is against any general rule that a loss ceases directly to flow just because there is an interposed causal step. In fact, Aickin J rejected an argument in Air Express that the loss claimed by the appellant ('Air Express') was not recoverable because it was only an indirect result of the interlocutory injunction. By way of background, in that case the respondent ('Ansett') obtained an interlocutory injunction restraining a Commonwealth government official from issuing to Air Express a licence to import two planes. The injunction had no direct effect on Air Express in the sense that it did not require it to do anything or to refrain from doing anything. Indeed, Ansett did not join Air Express as a defendant and it became one only on its own application. Despite that, the form of the injunction never impacted directly on it in terms of legal rights, interests or obligations. 66 When the interlocutory injunction was eventually dissolved, the claim then made by Air Express under Ansett's undertaking as to damages principally related to the harm done to its business when it was unable to import the two planes. At 283, Aickin J recorded Ansett's argument that Air Express had been 'merely affected indirectly' and rejected it in these terms: Another argument was based on the fact that Air Express was not a party to the original proceedings and became a party on its own application in a situation in which no claim was made against it or could have been made against it. It was not restrained from taking any step itself and was merely affected indirectly by the injunction obtained against the Commonwealth and the Secretary to the Department. I think this argument is misconceived and is really an argument that it should not have been joined as a defendant at all. It would no doubt have been possible for Air Express to have been permitted to intervene in the argument without being made a party but the fact is that it was joined as defendant and that there was no appeal from that order. There is no direct authority on this point, but the decision in Tucker v. New Brunswick Trading Co. of London suggests that parties against whom no injunction is granted may nonetheless claim pursuant to an undertaking given in respect of them, though in that case the defendant in question was named as a defendant when the writ was issued. (citations omitted) 67 If his Honour had intended by the expression 'flow directly' necessarily to exclude cases where there was an interposed causal step, it is difficult to see why the loss claimed by Air Express would not have fallen foul of that requirement. On that view of affairs, Ansett's interlocutory injunction caused the official not to issue the licence and the non-issue of the licence then caused the loss claimed. That would be the same kind of interposed causal step which Sanofi now contends defeats a claim on the undertaking. This is an indication that his Honour did not intend that the 'flow directly' requirement should forbid such an interposed causal step. 68 To that we would add the observation that in other areas of the law where one encounters the 'flow directly' concept, claimants have nonetheless been entitled to recover consequential losses which points against the stricture for which Sanofi contends: e.g. Sellars; Wyzenbeek; Smith New Court (above). 69 A few remarks should also be made about the decision of Jagot J in Sigma Pharmaceuticals (Australia) Pty Ltd v Wyeth [2018] FCA 1556; 136 IPR 8 ('Sigma v Wyeth') to which both parties referred in submissions. 70 The facts of Sigma v Wyeth bear some similarity to the present appeal. The respondent ('Wyeth') was the proprietor of a method patent for venlafaxine, an anti-depressant drug which it sold under the brand name 'Effexor-XR'. The appellants (Sigma Pharmaceuticals (Australia) Pty Ltd, Generic Health Pty Ltd and Alphapharm Pty Ltd; together, 'the generics') brought proceedings against Wyeth challenging the validity of the patent. Wyeth counterclaimed for infringement and threatened infringement premised on the supply or threatened supply by the generics of their own venlafaxine products. Upon the application of Wyeth and supported by an undertaking as to damages which it gave, Jagot J granted interlocutory injunctions restraining the generics from supplying their venlafaxine products. Jagot J later upheld the validity of the patent and granted permanent injunctions. The generics succeeded on appeal to the Full Court, whereupon the patent was held to be invalid and the matter was remitted to Jagot J to determine claims on Wyeth's undertaking as to damages. 71 Importantly for present purposes, the generics were not the only claimants on Wyeth's undertaking. Three non-parties also claimed: two entities which manufactured and supplied venlafaxine to the generics ('the suppliers'); and the Commonwealth of Australia. The Commonwealth's claim was settled after the hearing and therefore did not remain for determination when Jagot J delivered her reasons. 72 Attention may presently be confined to the claims made by the suppliers. Wyeth resisted these claims on alternative bases: first, that the requirement that losses should 'flow directly' precluded the recovery by non-parties to the proceeding in which the injunction was granted; secondly, that the losses claimed by the suppliers were 'merely consequential upon the effect of the orders upon the generics and thus could not be said to be losses flowing directly from the orders': [219]-[220]. Her Honour disposed of the first contention on the basis that it was contradicted by principle and by the terms of the undertaking which referred to 'any person, whether or not a party, adversely affected' by the operation of the injunction: [219]. In respect of the second contention, her Honour said at [220]: The description of the manufacturers/suppliers' losses as consequential is accurate. If the generics are right that the orders prevented them from supplying their products when they otherwise would have done then the orders necessarily had consequential impacts on the manufacturers/suppliers. But this description does not answer the question whether the claimed loss of the manufacturers/suppliers flowed directly from the interlocutory orders or are too remote to be the subject of a just order for compensation. 73 We return shortly to the significance of the finding that the suppliers' consequential losses could nonetheless be 'direct' in the requisite sense. It is convenient at this point to say something about the concept of 'remoteness' as it appears in the passage just quoted and elsewhere in Jagot J's reasons. The Commonwealth is probably right to say, as it does in its reply submissions at [20], that Jagot J seemed to view the concept of 'directness' in this context as a remoteness limitation additional and complementary to the requirement that losses of the kind alleged must have been reasonably foreseeable: [12], [217], [222]-[225]. At the same time, however, her Honour identified (at [219]) the rationale of the 'flow directly' requirement in a manner which closely resembles the causation principle articulated in Air Express and European Bank: 'The concept of loss flowing directly from an interlocutory injunction reinforces the necessity of focusing on the allegedly adverse effects of the interlocutory orders alone, rather than the potentially adverse effects of the litigation'. As we have explained above, the taxonomic question of remoteness versus causation, while perhaps an interesting academic exercise, is not ultimately useful in identifying the content of the requirement of 'directness'. 74 What is noteworthy for present purposes is how Jagot J dealt with the suppliers' claims for consequential losses. Sanofi relies on her Honour's finding at [227] that the suppliers could not sustain claims based on supply contracts which did not exist and were contingent on future negotiations at the time the interlocutory injunctions were granted. Such alleged losses did not satisfy the requirement of directness even if the suppliers could show, with the benefit of hindsight, that the contracts would in fact have come about had the injunctions not been granted: [227]. 75 However, Sanofi did not refer to her Honour's finding in respect of supply contracts which did exist at the time the injunctions were granted (at [226]): … to the extent that the claims of the manufacturers/suppliers are based on contracts for supply of the products to Sigma, Alphapharm and Generic Health which existed before the interlocutory injunctions were granted, I am unable to characterise the claimed loss from the generics not having ordered products as anything other than loss which is the direct and natural consequence of the interlocutory orders which prevented supply. The products as supplied to the generics were the complete products in packages ready for supply to pharmacists. The fact of supply to the generics or not was a natural and ordinary consequence of and directly related to their sales to pharmacists. If restrained as they were by the interlocutory injunctions, it was inevitable that the generics would not place orders with their suppliers. 76 In other words, her Honour accepted that a claim on the undertaking could be made good even where the claimant was a non-party to whom the injunction was not expressly or impliedly addressed and whose losses were 'consequential' in the sense that they came about via a series of steps linking them to the parties restrained: in that case, the generics. This is very difficult, indeed impossible, to square with Sanofi's submission at [28] of its written outline that '[t]he fact that several successive steps are involved is itself sufficient to prevent the relationship between the injunction and the loss from meeting the requirements for directness'. That submission cannot be accepted. 77 We would therefore reject the proposition that the presence of an interposed causal step inevitably prevents a loss flowing directly from an injunction. Other than to make that observation we do not think it appropriate further to go. On the findings we have made, the Apotex undertaking not to seek PBS listing did not constitute an interposed causal step because it was in fact not a cause in the 'but for' sense required by Air Express. We would defer a determination of what kind of interposed causal step is sufficient to prevent a loss flowing directly from an interlocutory injunction to a time when that question needs to be answered in a concrete fashion. 78 Thirdly, the fact that the Apotex undertaking not to list was not supported by Sanofi's undertaking as to damages did not provide contextual support for his Honour's conclusions once one accepts, as we have, that the loss claimed flowed directly from the interlocutory injunction. The orders and undertaking must be interpreted and applied in accordance with their terms. Because the Commonwealth would not have been able to recover under Sanofi's undertaking as to damages even if it had been proffered in support of Apotex's undertaking not to seek listing (because the Apotex undertaking could not be the sine qua non of the Commonwealth's loss), the existence of the undertaking not to list has no impact on whether the loss claimed by the Commonwealth flowed directly from the interlocutory injunction or not. It becomes material only when it is possible to say that the loss claimed flowed directly from the undertaking not to seek listing. Of course, the trial judge did so find but, as we have explained, in our view it was erroneous for him to do so. 79 Fourthly, the question in this appeal is concerned with the scope of Sanofi's undertaking as to damages proffered not only to compensate Apotex in the event the injunction was dissolved but also any third party. If the effect of Apotex's proffering of an undertaking not to seek listing on the PBS and Sanofi's non-proffering of an undertaking as to damages in respect of that undertaking was to reduce the ambit of the undertaking as to damages in relation to third parties, it is difficult to see why this was not a matter which was immediately disclosed to Gyles J. More is this so when it is clear that his Honour was exercised about the impact of the injunction on the PBS and took that into account in considering the balance of convenience ('However, I am much influenced by the effects of disturbing the status quo, particularly as it relates to the operation of the PBS': GenRx Pty Ltd v Sanofi-Aventis [2007] FCA 1485 at [15]). 80 The third parties having the benefit of Sanofi's undertaking as to damages were not before the Court. No suggestion was made by Sanofi to Gyles J on 18, 21 or 25 September 2007 that if the injunction caused loss to a third party relating to the inability of Apotex to seek PBS listing then this would not be recoverable under its undertaking as to damages. If that truly was the case it was a matter which required explicit disclosure. 81 This is significant because the role of the Court in assessing damages is to award compensation which is just and equitable. No doubt in doing so it is necessary to take into account the undertakings proffered and the injunction ordered. But the Court must also take into account the justice of the position of third parties being adversely affected by inter partes arrangements together with the fact that no such adverse affectation was disclosed to the judge to whom the undertaking as to damages was proffered. 82 There were three aspects to the Commonwealth's contention under Ground 1. These were that the trial judge erred because he had: (a) identified an inappropriate causal test or standard of causal connection; (b) erred in assessing whether the Commonwealth's loss flowed directly from the interlocutory injunction; and (c) ought to have found that the Commonwealth's loss did flow directly from the terms of the interlocutory injunction. 83 We have upheld propositions (b) and (c). This is sufficient to conclude that had it arisen we would have upheld Ground 1. It is not necessary to consider whether (a) is made good in light of that conclusion. We now explain why Ground 1 does not arise.

THE APOTEX LAUNCH AND LISTING ISSUE: NOTICE OF APPEAL GROUND 2

Introduction 84 Between J [197]-[351], the primary judgment considered a substantive body of evidence both documentary and testimonial. This evidence concerned the question of whether Apotex, had it not been restrained by the interlocutory injunction, would have applied to list clopidogrel on the PBS and would then have launched its products on and from 1 April 2008. At J [351], his Honour reached this conclusion on this topic: In the result, I am not persuaded that Apotex Australia would have sought and obtained a PBS listing of its clopidogrel products from 1 April 2008 even if the interlocutory injunction had not been granted. It follows that the Commonwealth's claim for compensation must be dismissed. 85 By Ground 2, the Commonwealth contends that this conclusion was erroneous for four substantive reasons. First, the trial judge had erred in failing to conclude that the Commonwealth's onus was only to establish a prima facie case that its damage flowed directly from the grant of the interlocutory injunction (Ground 2(a)). Secondly, the trial judge had failed to have regard to: relevant contemporaneous evidence concerning Apotex's intention to launch if it was not restrained; relevant evidence from witnesses called by the Commonwealth; and associated submissions made by the Commonwealth about these matters (Ground 2(b)). Thirdly, his Honour had erred: in concluding that the Commonwealth's case suffered an evidentiary deficiency which could only be made good by its calling Dr Sherman to give evidence; in drawing a Jones v Dunkel (1959) 101 CLR 298 ('Jones v Dunkel') inference against the Commonwealth for failing to call Dr Sherman; and in failing to address the Commonwealth's associated submissions about these matters (Ground 2(c)). Fourthly, it was said that his Honour had lost the advantage usually afforded to a trial judge in relation to the assessment of the credit of witnesses by reason of the long delay in the production of his Honour's reasons (Ground 2(d)). The Commonwealth raised a fifth contention that the trial judge ought to have found that Apotex would have sought PBS listing had the interlocutory injunction not been granted (Ground 2(e)), however, this does not arise where no error is found in relation to the other particulars. 86 We begin with Ground 2(a).

GROUND 2(A): WHAT WAS THE COMMONWEALTH REQUIRED TO PROVE? 87 The Commonwealth puts its argument on Ground 2(a) this way: the compensation sought by a claimant pursuant to an undertaking as to damages is in the nature of equitable relief for the wrongful grant of an injunction. Where it appears that an injunction has been wrongly granted, the claimant for compensation is put in the difficult position of having to prove what would have happened if the injunction had not been granted. Equity comes to the assistance of the wronged party in that situation by requiring it to prove no more than a prima facie case of the loss caused by the non-performance of the act which was restrained by the injunction. It does this by inferring that the party who was restrained would have done the act which was restrained but for the injunction. Once that prima facie case is established, the evidentiary burden shifts to the respondent to the claim to prove that the prima facie case is wrong and that the loss claimed would not have occurred if the injunction had not been granted. This principle was said to be similar to that which occurs in the realm of causation in claims under s 52 of the Trade Practices Act 1974 (Cth) where it suffices for an applicant to make good a prima facie case on causation at which point the evidentiary burden shifts to the respondent to show otherwise: cf. Berry v CCL Secure Pty Ltd [2020] HCA 27; 271 CLR 151 at [29], [39]. The Commonwealth submits that the principle is established by two passages in Air Express. The first is in the reasons of Mason J at 322 and the second in the reasons of Stephen J at 320. 88 We do not accept the Commonwealth's submissions on Ground 2(a) for two reasons: (a) if the principle exists it has no application to this case; and (b) in fact, the principle is not established by any binding or persuasive authority.

(a) The principle has no application to this case 89 Assuming the principle to exist, Sanofi submitted that all it would require is for it to lead some evidence to discharge its evidential burden. Once that was done, the persuasive burden would remain unchanged and would rest on the Commonwealth: Purkess v Crittenden (1965) 114 CLR 164 at 167-168 per Barwick CJ, Kitto and Taylor JJ and at 170-171 per Windeyer J ('Purkess v Crittenden'). 90 Sanofi then submitted that it had in fact led evidence which discharged its evidential burden. We discuss these matters in more detail later, but for present purposes they included evidence pertaining to at least two matters: first, evidence tending to suggest that Apotex was better off being restrained with the benefit of the undertaking as to damages than it would have been if it had launched at risk and was exposed to the risk of damages for patent infringement; and, second, evidence that once it was no longer restrained by the injunction after winning its appeal to the Full Court, it still did not launch at risk whilst Sanofi's special leave application was pending in the High Court. This was said to be actual evidence of what Apotex would have done in the counterfactual. 91 We did not apprehend the Commonwealth to submit that this evidence did not have the effect of discharging Sanofi's evidentiary burden. Rather, the Commonwealth accepted the effect of Purkess v Crittenden but submitted that it was not enough, once the Commonwealth had established a prima facie case, for Sanofi to speculate about what Apotex might have done if it had not been restrained. Rather, this had to be the subject of evidence: T315.34-36. However, this submission is unresponsive to Sanofi's submission which referenced actual evidence about the question. Since the Commonwealth did not submit that the evidence relied upon by Sanofi did not have the effect of discharging the evidentiary burden, Sanofi's submission stands unanswered. Alternatively, if the Commonwealth's submission at T315.34-36 is to be understood as a denial that the evidence relied upon by Sanofi discharged the evidentiary burden then we reject the denial. Both matters referred to by Sanofi were evidential in nature rather than speculative. One concerned the financial position of Apotex and the incentives that would have been created. This is material from which an inference can be drawn. The other concerned actual evidence about how Apotex had behaved in a factual circumstance similar to the counterfactual. Again this is evidence from which an inference can be drawn. 92 In that circumstance, the effect of Purkess v Crittenden is not in dispute and either the Commonwealth has not met Sanofi's submission that there was evidence which discharged its evidential burden or, if it has, that submission should be rejected. Consequently, we accept that the effect of Purkess v Crittenden is that once both sides had adduced evidence on the topic of what Apotex would have done if not restrained, it remained for the Commonwealth to discharge its persuasive burden by showing, on all of the evidence, that Apotex would have sought to list on the PBS and then distribute its products in Australia. 93 As such, we do not think that the principle invoked by the Commonwealth has any relevance to this proceeding even if the principle exists.

(b) The principle is not required by any binding or persuasive authority 94 The Commonwealth relies upon the judgment of Mason J and the judgment of Stephen J in Air Express as the source of the principle. Dealing first with the reasons of Stephen J, we do not accept that his Honour said anywhere that it could be inferred from the fact that Air Express suffered loss because it did not receive an import licence, that the Secretary of the Department of Transport would have issued the permit to Air Express if he had not been restrained by the interlocutory injunction. In fact, Stephen J dismissed the appeal which is inconsistent with the drawing of any such inference. 95 As for the reasons of Mason J, we accept that if they represented the ratio decidendi of Air Express then they would be authority for the principle for which the Commonwealth contends. However, they do not represent the ratio decidendi of the decision. Rather, they are reasons which were delivered in dissent. Further, it is apparent that the matter which caused Mason J to dissent is the same matter which caused all of the judges to dismiss Air Express's appeal. This matter is the absence of any evidence as to what the Secretary for the Department of Transport would have done if he had not been restrained by the injunction. Only Mason J was willing to infer from the fact that Air Express had suffered loss by reason of the non-performance of the restrained act that had the injunction not been issued the Secretary would have issued the permit. All other Justices declined to take this step. 96 The Commonwealth did not take the Court to any other authority dealing with the undertaking as to damages to make good its argument. We conclude that the principle is not required by any binding or persuasive authority. That conclusion is consistent with the way in which the issue has been approached at first instance: see Coshott v Principal Strategic Options Pty Ltd [2004] FCAFC 50 at [18(e)]; Specsavers Pty Ltd v The Optical Superstore Pty Ltd (No 3) [2012] FCA504; 290 ALR 263 at [7] and [15]; and Sigma Pharmaceuticals (Australia) Pty Ltd v Wyeth [2018] FCA 1556; 136 IPR 8 at [122] and [144]. 97 Once it is accepted that both sides have adduced substantial evidence on the question of what Apotex would have done had it not been restrained, Air Express is not relevant to the question of onus. The question for the trial judge and, assuming its review jurisdiction is engaged, this Court, is whether on all the evidence, the Commonwealth has satisfied its persuasive burden of demonstrating that if it had not been restrained Apotex would have sought to list on the PBS and then distribute its products in Australia. Since this is precisely what the trial judge in fact did it follows that Ground 2(a) must fail. 98 It remains to note that Air Express continues to be relevant to Ground 2(c) which concerns the significance to be attributed to the fact that the person at Apotex who was the relevant decision maker, Dr Sherman, was not called to give evidence. After the next section, where the relevance of Dr Sherman becomes clearer, we will return to Air Express.

GROUND 2(B): WOULD APOTEX HAVE LAUNCHED IF THE INJUNCTION HAD NOT BEEN GRANTED? 99 On this part of the appeal, the parties prepared an agreed chronology which sets out the documents and ID references for the documents which the trial judge referred to as well as those in which his Honour did not refer ('the Chronology'). The item numbers in the Chronology are a convenient way of referring to the documents which are, in some cases, located in multiple locations in the evidence. 100 Ground 2 of the amended notice of appeal is as follows: The primary judge erred in concluding (at [351]) that the Commonwealth had not established that Apotex would have sought and obtained a PBS listing of its clopidogrel products from 1 April 2008 if the interlocutory injunction had not been granted. 101 This grounds asserts an error in fact finding at J [351]: In the result, I am not persuaded that Apotex Australia would have sought and obtained a PBS listing of its clopidogrel products from 1 April 2008 even if the interlocutory injunction had not been granted. It follows that the Commonwealth's claim for compensation must be dismissed. 102 Relevantly for the purposes of the present discussion, Particular (b) to Ground 2 sets out one way the Commonwealth seeks to make good this contention. It is as follows: (b) The primary judge erred by: (i) failing to have regard, including when making findings of credit (at [248]-[251], [317], [322]-[324], [338], [340]), and speculating as to what Apotex might have done had the interlocutory injunction not been granted (at [243], [251], [278]-[289], [292]-[296], [300], [308]-[317], [329]-[330], [340]-[341], [348], [350]), to: (1) relevant, contemporaneous evidence of Apotex's intention to commence selling clopidogrel products and to apply for a PBS listing if no interlocutory injunction was granted; and (2) relevant evidence from witnesses called by the Commonwealth, and (ii) failing to address, either adequately or at all, submissions of the Commonwealth in relation thereto. 103 The Commonwealth's case on Ground 2(b) is therefore that the trial judge erred in fact at J [351] because he failed to have regard to relevant contemporaneous evidence and failed to address submissions advanced at trial. In writing, the Commonwealth submitted that there was evidence not evaluated by the trial judge which contradicted his conclusion and which was sufficiently material that his Honour's failure to consider it constituted an error in the fact finding process: AS [44]. Once this evidence was brought to account, the Commonwealth submitted that the only conclusion open on the evidence was that if the injunction had not been issued, then Apotex would have sought to list its product on the PBS from 1 April 2008 (AS [47]) and would have started taking orders in January 2008: AS [69]. 104 The Commonwealth's attack is therefore upon the finding at J [351]. This should be emphasised. Ground 2(b) does not encompass omnibus attacks on other paragraphs of the reasons for judgment. Despite this, the Commonwealth took issue in its written submissions with what it alleged were three intermediate conclusions of the trial judge. These were that: (a) Dr Sherman had made only a provisional decision to launch at risk at J [340] (AS [44]); (b) it was not unlikely that the Dr Sherman would change his mind about whether to launch at risk at J [250]-[251] (AS [74]); and (c) Dr Sherman might have changed his mind after it became clear that the trial would be in April 2008 with a judgment given by August 2008 at J [289], [341] (AS [75], [77]). 105 None of these are covered by Ground 2(b). While we would accept that each of these findings was a tributary which fed into the ultimate finding at J [351], the better view may be that framed in this way, the Commonwealth's submissions lie outside the amended notice of appeal. However, Sanofi met the Commonwealth's arguments and took no point about its approach. In that circumstance, it is appropriate to accept that the Commonwealth is entitled to challenge the finding at J [351] indirectly in this way. 106 A limitation on this conclusion should, however, be noted. The language of Ground 2(b) does not encompass challenges to other findings of fact made by the trial judge where the challenge is not premised on an alleged overlooking by the trial judge of evidence or submissions. This too should be emphasised. Some of the Commonwealth's written submissions involved an extensive repetition of its case at trial with invitations to this Court to take a different view of the facts to that taken by the trial judge. Such submissions are not relevant to the error for which the Commonwealth contends. We do accept, however, that submissions of this kind would become relevant if the Court were to conclude that an error had occurred and then decided to enter upon the fact finding process itself. 107 In relation to the three alleged findings set out above, the Commonwealth's case on appeal is therefore confined to demonstrating that each was reached in circumstances where the trial judge overlooked evidence or submissions which were material. Dealing with each in turn:

1. Dr Sherman had made only a provisional decision to launch 108 The relevant finding by the trial judge is at J [340] in these terms: I do not accept that the 20 February 2007 email was an instruction on which Mr Millichamp would have considered himself entitled to act in late 2007 without reverting, either directly or indirectly, to Dr Sherman for the purpose of obtaining confirmation of that instruction and the necessary final approval from him authorising a launch at risk. The suggestion that Mr Millichamp did not need to obtain any further approval from Dr Sherman before launching at risk is, in my view, inconsistent with the email correspondence between Mr Kay and Dr Sherman and Mr Millichamp and Mr Haas. 109 The email from Dr Sherman of 20 February 2007 was in these terms: Plan is as follows: [REDACTED] 2. In May or June, we will file suit in Australia to invalidate the patent. [REDACTED] 3. We will then advise Sanofi that we will launch unless they move for and obtain an injunction, in which case they will have to give an undertaking for our damages. 4. If they do not give an undertaking for our damages and do not get an injunction, we will launch. 110 The Commonwealth's case was that this email demonstrated that Dr Sherman had decided to launch on 20 February 2007 and that Mr Millichamp thereafter needed no further approval from him to launch at risk. As will be seen, Mr Millichamp gave evidence to this effect which the trial judge rejected. 111 The Commonwealth's written submission about J [340] proceeded down two paths. First, it submitted that in making the finding, the trial judge had erroneously taken into account the opinions of Mr Kay rather than those of Dr Sherman; and secondly, that his Honour had overlooked material evidence.

First path: An erroneous reliance on the opinions of Mr Kay rather than those of Dr Sherman 112 As to the first path, the Commonwealth began by submitting in writing that the trial judge wrongly had regard to Mr Kay's opinions at J [340] because the decision maker was not Mr Kay but rather Dr Sherman: AS [72]. There is nothing in this submission. There was no debate before the trial judge that the decision maker was Dr Sherman and it is clear that the trial judge shared this view since he found at J [199] that Dr Sherman was 'the ultimate controller of the Apotex Group' and at J [339] that he was 'the key decision-maker'. The finding at J [339] was located in the section of the trial judge's reasons which dealt with the question of whether Apotex would have launched at risk had it not been restrained, that is to say, the issue presently under consideration. To the extent that the Commonwealth's submission carries with it an insinuation that the trial judge was confused as to who the decision maker actually was, this should be rejected. His Honour was not. 113 The Commonwealth identifies in its written submissions that the paragraph at which the trial judge took into account the opinions of Mr Kay rather than those of Dr Sherman is J [340] (set out above). The first point about J [340], which is obvious, is that the trial judge did not say that he was taking into account the opinions of Mr Kay over those of Dr Sherman or even that he was taking into account the opinions of Mr Kay at all. What he said was that Mr Millichamp's view that he did not need to obtain any further approval from Dr Sherman was inconsistent with the emails which had passed between Mr Kay and Dr Sherman. This is a different proposition. 114 The second point is that the emails were identified by the trial judge at J [241]-[244]. These were emails passing between Dr Sherman and Mr Kay on 25 June 2007 and 27 June 2007. What they showed was that at those dates, Mr Kay needed Dr Sherman to make a decision as to whether there should be a launch at risk in Australia. In the first email, Mr Kay said to Dr Sherman: 'Would you please re-confirm or otherwise our approach in Australia', and then at the end of the email: 'Do you want us to continue to launch and to invest in the revocation proceedings?' In the second email, Mr Kay asked Dr Sherman whether he had yet made a decision on the issue. The trial judge did not therefore act on Mr Kay's opinion rather than Dr Sherman's. Rather, his Honour thought that Mr Millichamp's view that nothing further was necessary from Dr Sherman did not comport with the emails passing between Dr Sherman and Mr Kay. Once that is appreciated, the Commonwealth's written submission that the trial judge had acted on the opinions of Mr Kay rather than those of Dr Sherman must be rejected because it is premised upon an inaccurate reading of the trial judge's reasons. 115 The third point is that, even if that were not so, the submission lies outside the Commonwealth's amended notice of appeal which contains no reference to such an error having been made by the trial judge. In particular, the Commonwealth's submission is not an argument that the trial judge overlooked evidence or a submission but, rather, that his Honour had misunderstood the evidence. As will be evident from the above, the trial judge did neither. The first path of the Commonwealth's written submissions is therefore foreclosed.

Second path: A failure to advert to relevant evidence 116 Turning then to the second path by which the Commonwealth seeks to impugn the trial judge's conclusion that Dr Sherman had made only a defeasible decision, the Commonwealth identifies the following five matters as not having been referred to by the trial judge (AS [72]):

(a) An email of 28 June 2007 117 The first item of overlooked evidence is said to be an email of 28 June 2007 which, according to the Commonwealth's written submissions, 'recorded Dr Sherman's "instructions" after being asked to reconfirm them by Mr Kay'. This is item 29 of the Chronology which we discuss later at [137]-[143]. For reasons given below, this email is not one which the trial judge ought to have considered. We should say that we do not accept the premise of the Commonwealth's argument that merely because the trial judge did not refer to an item of evidence it follows that the evidence was not considered. An inference which is also open is that the trial judge considered the evidence but did not regard it as material. In the case of this email that is the inference we would draw.

(b) Apotex's public announcements 118 Secondly, the Commonwealth submitted that the trial judge had overlooked Apotex's public communications to its customers and pharmacists that it would soon be launching its clopidogrel products: AS [72]. In fact, the trial judge dealt in some detail with Apotex's public communications and the evidence relating to it. At J [260], his Honour found that Mr Millichamp provided a letter to be sent to customers to his sales team on 17 August 2007: item 45. In the same paragraph, the trial judge set out the full text of this letter together with a finding that the letter had in fact been sent. The terms of the letter included the statement that 'the decision whether to launch these products will be delayed until the outcome of that application has been determined'. At J [288], the trial judge analysed the text of this letter to make the point, correctly in our view, that 'In its written communications with pharmacists, Apotex Australia had not committed itself to launching in the event no interlocutory injunction was granted'. The submission that the trial judge overlooked Apotex's public communications to its customers and pharmacists is therefore incorrect. 119 The Commonwealth then collaterally submitted in writing (AS [65]) that the trial judge had failed to advert to an email chain dated 21 August 2007 apparently touching on the question of Apotex's public communications to its customers. This email chain consisted of various emails between several employees of Sanofi about Apotex's public announcement that day. The last of these is from Sanofi's General Manager, Mr Moulding, and contains this statement: 'This seems to be very clear evidence of intention to launch and the urgency for an interim injunction': item 48. 120 About this email chain the Commonwealth began by submitting that the announcement was a 'critical step' since it was a public commitment to the market that Apotex had made the decision to launch at risk if not enjoined: AS [65]. However, Mr Moulding's email does not make good this point. It shows only that Sanofi was aware of the announcement. Since this has never been in dispute, it is difficult to see how Mr Moulding's email furthers this submission. The only aspect of Mr Moulding's email that is useful from the Commonwealth's perspective is that the trial judge did not refer to it in his reasons (whereas he plainly did refer to the announcement). In any event, making the assumption that that submission was really intended as an observation about the terms of the announcement, as the trial judge correctly found at J [288], this was not in fact what the announcement said at all. What it actually said was that if Sanofi applied for an interlocutory injunction 'then the decision whether to launch these products will be delayed until the outcome of that application has been determined': J [260]. The terms of the letter directly contradict the Commonwealth's contention that no further decision remained to be made. Using the language of the Commonwealth's own submission, it was 'a public commitment' that a decision whether to launch the product would be made only once the outcome of the application was known. Thus one may assume to be correct the Commonwealth's submission that there was 'evidence that those kinds of public statements to customers are not lightly made' (although the submission did not identify what this evidence was or where in the appeal papers it might be found) because it only serves to underscore the significance of the fact that what Apotex actually told the market was that the decision whether to launch would be made once the outcome of the application was known. Far from assisting the Commonwealth, the terms of the public announcement are inconsistent with its case. 121 The Commonwealth then sought to rely directly on the terms of Mr Moulding's email: 'This seems to be very clear evidence of intention to launch and the urgency for an interim injunction'. However, the Commonwealth's written submissions did not explain why this evidence was material to the trial judge's conclusion at J [340] that he did not accept that the email from Dr Sherman dated 20 February 2007 was final approval for a launch at risk (recalling that this is the finding presently under challenge). Similarly, unaided by a submission as to why Mr Moulding's statement was material to J [351], we do not see how it could be. The fact that the General Manager of Sanofi thought that Apotex's public communication was clear evidence of an intention to launch is not rationally capable of bearing upon the state of Dr Sherman's mind whereas the terms of the actual announcement made by the company he controlled are. Indeed, had objection been taken to the tender of Mr Moulding's email on the grounds of relevance, it seems to us that its tender would have been rejected. The failure of the trial judge to advert to irrelevant evidence is not a matter warranting the intervention of this Court.

(c) Apotex's listing application 122 It was then said that the trial judge had not taken into account Apotex's application for PBS listing made on 1 September 2007 and the detailed arrangements for supply it had made in the event that it was not enjoined: item 51. As to the application for listing, again this is wrong. The trial judge dealt with the listing application at J [261]: Apotex Australia submitted an application to list its clopidogrel products on the PBS with effect from 1 December 2007. The application was dated 1 September 2007 but was not received by PBAC until 4 September 2007. 123 At J [262]-[263], his Honour went on to make further findings about the PBS application to the effect that it could have been withdrawn until 12 October 2007. The submission that the trial judge had overlooked this matter must therefore be rejected. 124 As to the suggestion that the trial judge had not taken into account the 'detailed arrangements for supply' it made in the event that it was not enjoined, the Commonwealth's written submissions did not identify what these arrangements were: AS [72]. At [63] of its written submissions, the Commonwealth did set out document ID codes for a number of documents which it said were related to the supply issue without indicating the nature of the documents or the dates they bore or whether they had been referred to by the trial judge or not. By comparing the document ID codes given for these documents with the Appeal Book Part C Index, it is possible to determine what these documents are and the date of each. When this information is then cross-referenced with the Chronology, it becomes possible to ascertain that most of them were actually dealt with by the trial judge. However, ten were not. These are items 26, 30 (including 30A), 31, 32, 37A, 38, 39, 40, 56 and part of item 65 in the Chronology. Unpacking AS [63] in this way, it then appears that the Commonwealth's submission is that the trial judge failed to consider the following documents which are said to relate to 'detailed arrangements for supply': (a) an email from Mr Haas to Mr Millichamp sent at 12.00 am on 28 June 2007 (item 26); (b) an email from Dr Desai to a number of Apotex Canada employees sent at 5.19 am on 29 June 2007 and Mr Haas' email forwarding to Mr Millichamp the email from Dr Desai, sent at 5.30 am on 29 June 2007 (items 30 and 30A); (c) an email from Mr Millichamp to Mr Haas sent at 8.13 pm on 29 June 2007 (item 31); (d) a 'Second Transfer Pricing Agreement' between Apotex and Apotex Canada dated 10 July 2007 (item 32); (e) an email from Mr Millichamp to Mr Kay, Dr Sherman and Mr Haas (amongst others) sent at 12.57 am on 9 August 2007 (item 37A); (f) an email from Mr Haas to Mr Millichamp in response to item 37A sent at 8.47 pm on 9 August 2007 (item 38); (g) an email from Mr Kay and an email from Mr Caccamo sent at 1.29 am and 1.30 am respectively on 10 August 2007 responding to an email from Mr Millichamp (item 39); (h) an email from Mr Kay to Mr Millichamp sent at 1.28 am on 10 August 2007 (item 40); (i) an email from Mr Haas to Mr Millichamp sent on 5 September 2007 (item 56); and (j) an email from Mr Haas to Dr Sherman sent at 2.37 am on 15 September 2007 (item 65). 125 Leaving aside for now the position of the transfer pricing agreement (item 32), each of these emails was part of a much larger series of communications leading up to the application for the interlocutory injunction. The trial judge dealt with many of these communications although not all of them. At this point, it is necessary to recall the forensic relevance which the Commonwealth ascribes to these supply arrangements. It submits that by failing to consider the evidence concerning the supply arrangements, the trial judge erred in concluding at J [340] that Dr Sherman's email of 20 February 2007 was not one which Mr Millichamp would have considered himself as entitled to act upon without first reverting to Dr Sherman. In effect, the contention is that the alleged fact that Apotex had made detailed arrangements for supply was a matter which the trial judge had to consider because it made more likely the hypothesis that Dr Sherman's email was final in nature and that Mr Millichamp did not thereafter need to revert to him for further approval. 126 However, consideration of the 'detailed arrangements for supply' cannot be divorced from the trial judge's assessment of the larger set of emails and, in particular, the conclusions his Honour drew from them. In what follows, we therefore assay the larger set of emails to provide that context, dealing with each of the allegedly overlooked emails as we proceed. 127 Mr Kay emailed Dr Sherman on 25 June 2007 asking him to 're-confirm or otherwise our approach in Australia': item 20. In this email, parts of which were redacted for privilege, Mr Kay explained to Dr Sherman that a PBS listing would trigger a 12.5% price reduction in the product and that the cost of the revocation proceedings would be likely be up to $2 million. He concluded that by saying 'Do you want us to continue to launch and to invest in the revocation proceedings?' This email was considered by the trial judge at J [241]-[242]. 128 This email triggered a cascade of emails between Dr Sherman, Mr Kay and Mr Millichamp on 26 and 27 June 2007. Most of these are redacted in part or in whole as a result of claims made by Apotex for legal professional privilege. One of them was sent by Mr Kay to Dr Sherman at 11.54 am on 26 June 2007: item 21. Mr Kay asked Dr Sherman to 'advise when you have decided whether to pursue revocation of the … patent, and if you wish us to move to launch at risk'. The trial judge attached significance to this email because it reflected an understanding on Mr Kay's part that it was necessary, or at least desirable, to obtain confirmation from Dr Sherman as to his intention with respect to Australia and whether or not Apotex should seek to revoke the patent and launch at risk: J [243]. 129 That email was followed by further email exchanges. Two of these were emails from Dr Sherman sent on 27 June 2007 at 4.44 am and 5.12 am. Their contents are unknown because they have been redacted for privilege. However, Mr Millichamp forwarded Dr Sherman's emails to a number of other persons at 11.23 am on 27 June 2007: item 24. The forwarded emails are redacted for privilege however Mr Millichamp's email contains the opening line, 'FYI - Game on !!!', which the trial judge noted at J [245]. Although the Commonwealth made much of this email, it begs the question of what the game was. There are at least two possibilities. One 'game' was to seek to launch at risk. Another 'game' was to signal to the market that Apotex was going to launch in order to goad Sanofi into seeking an interlocutory injunction and proffering an undertaking as to damages. Which of these was the 'game' to which Mr Millichamp was referring rather turns on the contents of Dr Sherman's two emails which are unknown. 130 The trial judge was much more impressed by Mr Millichamp's next email which was sent two hours later at 1.59 pm to Mr Haas: item 25. Mr Haas was involved with supply: T52.4-5. Most of the email is redacted but it contains a statement which is inconsistent with the Commonwealth's contention that Dr Sherman had made a final decision to launch at risk on 20 February 2017. What Mr Millichamp told Mr Haas was this: [REDACTED]. If we are successful in avoiding an injunction we will plan to launch subject to Barry's further advice/approval. If anything changes I'll let you know. 131 It was relevant in the trial judge's view because Mr Millichamp had testified before him that he understood from Dr Sherman's email of 20 February 2007 that no further approval from Dr Sherman was necessary. Mr Millichamp was cross-examined about the email and the trial judge was not impressed with his evidence about it, preferring the email: J [248]-[251]. This is the subject of a separate challenge to which we will return in due course. 132 Mr Haas replied to this email the next day, 28 June 2007, thanking Mr Millichamp for his update, asking him for an estimated launch date and informing him that Mr Haas would now 'initiate blister tooling activities to determine timing for providing you with pack count desired for AUS market … Coldform tooling leadtimes can be as long as 6 months': item 26. Item 26 is one of the emails which the Commonwealth now identifies as evidence that the trial judge overlooked relating to Apotex's 'detailed arrangements for supply'. As noted above, the Commonwealth did not develop any explanation of what these detailed arrangements were other than to assert that they were detailed and that they were evidenced in the emails presently under discussion (and the transfer pricing agreement). Each of these documents should therefore be assessed through two lenses. The first lens is whether the particular document by itself could be material to the overall finding at J [340] that Dr Sherman's email of 20 February 2007 was not an instruction that Mr Millichamp would have considered himself entitled to act upon without first reverting to Dr Sherman. The second lens consists of considering all of the suggested detailed arrangements for supply (including this particular email) which had been made and asking whether, viewed as a whole, they were a corpus of evidence which was material to the finding at J [340]. 133 In the case of the email of 28 June 2007 (item 26), we do not think that it was, by itself, material to the trial judge's finding at J [340]. That Mr Haas told Mr Millichamp that he would initiate blister tooling activities to determine the time it would take to provide the pack count contemplated by Mr Millichamp, by itself, proves nothing relating to Dr Sherman's email of 20 February 2007. We will return to consider this email, again, through the second lens and, in that context, it is convenient to refer to this email as the 'First Supply Arrangement'. 134 On the same day, that is 28 June 2007, a Ms McTavish replied to Mr Millichamp's 'Game on !!!' email asking, 'So have they now been notified?': item 27. It will be observed that this email is consistent with both of the 'games' referred to above. Mr Millichamp responded the same day saying 'Karen. Not yet.' (item 28) and the same may be said of it. Neither of these emails was referred to by the trial judge but, in our opinion, they could not be material to the conclusion he reached at J [340] (a strictly unnecessary observation since this email is not alleged to be one of the 'detailed arrangements for supply'). We would accept that they demonstrate that the 'game', whatever it was, included notifying Sanofi of a launch proposal. But this does not prove anything about Dr Sherman's state of mind. The most which can be extracted from this exchange is that Apotex was intending to notify Sanofi of its intention to launch. So viewed, however, this is of little importance. There was plenty of other evidence that the trial judge did refer to that indicated that this was, indeed, Apotex's plan. For example, Mr Kay told Dr Sherman on 8 August 2007 that '[t]he current plan is to put Sanofi on notice of our intention to launch and so invite them to seek an interlocutory against us launching' which the trial judge referred to at J [252]. 135 Then there is the undeniable fact that Apotex did put Sanofi on notice of its intention to launch when the letter written by Mr Millichamp was sent to Apotex's customers and pharmacists. As we have noted above, the trial judge set that letter out at J [260]. Just as important, however, are the terms of that letter which, as we have previously mentioned, contained the statement that 'the decision whether to launch these products will be delayed until the outcome of that application'. This was consistent with what Mr Millichamp had said to Mr Haas in his email of 27 June 2007: 'If we are successful in avoiding an injunction we will plan to launch subject to Barry's further advice/approval'. In any event, the point for present purposes is that the trial judge dealt with the fact that Apotex signalled its intention to launch to Sanofi. The two emails in items 27 and 28 of the Chronology, although not dealt with by the trial judge, add nothing to that picture. We have dealt with these two emails, although they are not alleged to form part of the 'detailed arrangements for supply', because they do appear in the Chronology as documents not referred to by the trial judge. Whilst we do not accept that the Commonwealth's case on appeal was broadened beyond its written submissions to include what was contained in the Chronology, we have dealt with them nonetheless to forestall any further debate about the matter. In any event, no error was made by the trial judge in not referring to them. 136 The next email is from Mr Chalk to Mr Hughes and was sent on 28 June 2007: item 28A. Although it is included in the list of the documents which the Commonwealth submits was overlooked by the trial judge, its entire contents are redacted. In that circumstance, it is difficult to see how it could have been material to his Honour's finding at J [340]. Further, item 28A is not one of the documents identified as forming part of the 'detailed arrangements for supply'. 137 On 28 June 2007 at 10.28 pm Mr Millichamp sent an email to Mr Haas enclosing the redacted email sent by Mr Chalk to Mr Hughes (item 28A): item 29. This email recites the launch plan previously mentioned by Mr Millichamp including the statement: 'If we are successful in defending our position vs Sanofi and an injunction is not granted by the courts then we will launch'. In relation to the topic of arranging supply, the email also said this: 5) Therefore if we are successful and are able to launch we will need product (blister packed by Apotex) in Australia ideally end October but latest Mid November. 6) Initial quantities and follow up / replenishment orders etc can all be discussed and arranged in a way that is best for you bearing in mind timings and blister constraints. Could you please continue to work with Marnie, Paddy and Karen etc on the packaging options. Please confirm when you would like purchase orders raised. We are not sure that we can launch yet so need to know if we should place orders in anticipation or should wait. Barry has made it clear that he does not want to waste money on launch activities until we are clear on what we can do. At our next demand meeting we should talk through the various options please. (emphasis added) 138 This email was not referred to by the trial judge. Again, it is not identified as part of the detailed arrangements for supply but it does appear in the Chronology as a document not referred to by the trial judge. For the reasons already given, we will consider the significance of the fact that it was not referred to by the trial judge. There are three aspects of this email to note. The first concerns Mr Millichamp's observation that if Apotex successfully resisted the injunction application 'then we will launch'. On the face of it, this is a clear statement from Mr Millichamp that Apotex would launch if the injunction was refused. The relevance of this evidence is that it is consistent with Mr Millichamp's claim in his oral evidence that he understood that if there was no injunction, then Apotex would launch at risk without any further approval from Dr Sherman. As such, it was material which could have borne upon the trial judge's conclusion that this aspect of Mr Millichamp's evidence should not be accepted because it was contradicted by his 27 June 2007 email ('If we are successful in avoiding an injunction we will plan to launch subject to Barry's further advice/approval': item 25). Put another way, Mr Millichamp's oral evidence might have been inconsistent with his 27 June 2007 email (item 25) but it was relevant in assessing his credit to know that what he was saying was consistent with his 28 June 2007 email: item 29. 139 However, framed this way we do not accept the argument (assuming the argument was put). The trial judge accepted that Mr Millichamp was keen to launch at risk. He found as much at J [290]: I do not think there is any doubt that Mr Millichamp was eager to launch at risk. He believed that if Apotex Australia could get its clopidogrel product on the market in early 2008 followed by a PBS listing on 1 April 2008 then it would lead to a substantial increase in sales. However, his evidence shows that he was heavily focused on the positive consequences of getting to market quickly, but much less focused on the negative consequences of doing so, and the potential financial consequences should the challenge to the validity of the Patent fail. 140 His Honour referred to evidence where Mr Millichamp proceeded on the basis that if the injunction were not granted then Apotex would launch. For example, the trial judge referred at J [267] to an email sent by Mr Millichamp to Mr Kay on the first return of the injunction on 13 September 2007: 'Our plan is to launch to customers (take orders) immediately, assuming that we are free to sell, and then supply product in early 2008 for an April PBS listing'. And the same theme appears in an email referred to by the trial judge at J [237] sent by Mr Millichamp to Mr Kay on 25 June 2007: Our expected timetable of events going forward is as follows: * Registration (ARTG) approval date - Expected towards the end of July, latest early August (Please note however that the TGA have been surprising us and approving our dossiers earlier than expected. Gabapentin 600mg and 800mg tablets are likely to be approved 3 months ahead of our expected timing). * We propose launching to the trade, by taking orders, on the day that we appear on the ARTG i.e. have regulatory approval. * We propose notifying Sanofi of our planned launch by applying to revoke their patent either on the same day that we launch or a few days earlier. The idea is to notify Sanofi as soon as we know for sure that we can launch, i.e. when we know that TGA approval will be granted but not give Sanofi time to try and block our initial launch efforts. * We will submit our application for PBS listing once we are sure that no injunction will be imposed by the courts (PBS listing is granted roughly three months after initial application).We will, of course, have to get to court before we know this outcome. If we apply for PBS listing as soon as we launch we will trigger the 12.5% price drop and Sanofi will argue that pricing has been dropped and therefore their damages cannot be readily measured, will cause irreparable harm and the decision cannot be reversed etc. When we get to court we will need to let the judge know that we plan to PBS list. In court the idea, in relation to the 12.5% drop and further launch activities, is that we will argue that damages are an adequate remedy, give undertakings that we will record all sales, and provide a bank guarantee in order to pay costs and damages in the event that we are not successful at final trial. We will have to give a bank guarantee that is effective in Australia, but supported by Apotex, as GenRx have no assets or profits at this stage in this jurisdiction to prove that we can pay out in the event of a Sanofi success. * It is difficult to forecast the amount that we would have to put up as a bank guarantee but we estimate that it could be in the range of 50 -70 Million AUD (for one year's damages) depending on what Sanofi argues in court. * We would not propose continuing with launch activities i.e. takings orders and supplying stock if we do not gain PBS listing. Our sales will be very low, virtually zero without PBS reimbursement. In this case we would continue with our patent revocation case but not apply for PBS listing and so not trigger the 12.5% price reduction. We will of course be notifying our customers when we launch that this is a high risk launch and subject to legal action so that they are not misled if we are not able to deliver stock for a considerable time. Andrew, this is what we are planning but if there is a different approach that is proposed from your side we will be happy to take that advice and recommendation. 141 Thus the trial judge was therefore fully aware that Mr Millichamp was keen to launch and had made statements to the effect that if the injunction were refused, Apotex would launch. His Honour's answer to these statements was that Mr Millichamp's enthusiasm for launch was not an answer to the proposition that Dr Sherman's final approval before a launch would be necessary. Mr Millichamp had said that in his email of 27 June 2007: item 25. Thus, whilst it is true that the trial judge did not refer to this email of 28 June 2007 (item 29), we are not satisfied that it adds anything to the other emails from Mr Millichamp to the same effect referred to at J [237] and J [267]. We would infer that the trial judge did not refer to this evidence because he correctly did not think it was material. 142 The second aspect of this email is that it can be inferred from it that Mr Millichamp was aware that Dr Sherman did not wish to waste money on launch activities until the outcome of the injunction was known. Mr Millichamp was advancing the launch process as far as he could subject to that constraint. In our view, this does not assist the Commonwealth and, in fact, tells against it. The statement 'Barry has made it clear that he does not want to waste money on launch activities until we are clear on what we can do' is consistent with the idea that the person who would make the decision was Dr Sherman. It was, after all, Dr Sherman who did not wish to waste money. The statement is also consistent with Apotex's public announcement that the decision whether to launch would be made after the outcome of the application for the interlocutory injunction was known. The email sits uncomfortably with the notion that Dr Sherman had decided in irrevocable terms in his 20 February 2007 email to launch at risk and together with it, the idea that Mr Millichamp did not need to seek any further advice or approval from Dr Sherman. This is so because it suggests that Dr Sherman remained engaged in the launch process in a way which makes little sense if he had made his final decision on 20 February 2007. On the other hand, Mr Millichamp's email of 28 June 2007 comports with his own words in his email of 27 June 2007 (item 25): 'If we are successful in avoiding an injunction we will plan to launch subject to Barry's further advice/approval'. As such, we do not think that the trial judge erred in failing to advert to this email (item 29) (assuming such an argument was put). In our view, had he done so it would have provided another reason for confirming his Honour's conclusion at J [340]. 143 The third aspect to note relates to the Commonwealth's written submission at AS [72] as to the failure of the trial judge to advert to the 'detailed arrangements for supply if not enjoined'. Whilst we do not accept that the email of 28 June 2007 (item 29), by itself, is something which the trial judge ought to have adverted to, we do accept that it may have a secondary relevance to the Commonwealth's contention inasmuch as it is one of the emails which identifies a supply arrangement. As mentioned above, the Commonwealth did not advance this email as one of the items of evidence upon which it relied to make good the existence of the suggested 'detailed arrangements for supply' although it does appear in the Chronology as one of the documents said not have been considered by the trial judge. Whilst the document was not therefore formally advanced, we nevertheless consider it prudent to take it into account as one of the detailed supply arrangements. The supply arrangements referred to in this email consist of: (a) an aim to launch in October or November with blister packs; (b) a commitment to discuss the various options at the next meeting including initial quantities and follow up/replenishment orders; and (c) an instruction by Mr Millichamp to other employees to continue working on packaging options. For convenience we will refer to this as the 'Second Supply Arrangement'. We will return to summarise all these supply arrangements when each has been identified. 144 Mr Haas responded to Mr Millichamp's email on 28 June 2007 (item 29AA) thanking him for the update and saying: Roger, Thank you for the update. As indicated previously, I will set up a blister meeting next week with packaging engineering to firm up blister tooling leadtimes to be able to provide pack size of 28's. Currently we only have 10's blister tooling in house. Once I have details I will communicate back to Marnie, Paddy, Karen and yourself. Strategy from my side is to get everything in place so that we can proceed with commercial manufacture immediately once given the go-ahead. I will revert a.s.a.p. regarding the purchase orders once I have confirmed batch sizes. I will create a slide in the Demand presentation to address Clopidogrel and suggest that those from your side do the same to facilitate discussion and issue resolution. Best regards Stephen 145 This email was not referred to by the trial judge. We do not consider that this email adds anything to Mr Millichamp's email of 28 June 2007 in terms of throwing light on the state of Dr Sherman's mind and as such is not material. As in the case of Mr Millichamp's email to which it responds, this email does however constitute a supply arrangement, namely, that Mr Haas would set up a meeting with 'packaging engineering' to firm up blister tooling lead times (the 'Third Supply Arrangement'). 146 Mr Haas then replied a second time to Mr Millichamp's email of 28 June 2007 (item 29) on the same day: item 29A. Recalling that in his email Mr Millichamp had said, in point (5), that if Apotex successfully resisted the injunction 'and are able to launch we will need product (blister packed by Apotex) in Australia ideally end October but latest Mid November'. Mr Haas now replied: 'One further question … Is Barry in the loop the strategy below (potentially launch end October but latest Mid November)?' When the two emails are put next to each other, it can be seen that what Mr Haas was asking Mr Millichamp was whether Dr Sherman was aware of the proposal to launch at the end of October or mid-November. This email does not, therefore, throw any light on what Dr Sherman's thoughts were on the question of launching at risk but rather is an inquiry about Dr Sherman's thinking on the proposed October or November launch date. As such, we do not accept that this email was material to the trial judge's finding at J [340]. 147 Mr Haas' question to Mr Millichamp in terms invited a response. Before Mr Millichamp could give that response, Mr Haas received an email which had not been addressed to him and which must have been forwarded to him by someone else. The email was from Dr Desai, an employee of Apotex Canada, to a number of senior personnel at Apotex Canada but did not include Mr Millichamp. It was sent at 5.19 am on 29 June 2007: item 30A. This email was not referred to by the trial judge. It said this: I just want to clarify our position for Australia. 1. We will be notifying Sanofi that we intend to launch in Aus with the bisulphate salt. 2. We expect Sanofi to sue us. 3. We will litigate the patent in court 4. We will not be making any launch stocks of the bisulphate salt even though we will expect approval in the next couple of months. 5. We will develop the Hbr salt for Aus and file ASAP. Please let me know if you require any further clarification. Jeremy 148 The key point concerns the reference in point (4) to the bisulfate salt. The trial judge found that this salt was the active ingredient in the clopidogrel product launched in the US (J [206]) and the product attempted to be launched in Canada: J [201]. It was also to be the active ingredient in Apotex's Australian clopidogrel product: J [101]. Ultimately, the final injunction granted by Gyles J was in respect of the bisulfate salt: J [105]. As such, point (4) is a statement that Apotex would not be making any launch stock of the active ingredient in the product which was to be launched. 149 This of course does not entail that there was no stock of the bisulfate salt available. As later emails hint at, there was stock on hand of the bisulfate salt in the US. The observation that Apotex would not be making any bilsulfate stock is in line with Mr Millichamp's statement in his email of 28 June 2007 (item 29) that 'Barry … does not want to waste money on launch activities until we are clear on what we can do'. The point for present purposes is that as at 29 June 2008 when Dr Desai sent this email, the launch plan did not yet contain an element which explained where the bilsulfate stock was going to come from. We do not accept that Dr Desai's email advances the Commonwealth's case because it throws no light on what Dr Sherman was thinking beyond what Mr Millichamp's earlier email (which, for reasons set out above, we have found not to be material) demonstrates, namely, Dr Sherman's aversion to potentially wasteful expenditure. As such, we do not accept that this email was material in a direct sense to the finding at J [340]. It does have a secondary significance as one of the supply arrangements because it shows that the arrangements did not at this time include any identification of where the bisulfate salt was going to come from. It is convenient in that context to refer to the decision that the bilsulfate salt would not be manufactured for the launch as the 'Fourth Supply Arrangement'. 150 It will be recalled that Mr Haas had already sent an email to Mr Millichamp querying the October or November launch date and asking whether Dr Sherman was 'in the loop' in relation to it: item 29A. Before Mr Millichamp replied to that email, Mr Haas sent to Mr Millichamp the email from Dr Desai (referring to the fact that no bisulfate salt would be made for the launch). Mr Haas' email was sent at 5.30 am on 29 June 2007: item 30. This email was not referred to by the trial judge. Mr Haas noted Dr Desai's observation that Apotex would not be making any of the bisulfate salt for the launch and then said to Mr Millichamp: 'This does not align with what you had provided to me earlier today. Can you clarify to ensure everyone (R&D, Execs, Demand etc) is on the same page'. What Mr Haas was referring to here was Mr Millichamp's email sent at 10.28 pm on 28 June 2007 (item 29) at point (5) where Mr Millichamp had said that Apotex would 'need product (blister packed by Apotex) in Australia ideally end October but latest Mid November'. The tension Mr Haas perceived was between Dr Desai's statement that no launch stock of the bisulfate salt would be made and Mr Millichamp's statement that he wanted blister packed products for an October or November launch. 151 We do not accept that this email (item 30) was material to which the trial judge had to refer. It threw no light on either Mr Millichamp's state of mind or, for that matter, Dr Sherman's. 152 Mr Millichamp now had three emails in his inbox from Mr Haas to which he had not yet replied. All concerned his statement from the day before that he wished to see blister pack products for an October or November launch. The first was Mr Haas indicating that he would set up a blister meeting in the next week with packaging engineering: item 29AA. The second asked whether Dr Sherman knew about the October/November launch strategy: item 29A. The third concerned Dr Desai's statement that no bisulfate salt would be manufactured for the launch: item 30. 153 At 8.13 pm on 29 June 2007, Mr Millichamp responded to Mr Haas: item 31. This email was not referred to by the trial judge. It is worth setting out his response because, as will be seen, it was unresponsive. It is possible that this may be because of its partial redaction for privilege. The available unredacted text is as follows: Hello Stephen, [REDACTED] I was probably not clear enough in my mail to you. We are not requesting launch stocks to be made before we commence litigation (in the next month or so) and start selling to customers as we have done with previous launches e.g. Perindopril and Carvedilol. [REDACTED] [REDACTED] The scenario I referred to was in the event that we avoid the imposition of an injunction granted in favour of Sanofi. We will be free to sell until final trial, and after, if we are successful. In the instance that we are successful we will need stock ready for a December PBS listing. [REDACTED] [REDACTED]. Thus if we are successful we will go ahead and launch (the bisulphate salt). If we are not successful in avoiding an injunction then the matter will go to final trial which could be a year or so (or longer) away. That's why Barry does not want us to make stock right now and incur unnecessary expense. [REDACTED] Best regards, Roger 154 Contrary to Mr Gleeson SC's submission at T56.9-12, the email did not appear to answer any of Mr Haas' questions. It did not, for example, say whether Dr Sherman was aware of the proposed October/November launch date. Nor did it even refer to those launch dates. Instead, what Mr Millichamp now said was that stock had to be ready for a December 2007 PBS listing. Mr Millichamp also did not respond to Dr Desai's statement that no bisulfate salt would be manufactured for the launch. 155 Again, this email is consistent with other statements made by Mr Millichamp that the proposal was to launch if the injunction were refused (i.e. J [237] and J [267]). As we have explained, the trial judge was aware that this was Mr Millichamp's attitude but was unpersuaded that it could undo the effect of Mr Millichamp's 27 June 2007 email (item 25) where he had referred to the need for Dr Sherman's 'further advice/approval'. To that we would add that it was inconsistent with the terms of Apotex's public announcement that the decision whether to launch would be made once the outcome of the interlocutory injunction application was known. As such we do not accept that, in assessing Mr Millichamp's oral evidence to the effect that he believed that he could launch at risk without any further approval from Dr Sherman, this email added to the picture in any way which the trial judge had not already considered. The inference we would draw is that the trial judge did not refer to this evidence because his Honour correctly thought that it did not matter. As such we do not accept that the trial judge erred in not referring to this evidence. 156 In terms of the secondary significance of the four emails just discussed, they contain nothing in terms about supply preparations beyond matters already sufficiently covered by the Second, Third and Fourth Supply Arrangements. 157 The next emails relating to the arrangements for supply which the Commonwealth says the trial judge overlooked are two emails sent at 1.29 am and 1.30 am on 10 August 2007: item 39. Both emails are in reply to an email from Mr Millichamp. One reply is from Mr Kay, the other from a Mr Caccamo who was also involved in supply. In the interests of clarity, it is useful to note that these two emails are the subject of two separate contentions by the Commonwealth. One concerns the Commonwealth's submission that the trial judge had failed to consider the detailed arrangements for supply which had been made. They also feature, however, as an element in the Commonwealth's submission that the trial judge failed to deal with 'the full email chain' of 8-10 August 2007 to which we will in due course come. The present discussion concerns the first point. 158 For present purposes, it is useful to know that both emails are in response to a report from Mr Millichamp. Mr Kay's response says: Roger Thanks for the comprehensive update OK by me Regards Andrew 159 Mr Caccamo's email says: We have aligned our launch plans accordingly. 160 These two emails could have no impact on the trial judge's conclusion that he did not accept that Mr Millichamp thought that he could launch at risk without any further approval from Dr Sherman. As such, the fact that the trial judge did not refer to them is without significance. It should also be noted that although nominated in the Commonwealth's written submissions as one of the emails evidencing the detailed arrangements for supply which had been made (AS [63]), neither email contains such an arrangement. 161 In his oral submissions, Mr Gleeson also took the Court to part of this email chain which was not listed in the Chronology as one of matters which had been overlooked by the trial judge. Mr Gleeson drew the Court's attention to the fact that the email chain had been commenced by Mr Haas at 10.36 am on 8 August 2007. Mr Gleeson noted that Mr Haas had noted a need for 14.4 million doses. Mr Gleeson submitted, 'What your Honours will note from those bullet points is that it was to be a heavy front-end load': T70.29-30. However, to the extent that this suggested that Apotex had actually put in place such an arrangement, it is clear from the same email chain that there was to be no heavy frontload all. Mr Kay said in response at 10.58 am on the same day that it was expected that Sanofi would obtain an injunction and that Apotex 'are thus not planning to launch at this stage. I'm not sure why this information has not reached you': item 35. 162 The next two emails the Commonwealth relies upon to identify the existence of detailed supply arrangements are Mr Millichamp's email to Mr Haas, Dr Sherman and Mr Kay (amongst others) of 9 August 2007 (item 37A) and Mr Haas' response (item 38). We deal in more detail with these emails below at [183]-[185]. As that examination shows, Mr Millichamp's email contains no evidence of a supply arrangement that was in place. Indeed, quite to the contrary what it says is: 'We were not expecting any stock to be made or packed until we know the outcome of the hearing for interlocutory relief which should be approximately 2 to 6 weeks after we file for revocation'. Mr Haas' response ('thank you for providing further clarity') likewise does not comport with the Commonwealth's submission. 163 The next email the Commonwealth submits contains evidence of detailed supply arrangements overlooked by the trial judge is Mr Kay's email to Mr Millichamp of 10 August 2007: item 40. Like Mr Kay's and Mr Caccamo's emails, this is a response to Mr Millichamp's report. Amongst other things not relevant to the present matter, Mr Kay simply says, 'This looks fine'. There was no reason for the trial judge to refer to this email and he did not err in failing to do so. 164 The next email is one relied upon by the Commonwealth in its written submissions concerning the 'detailed arrangements for supply'. It is Mr Haas' email to Mr Millichamp of 5 September 2007: item 56. This email was not referred to by the trial judge. It was sent in response to an email from Mr Millichamp to Mr Kay dated 4 September 2007: item 54. Mr Millichamp's email was referred to by the trial judge at J [279]: The first of these occurred in early September when Apotex Australia became aware that its application to list its clopidogrel products on the PBS on 1 December 2007 had been filed too late for that to occur, and that the earliest possible date for a PBS listing was 1 April 2008. Mr Millichamp advised Mr Kay of this fact in an email sent on 4 September 2007. The email, which was not copied to Dr Sherman, states: We have had a conversation with the Pricing and PBS listing Department of Health and Ageing. The new guidelines on the 12.5% price drop and timings for PBS listing have resulted in us being unable to get listed in the PBS book for December. In the event that we are successful in defending against an application for interlocutory relief from Sanofi the earliest time that we can get PBS reimbursement is now April 2008. In this case we will require stock in Australia towards the end of February / Start March 2008 assuming that we are able to launch. We please request that the supply team remain on standby to supply us if we are successful in the initial stages of our litigation. The timing has just moved back three months later than originally communicated. 165 In response to this, Mr Haas wrote back to Mr Millichamp in these terms (item 56): Roger, thank you for the update. Target launch date and forecasts will now be shifted out to Mar '08. Marnie: please advise when we will have customer approval for CTN dieline (BLS footprint has already been approved). Also, please have artwork provided so that we can have it fit to our dielines. Considering the magnitude of this launch, I would like to have all upfront activities completed so that we are in a favorable position once we are given the green light to manufacture. We will not order pkg components until we have a favorable litigation outcome Regards Stephen 166 The first question is whether this is material which the trial judge ought to have considered before making the finding at J [340]. The second question is whether it is evidence of the detailed arrangements for supply which the Commonwealth submits that Apotex had made. As to the first question, the short of the matter is that Mr Haas was replying to Mr Millichamp's email and the trial judge dealt with Mr Millichamp's email at J [279]. We do not accept that Mr Haas' response to that email could add anything to what Mr Millichamp had in his mind beyond that which Mr Millichamp's own email showed. As such we do not accept that the fact that the trial judge did not advert to this email has any significance. 167 As to the second question, we do accept that Mr Haas' email is evidence of supply arrangements. The arrangement was that: (a) the target launch date would be moved to March 2008; (b) Ms Peterson was asked to advise when Apotex would have 'customer approval for CTN dieline (BLS footprint has already been approved)' (it is unclear what this means); (c) someone was to provide the artwork; and (d) that no package components would be ordered until there was a favourable outcome in the litigation. There was also a desire to have all 'upfront activities' completed. The upfront activities were not identified by Mr Haas. However, it may be readily inferred that whatever they were, they did not involve the incurring of any substantial expense. So much had been prohibited by Dr Sherman as Mr Millichamp had himself said in his email of 28 June 2007 (item 29) that 'Barry … does not want to waste money on launch activities'. It is convenient to refer to these as the 'Fifth Supply Arrangement'. 168 The final email included in the Commonwealth's written submissions as a document relating to supply arrangements said to have been overlooked by the trial judge was an email sent at 2.37 am on 15 September 2007 from Mr Haas to Dr Sherman: item 65. Mr Haas' email to Dr Sherman was in response to an email sent by Dr Sherman on 13 September 2007. The trial judge dealt with Dr Sherman's email at J [270] recording that in the email he had said '[w]e may be launching in Australia soon' and asked whether 'the inventory of tablets that we have on hand for US [is] is saleable in Australia. i.e. can we repackage for Australian launch?' Mr Haas' email in response to this email was referred to in part by the trial judge at J [270]. The relevant portion of J [270] is a paraphrase of part of the email in these terms: 'Mr Haas informed Dr Sherman that he was looking into this and that Apotex Australia had been "unable to get listed in the PBS book for December"'. The trial judge also recorded that Mr Haas told Dr Sherman the formulation submitted to Australia was identical to that of the US and that the approved shelf life was two years. 169 The Commonwealth's position on this email was a little unclear. At AS [63], it referred to it as evidence for the proposition that '[t]his included, at Dr Sherman's request, looking at whether tablets intended for the US market could be sold in Australia'. If this is what the Commonwealth seeks to get from Mr Haas' email then the submission appears somewhat pointless because it is plain that this is precisely what the trial judge extracted from it at J [270]. However, in the Chronology the Commonwealth appeared to place emphasis on another part of the email which was in these terms: looking into this to the best of my knowledge expiry of existing US inventory is Aug 2008 but I will confirm. Last communication rec'd from Roger Millichamp was: "The new guidelines on the 12.5% price drop and timings for PBS listing have resulted in us being unable to get listed in the PBS book for December. In the event that we are successful in defending against an application for interlocutory relief from Sanofi the earliest time that we can get PBS reimbursement is now April 2008. In this case we will require stock in Australia towards the end of February/Start March 2008 assuming that we are able to launch" (emphasis in original) 170 The trial judge, it is true, did not refer to this statement. The question which arises is whether this part of the email was material to the trial judge's conclusion that the Commonwealth had not shown that Dr Sherman might not change his mind once apprised of the new launch date of 1 April 2008 and the trial date of 28 April 2008. All that one obtains from this email, however, is the unremarkable information that the listing date was now 1 April 2008 which was the very development of which the trial judge was speaking. As such, if this is what the Commonwealth intended by its submission, it seems to make little sense. The only aspect of the email of any moment is the statement that stock would be required by the end of February or the beginning of March to meet that launch date. It is possible to describe this as a supply arrangement and, as such, we will refer to it as the 'Sixth Supply Arrangement'. The arrangement was a request to Dr Sherman, if the injunction were refused, to have stock on hand in Australia towards the end of February or start of March 2008. 171 Having surveyed each of the emails which the Commonwealth submits the trial judge had failed to advert to in this first branch of the Commonwealth's argument, we are not satisfied that any of them were a matter which the trial judge had to consider before making the finding at J [340]. In addition to these emails, the Commonwealth also contends that the trial judge had failed to advert to the transfer pricing agreement between Apotex and Apotex Canada dated 10 July 2007: item 32. No submissions were made about this agreement by the parties at trial. It is not surprising therefore that the trial judge did not refer to it. We detect no error in his Honour's not doing so. 172 It remains then to deal with the Commonwealth's submission that the trial judge had failed to deal with the detailed arrangements for supply which Apotex had made (i.e. the 'second lens' through which to consider the documents identified by the Commonwealth). As we have already noted, the Commonwealth did not expand on what these detailed arrangements for supply were. However, from the foregoing, there are only six supply arrangements referred to in the emails said to have been overlooked. It is useful to recall the six supply arrangements we have identified in the emails upon which the Commonwealth relies. They are: (a) The First Supply Arrangement. This is identified in Mr Haas' email of 28 June 2007: item 26. The arrangement was that Mr Haas would initiate blister tooling activities to determine the time it would take to provide the pack count contemplated by Mr Millichamp for the Australian market. (b) The Second Supply Arrangement. This is identified in Mr Millichamp's email of 28 June 2007: item 29. The arrangement was an aim to launch in October or November with blister packs, a commitment to discuss the various options at the next meeting including initial quantities and follow up/replenishment orders, and an instruction by Mr Millichamp to other employees to continue working on packaging options. (c) The Third Supply Arrangement. This is identified in Mr Haas' email of 28 June 2007: item 29AA. The arrangement was that Mr Haas would set up a meeting with 'packaging engineering' to firm up blister tool lead times. (d) The Fourth Supply Arrangement. This is identified in Dr Desai's email of 29 June 2007: item 30A. The arrangement was that Apotex was not going to manufacture any launch stock of the bisulfate salt. (e) The Fifth Supply Arrangement. This is identified in Mr Haas' email of 5 September 2007: item 56. The arrangement was that the target launch date would be moved to March 2008, Ms Peterson was asked to advise when Apotex would have 'customer approval for CTN dieline (BLS footprint has already been approved)', to provide the artwork, and that no package components would be ordered until there was a favourable outcome in the litigation. There was also a desire to have all upfront activities completed. (f) The Sixth Supply Arrangement. This is identified in Mr Haas' email of 15 September 2007: item 65. The arrangement was to have stock available in Australia towards the end of February or start of March 2008 should the injunction be refused. 173 Once 'the detailed arrangements for supply' are actually identified, the Commonwealth's submission loses much of its lustre. The critical matter is the one referred to in Mr Millichamp's email of 28 June 2007 (item 29) that Dr Sherman did not wish to 'waste money on launch activities until we are clear on what we can do' and his repetition of that in his email 29 June 2007 (item 31): 'That's why Barry does not want us to make stock right now and incur unnecessary expense'. The arrangements for supply were, in our view, the steps which could be taken without incurring any substantial expense. Although the Commonwealth referred to these as supply arrangements, what is striking about them is that the one thing they did not include was any actual arrangement for supply of the active ingredient in the product to be launched. The high watermark is Mr Haas' request to Dr Sherman on 15 September 2007 (item 65) that if the injunction were refused stock would be needed in February or March of the following year. 174 The endpoint of the Commonwealth's submission was that if there were detailed arrangements for supply then this made it more likely, from a forensic perspective, that Mr Millichamp's understanding was correct that he needed no further authority from Dr Sherman to launch. But this, in our view, is precisely what the arrangements do not show. The critical aspect of the supply arrangements, such as they were, was that it remained open to Apotex to decide not to launch without having incurred any substantial expense. Once that is appreciated, the Commonwealth's submission that the trial judge ought to have considered these matters before making the finding at J [340] must fail. Far from undercutting the trial judge's conclusion at J [340], the paucity of the arrangements for supply which had been made demonstrate only that Dr Sherman's instruction not to waste money on launch activities was consistent with Dr Sherman taking the position that he would wait and see what the outcome of the injunction application was before making the decision. This was precisely what Apotex had said in its announcement to its customers and pharmacists and it was precisely what Mr Millichamp had said in his own email to Mr Haas on 17 June 2007 (item 25): '… we will plan to launch subject to Barry's further advice/approval'. 175 In that circumstance, we do not accept that the trial judge erred in not referring to the documents identified by the Commonwealth.

(d) The submissions of Mr Catterns QC 176 Fourthly, it was submitted by the Commonwealth in writing that the trial judge had not had regard to the fact that at the hearing of Sanofi's application for an interlocutory injunction, Apotex's counsel, Mr Catterns QC, had informed the Court that unless it was restrained, Apotex would seek to distribute its clopidogrel products and also that Apotex profferred security of $50 million. Again, this lacks merit since the trial judge expressly dealt with it at J [272]: On 17 September 2007, Sanofi filed its defence and cross-claim against Apotex Australia alleging infringement of the Patent. The cross-claim included the claim for interlocutory injunctive relief. On 18 September 2007, Gyles J heard the application for the interlocutory injunction. At the hearing, Apotex offered security of A$50 million, to be increased as appropriate "as the proceeding progresses". Apotex also offered to maintain comprehensive accounting records of all clopidogrel products sold. During the course of the hearing, Mr Catterns QC was explicit and direct in stating that Apotex would apply for listing as at 1 April 2008 if there was no interlocutory injunction. 177 During the hearing of the appeal, Mr Gleeson advanced a variant of this submission which related to the affidavit of Mr Millichamp which was read before Gyles J on the hearing of the interlocutory application: T26.34. We deal with this argument later in these reasons when we come to that affidavit.

(e) The full email chain of 8-10 August 2007 178 Fifthly, the Commonwealth submitted that the trial judge had failed to refer to 'the full email chain' of 8-10 August 2007: AS [73]. It dealt with this email chain at [62] of its written submissions. From the Chronology, it is apparent that there are five emails in this chain to which the trial judge did not refer. The Commonwealth accepted that the trial judge dealt with the balance of the email chain at J [252]-[255]. Before turning to the five emails in question, it is useful to say a little bit about the emails in the chain to which the trial judge did refer. 179 There were a number of persons in the email chain including Mr Kay, Dr Sherman and Mr Millichamp. The trial judge was impressed at J [252]-[253] by Mr Kay's statement to Dr Sherman in one part of the chain: The current plan is to put Sanofi on notice of our intention to launch and so invite them to seek an interlocutory against us launching. We are assuming that this will be successful and are thus not planning to launch at this stage. I'm not sure why this information had not reached you. 180 As the trial judge observed at J [254], Dr Sherman replied to this a few minutes later: 'We should put Sanofi on notice ASAP. Who is taking care of it?' The trial judge noted that Dr Sherman had not said anything about Mr Kay's observation that Apotex was anticipating that it would lose the application and hence that Apotex Australia was not planning to launch at this stage. Mr Kay was Mr Millichamp's superior. Mr Kay's view that Apotex Australia was expecting to be enjoined and therefore that it was not planning to launch at this stage is, therefore, profoundly important for assessing the significance of the remainder of the email chain. The trial judge reasoned the same way at J [255]: On the basis of the emails exchanged between Mr Kay and Dr Sherman on 8 August 2007, and in the absence of evidence to the contrary, I infer that neither Dr Sherman nor Mr Kay were expecting that Apotex Australia would be successful in resisting an application for an interlocutory injunction by Sanofi. 181 The next email in the chain was from Mr Kay in response to Dr Sherman's 'who is taking care of it' email: item 37. The trial judge referred to this email at J [254] to note that it was entirely redacted. Correctly, the Commonwealth does not submit that the trial judge overlooked this email. 182 Turning then to the five emails in the chain which the Commonwealth says were not referred to by the trial judge, it should first be noted that they all post-date the exchange between Mr Kay and Dr Sherman on which the finding at J [255] was based. The five emails are as follows:

The first email: Email from Mr Millichamp to, inter alia, Dr Sherman, Mr Kay, Mr Haas and Mr Caccomo dated 9 August 2007 at 12.57 am (item 37A) 183 The Commonwealth says that Mr Haas and Mr Caccamo were responsible for supply which is convenient to accept: AS [62]; T52.4-5; T72.7. Most of Mr Millichamp's email is redacted for privilege. However, the second half is not and is in these terms: If Sanofi are successful and get an injunction then we will, of course, not launch and the legal process will run its course in court. If we are successful and defend against an application for interlocutory relief from Sanofi we would like to launch as soon as possible. The earliest date that we can get PBS (Government reimbursement)listing is Dec 07. We asked Stephen Hass to prepare us a launch plan so that in the event we are successful we can launch as soon as possible. We were not expecting any stock to be made or packed until we know the outcome of the hearing for interlocutory relief which should be approximately 2 to 6 weeks after we file for revocation. Please advise if this is an acceptable course of action. 184 The trial judge did not refer to this email, however, in its written submissions the Commonwealth did not rely upon this email. Mr Millichamp's email maps out two different paths of action depending on the outcome of the interlocutory injunction application. Where the trial judge had inferred from the earlier emails between Dr Sherman and Mr Kay that neither expected Apotex Australia to be successful in resisting the injunction application, the fact that Mr Millichamp mapped out two courses of action depending on the outcome of the application is not at all inconsistent with the inference drawn by the trial judge at J [255]. Consequently, we do not accept that Mr Millichamp's email was material to the conclusion at J [255]. The trial judge did not err by not referring to this email.

The second email: Email from Mr Haas to Mr Millichamp dated 9 August 2007 at 8.47 pm (item 38) 185 This email was sent by Mr Haas in response to the email in which Mr Millichamp had charted the two courses of action. In it Mr Haas said 'thank you for providing further clarity'. Once it is accepted that Mr Millichamp's email was not material in light of the trial judge's finding at J [255], it is more difficult to ascribe any significance to this email. Nothing in this brief statement is capable of undermining the significance of Mr Kay's statement to Dr Sherman that Apotex Australia expected to lose the injunction application and was not planning to launch at that stage. Consequently, the trial judge did not err in not referring to it.

The third email: Email from Mr Kay to Millichamp dated 10 August 2007 at 1.29 am (item 39) 186 This email was also copied to Dr Sherman and others. Again, this email is in response to Mr Millichamp's email charting the two courses of action depending on the outcome of the injunction application. What Mr Kay said to Mr Millichamp was this: 'Thanks for the comprehensive update. OK by me'. Again, this is incapable of undermining the significance of the earlier communications between Dr Sherman and Mr Kay or the trial judge's finding at J [255]. The trial judge did not err in not referring to it.

The fourth email: Email from Mr Cacaomo to Mr Millichamp dated 10 August 2007 at 1.30 am (item 39) 187 This email was also copied to Dr Sherman and others. This email is in response to Mr Millichamp's earlier email charting the two courses. Mr Caccamo, who was involved in supply (T72.7), simply said: 'We have aligned our launch plans accordingly'. What one may infer from this is that whatever Mr Caccamo's launch plans had previously been, they had now been aligned with what had been laid out in Mr Millichamp's email. However, Mr Millichamp's email laid out two paths, one involving a launch and one that did not. It is not possible to extract from this email that Mr Caccamo was saying that he was now going to proceed on the basis that a launch was imminent or likely. As such, it is probative of nothing. More particularly, it is incapable of detracting from the significance of the earlier exchange between Dr Sherman and Mr Kay that Apotex Australia expected to lose the injunction application and that they were not planning to launch at that stage. The trial judge did not err in not referring to this email.

The fifth email: Email from Mr Kay to Mr Millichamp dated 10 August 2007 at 1.28 am (item 40) 188 Mr Kay sent this email in response to Mr Millichamp's email outlining the two courses of action and relevantly said only: 'This looks fine'. The reasoning in relation to the first four emails applies with equal force to this email. The trial judge did not therefore err in failing to refer to this email.

Conclusion on the Commonwealth's contention that the trial judge erred in concluding that Dr Sherman's decision to launch at risk was only provisional 189 Much of the evidence which the Commonwealth submits was not referred to by the trial judge was in fact referred to. To the extent that some of it was not, none of it was material for the reasons we have given and we are unpersuaded that his Honour did not refer to it because he had overlooked it. The inference we would draw is that his Honour did consider it but, like us, did not regard it as material worth mentioning. We do not therefore accept that the five matters set out above made it 'abundantly clear' that Dr Sherman had reconfirmed the decision to launch at risk or that there was no evidence in the period from June to September 2007 (scil. 'the all-important months in 2007' (AS [73])) that Dr Sherman had wavered from his desire to launch at risk: AS [73]. On the contrary, the evidence strongly points to the correctness of the trial judge's conclusion. 190 For completeness, under this heading the Commonwealth developed a submission in writing at AS [73] which was, in effect, a general attack on the trial judge's reasoning. For example, the Commonwealth submitted: The trial judge was wrong, on the basis of a selective portion of the evidence, to conclude Dr Sherman's decision was provisional in any relevant sense, to disbelieve Mr Millichamp accordingly, and to refuse to draw the inferences as to what Dr Sherman would have done in the absence of an injunction, which naturally arose from the corpus of evidence properly and fully assessed. 191 There are at least two problems with this. First, it is couched in terms of such generality that the reader cannot know what the submission actually means. For example, what does the Commonwealth mean when it says that the trial judge examined only a 'selective portion of the evidence'? When attention is confined to the terms of Ground 2(b) and the Commonwealth's identification of the documents which it says were overlooked in the Chronology, we cannot interpret this submission as being any more than a somewhat pejorative restatement of its actual argument that the five matters just assessed were not taken into account by the trial judge. Given that much of the evidence the Commonwealth points to was in fact (plainly) considered by the trial judge and that the rest was immaterial, the suggestion that the trial judge had engaged in a selective consideration of the evidence is not one which we would accept. 192 Secondly, the reference to the inferences 'which naturally arose from the corpus of the evidence properly and fully assessed' is an invitation to this Court to draw different inferences to those drawn by the trial judge. Given that the submission is advanced as part of a reason why the Court should conclude that the trial judge erred by overlooking evidence, it is enough to observe that this lies outside of Ground 2(b).

2. Dr Sherman would not change his mind 193 The particular finding that the Commonwealth takes issue with is part of a single sentence in J [251] that his Honour did not 'accept that it was unlikely that Dr Sherman would change his mind'. The first point to make about this is the one already made above that we accept that a challenge to J [251] in this fashion is available to the Commonwealth because it may be seen as a finding which flows as a tributary into the ultimate finding of fact at J [351] that Apotex would not have launched if the injunction had not been issued. The second point is to emphasise again that Ground 2(b) is an allegation that the trial judge failed to deal with evidence or submissions that were material to the outcome; it is not a general warrant for disagreeing with the trial judge's reasons. The third point is that the finding in question actually formed part of the trial judge's reasons for rejecting certain aspects of Mr Millichamp's testimony given under cross-examination. This cross-examination related to Mr Millichamp's email of 27 June 2007 where he had said to Mr Haas: 'If we are successful in avoiding an injunction we will plan to launch subject to Barry's further advice/approval': item 25. Confronted with this email Mr Millichamp persisted with the view that no further approval from Dr Sherman was necessary and that Apotex Australia would have launched at risk unless it had been restrained by an injunction. Part of his evidence before the trial judge had been that 'Barry could change his mind. I mean, I … think he would. But - it would be very unlikely': J [249]. The particular finding that the Commonwealth attacks occurs in that section of these reasons where the trial judge explains why he did not accept this evidence. 194 In writing, the Commonwealth developed three lines of attack on the finding. The first concerned a suggestion that the trial judge had drawn the wrong inferences from the evidence. The second was a submission that the trial judge had engaged in speculation. The third was a submission that the trial judge's findings were contradicted by certain evidence and submissions. It is convenient to deal with each of these contentions separately:

First line of attack: The drawing of wrong inferences 195 Part of the trial judge's reasons for concluding that it was not unlikely that Dr Sherman might change his mind turned on his Honour's conclusion that Dr Sherman had already changed his mind once. The relevant finding is at J [251]: I do not accept the explanation given by Mr Millichamp concerning his email to Mr Haas. In particular I do not accept that the relevant statement was a "throw away" line or that Mr Millichamp was merely seeking to "cover all angles". Nor do I accept that it was unlikely that Dr Sherman would change his mind. On Mr Millichamp's own evidence, Dr Sherman had already changed his mind in that the revocation/launch strategy for Australia was originally predicated on a successful outcome of the litigation in the US and Canada. In my view, the terms of Mr Millichamp's email are more likely than not to be correct in so far as they suggest that Dr Sherman had not finally committed to a launch at risk in Australia and that it would be necessary for Mr Millichamp to obtain final approval to launch at risk in the event that no interlocutory injunction was granted. 196 The Commonwealth submitted in writing that the trial judge had drawn the wrong inference from this: AS [74]. The correct inference, it submitted, was that Dr Sherman was 'bullish'. The trial judge should have drawn that inference because: (a) Dr Sherman had decided to launch at risk in the US; and (b) Apotex Canada was willing to put up security when the interlocutory injunction application was eventually heard. 197 This submission does not involve a contention that the trial judge failed to consider evidence or submissions. Consequently, it lies outside Ground 2(b). Even if it had been within the amended notice of appeal, however, we still would not have accepted it. It is convenient to deal separately with: (a) the significance of the launch in the US; and (b) the fact that Apotex Canada was willing to provide security.

The significance of the launch in the US 198 The Commonwealth suggests that Dr Sherman was to be regarded as 'determined and bullish' because he had launched at risk in the US. However, that submission is directly contrary to the trial judge's findings. At J [206]-[218], the trial judge considered the evidence about what had occurred in the US proceedings. At J [218] his Honour recorded the Commonwealth's submission that Apotex Canada had launched at risk in the US and rejected it: Contrary to a suggestion made in the Commonwealth's opening submissions, I do not think the US launch was genuinely "at risk" in the sense those words were used in the evidence and submissions. Apotex's exposure depended not on how much profit Sanofi lost, but how much profit Apotex made. Apotex knew what its worst case would be in the US (as a percentage of its profit) should the validity of the US Patent be upheld. 199 No challenge to that finding is brought by the Commonwealth. In effect, the Commonwealth invites this Court to ignore the finding that Apotex Canada (through its US affiliate, Apotex US) had not genuinely launched at risk in the US, to conclude contrary to that finding that it had launched at risk, and then to infer from that substituted finding of fact that Dr Sherman was 'bullish'. In the absence of a challenge to J [218], this is not open to this Court even if had been encompassed within the amended notice of appeal. 200 In any case, even if there had been a challenge to J [218], we would not have accepted it. What the Commonwealth omitted from its submissions was any discussion of the amended settlement agreement reached between Sanofi and Apotex Canada with respect to those proceedings. As the trial judge observed at J [217], cl 14(ii) of the agreement set a limit on the amount of damages that Sanofi could recover from Apotex Canada at 50% of its net sales. Thus whilst Apotex Canada had been required to pay Sanofi US$442,209,362 it was entitled to retain the other $442,209,362 of sales which it had made. Of this the trial judge observed, correctly in our view, that '[t]his represented the realisation of the worst-case scenario for Apotex which, were it to have launched clopidogrel in the USA in the absence of the amended settlement agreement, could have resulted in an award of damages greatly in excess of that amount': J [217]. Thus we are unable to apprehend any error in the trial judge's conclusion that Apotex Canada had not genuinely launched at risk in the US. Consequently, this proposition is not available as an aid to the drawing of an inference that Dr Sherman was 'bullish'.

The significance of the security of AU$50 million 201 There is no dispute that at the hearing of the interlocutory injunction application, Apotex proffered an undertaking to provide AU$50 million for any damages and costs that might ultimately be awarded in favour of Sanofi. The trial judge found this at J [266]. At the hearing, Apotex indicated that this could be increased 'as the proceeding progresses' as the trial judge found at J [272]. 202 The Commonwealth's written submissions about this were distributed across three locations. The principal submission was contained elliptically at AS [74] and in slightly more detail at AS [67]. In the Chronology the Commonwealth also identified a number of documents relating to the topic of security which it says the trial judge did not advert to. These were items 48, 48A, 48B, 49 50, 50B and 50C. However, in this branch of the argument, the fact that these were not considered by the trial was not said to be relevant. 203 Instead, the Commonwealth makes three points in writing. First, Apotex Canada had arranged for the guarantee to be given: AS [67]. Secondly, the willingness of Apotex Canada to provide the guarantee showed 'just how determined Apotex was to launch its product at risk if not injuncted': AS [67]. Thirdly, the trial judge had said at J [350] that this evidence was irrelevant. 204 The first point appears to go nowhere. There was no dispute that an undertaking was proffered to provide security in the sum of AU$50 million. The Commonwealth referred at AS [67] to documents which showed that Apotex Canada was intending to provide the security by means of a bank guarantee to be issued by the National Australia Bank. In turn the National Australia Bank would be provided with a letter of credit issued by the Toronto-Dominion Bank. Accepting that to be so, it does not appear to advance matters. Obviously, putting up security of AU$50 million was not a small step. How large a step it might have been would have turned on the cost to Apotex Canada of procuring the letter of credit. None of the documents at AS [67] throw any light on this matter. Thus whilst we accept that this was not a minor step, it is not possible to say just what kind of financial commitment this was from Apotex Canada's perspective. 205 The second step is an inference the Commonwealth suggests should be drawn from the first. Here the thinking would be that if Apotex was willing to put up security in the sum of AU$50 million then this demonstrated a sufficient degree of enthusiasm on its part that it might more readily be inferred that it would launch if not restrained. We do not accept this submission for two reasons. First, because the cost to Apotex Canada of the letter of credit arrangement is not known, it is not possible to gauge what degree of enthusiasm the willingness to provide the security signified. Secondly, as Sanofi submitted, the proffering of an undertaking to provide security was consistent with Apotex keeping its options open: RS [76]. It is entirely possible as a matter of logic for Apotex to have been willing to proffer the undertaking to provide the security and, as yet, not to have made a final decision as to whether it was going to launch. 206 The third step was a submission at AS [67] that the trial judge had misunderstood the Commonwealth's case when he said that evidence showing that Apotex had arranged for the provision of the security was irrelevant. In fact, this is not what the trial judge said. The relevant part of the judgment is at J [350]: I should add that I do not regard Mr Fahner's evidence as to the capacity of Apotex Australia to provide security as a condition of avoiding an interlocutory injunction as relevant to any issue in the case. The critical question for present purposes is whether Apotex Australia would have applied for PBS listing in the absence of the interlocutory injunction. The question whether it would have been required to provide security as the price of avoiding an interlocutory injunction does not arise in the counterfactual analysis which presupposes that there was no entitlement to the interlocutory injunction that was granted. It would be incorrect, as a matter of principle, to instead presuppose that there was no interlocutory injunction if, and only if, Apotex Australia was able to provide security. 207 The finding therefore was not that the evidence that Apotex Canada was willing to provide security was irrelevant. It was that evidence about the capacity of Apotex Canada to provide that security was irrelevant. This is quite a different proposition. We reject the submission. 208 During his oral submissions at T75.22, Mr Gleeson also drew attention to an email sent by Mr Millichamp on 29 August 2007 (item 50) which was not referred to by the trial judge. It was in these terms: As planned, this will inevitably lead to an interlocutory injunction application from Sanofi. We want to defend against this application and launch. [REDACTED] If the court considerers that it may not be appropriate to impose an injunction, as we have foreshadowed, GenRx will need to offer to the court security in the form of an Australian bank guarantee. This guarantees that Gen Rx would be able to pay Sanofi upon the making of an enforceable (ie quantified) award of damages, an amount of money by way of damages. We made this arrangement with Carvedilol where National Australia Bank (NAB) gave the AUD 3mln bank guarantee and Toronto Dominion Bank in turn gave a conditional letter of credit "backing the payment". As discussed previously the amount of the bank guarantee can only be estimated at this stage but is likely to be in the range of AUD 50 -70 mln for the first year. In addition, if we are to avoid an injunction we will almost certainly be required to give undertakings to the court that we will keep detailed records of all our sales and allow inspection of the sales made by an independent third party. We also did this in the Carvedilol case. This email seeks your permission to initiate communication with regard to preparing for a Bank guarantee to be put in place. We propose that Paddy works with Gord Fahner and Jeff Adams to "prepare the ground". It will take a significant amount of time to get matters organised and the sooner we start to prepare the better. 209 Mr Gleeson submitted at T76.4 that this was further evidence of the seriousness of Apotex's intention to launch. However, this email does no more than repeat the substantive matters about the bank guarantee to which he had just referred. For the same reasons that they were not material in the relevant sense, this email is also not material.

Conclusions on the first line of attack that wrong inferences were drawn 210 This submission is not open on the amended notice of appeal. Even if it were, there is no substance to it.

Second line of attack: The trial judge engaged in unsubstantiated speculation 211 The trial judge found that: Apotex had missed the deadline for PBS listing on 1 December 2007 (J [279]); that the next date it could be listed on the PBS was in April 2008 (J [279]); that following the determination of the interlocutory injunction application, Gyles J had listed the matter for final hearing commencing on 28 April 2008 (J [281]); and that Gyles J was obliged to retire by 22 August 2008 and would therefore need to deliver judgment by that date (J [289]). 212 The trial judge's course of reasoning was as follows. First, his Honour found that Dr Sherman's 20 February 2007 email was not a sure guide as to what Dr Sherman would have done eight or nine months later confronted with changed circumstances. This appeared at J [341]: Nor do I consider that the email sent by Dr Sherman on 20 February 2007 is persuasive evidence as to what Dr Sherman's thinking may have been eight or nine months later. The 20 February 2007 email was sent prior to any decision by the US District Court in relation to the US Patent. Some of the developments that subsequently occurred in 2007 were expected; the ARTG registration of Apotex's clopidogrel products was one of these. However, the coincidence of events that would have led to both a PBS listing on 1 April 2008 (in the counterfactual scenario in which no interlocutory injunction was granted) followed by the commencement of the trial of the patent proceeding (as actually occurred) later that month was a result of developments that could not have been anticipated until about the time the interlocutory application was determined. 213 Secondly, the trial judge drew a Jones v Dunkel inference against the Commonwealth for failing to call Dr Sherman to give evidence: J [342]-[347]. The drawing of that inference is subject to a separate challenge by the Commonwealth under Ground 2(c) and may be passed over for now. 214 Thirdly, the trial judge concluded that he was not prepared to infer from Dr Sherman's 20 February 2007 email that he would have decided to launch at risk on 1 April 2008. This occurred at J [348]: I am not prepared to infer, based on the 20 February 2007 email, or any of the subsequent correspondence in evidence which was said to justify the drawing of such an inference, that Dr Sherman was likely to have instructed Mr Millichamp to procure the listing of Apotex's clopidogrel products with effect from 1 April 2008. 215 Fourthly, his Honour identified an evidentiary deficiency in the Commonwealth's case at J [349]: In my opinion, the Commonwealth's case suffers from an evidentiary deficiency which cannot be made good by drawing inferences from correspondence written by Dr Sherman in the lead up to the hearing of the interlocutory application. In particular, I do not think it can be inferred that if Dr Sherman had known that the trial of the patent proceeding would commence in the same month that Apotex Australia obtained a PBS listing of its clopidogrel products (triggering a 12.5% statutory price reduction), that he would have, in those circumstances, authorised Apotex Australia to obtain such a listing before judgment was delivered or, at least, until the trial had concluded (by which time he and his colleagues and his legal advisers may have had a clearer view of the strength of Sanofi's case). 216 The Commonwealth submitted on the basis of this that the trial judge had concluded that the Commonwealth had not established that Dr Sherman might not have changed his mind having regard to two facts: (a) the fact that following the determination by Gyles J of the application for the interlocutory injunction it was known that the PBS listing date would be 1 April 2008; and (b) that the matter would be tried commencing on 28 April 2008: AS [75]. It was said that this possible change of position by Dr Sherman based on these facts was the unsubstantiated speculation the Commonwealth alleged the trial judge had engaged in: AS [76]. 217 We would reject this argument for two reasons. First, it is not encompassed within Ground 2(b) as it is not a contention that the trial judge overlooked evidence or submissions. Secondly, the criticism that the trial judge engaged in unsubstantiated speculation is incorrect. There was no direct evidence of whether Dr Sherman would have decided to launch at risk on 1 April 2008 if the interlocutory injunction had been refused. If it was to be proved, by necessity it required the drawing of an inference about Dr Sherman's behaviour in the hypothetical state of affairs that the interlocutory injunction had been refused. Like any inference, that inference had to be drawn based on other evidence. The party seeking the drawing of that inference was the Commonwealth. It therefore bore the onus of proof. What was the conceivable universe of evidence from which such an inference might be drawn? It could consist only of: (a) indirect evidence of Dr Sherman's likely disposition towards launching at risk on 1 April 2008. This evidence consisted of correspondence to which Dr Sherman was a party, statements made by persons who were dealing with Dr Sherman and other objectively available evidence; and (b) direct evidence from Dr Sherman as to whether he would have decided to launch at risk on 1 April 2008 but for the interlocutory injunction. 218 Leaving to one side the question of whether a Jones v Dunkel inference might be drawn against the Commonwealth for failing to call Dr Sherman, the fact is that he was not called. Consequently, direct evidence of the kind referred to in (b) was not available. The question which therefore arose for the trial judge was whether it was to be inferred, as the Commonwealth contended, from the indirect evidence referred to in (a) that Dr Sherman would have decided to launch at risk on 1 April 2008. 219 The trial judge did not accept that the inference sought by the Commonwealth should be drawn. The first step was that his Honour found that the coincidence of the PBS listing date of 1 April 2008 and the trial date of 28 April 2008 was not something which could have been anticipated until the time that the interlocutory injunction application was determined: J [341]. The second step was to observe that his Honour was not minded to draw the inference that Dr Sherman would have launched at risk in the counterfactual on the basis of the evidence of the kind in (a): J [348]. The third step was to identify as an evidentiary lacuna the fact that Dr Sherman had not given evidence of the kind in (b). Contrary to the Commonwealth's submissions, his Honour did not speculate about what Dr Sherman would, or would not have done, once apprised of the 1 April 2008 launch date and 28 April 2008 trial date. Rather, what his Honour said at J [349] was that he was not going to draw any inference about it in the absence of evidence from Dr Sherman. 220 Thus the Commonwealth's submission that the trial judge had speculated about the impact on Dr Sherman of the 1 April 2008 launch date and 28 April 2008 trial date is quite wrong. Rather, his Honour declined to speculate about what Dr Sherman would have done confronted with this information in the absence of Dr Sherman. We see nothing erroneous in this approach. 221 The only answer to this would be if the indirect evidence in (a) were in such a state that the only inference available was that Dr Sherman would have decided to launch at risk on 1 April 2008. In our opinion, such a submission would be untenable. At trial, the Commonwealth advanced a case that Apotex was an aggressive generic pharmaceutical company with a propensity towards launching at risk. On appeal, the Commonwealth advanced a similar argument together with the argument that Apotex had launched clopidogrel at risk in the US. As we have noted, the trial judge rejected the latter contention finding that the US launch was not truly at risk at all. 222 At trial, the Commonwealth advanced evidence from Mr Millichamp that Apotex was always trying to 'push the boundaries' and always trying to 'get on the market at the earliest time possible for our commercial advantage'. The trial judge noted this evidence at J [316] but was unimpressed by it for a number of reasons. The first of these was that it was inconsistent with Apotex's ruminations immediately prior to the time when Gyles J was to hand down his judgment. One possibility that Apotex sought to wrestle with at that time was whether, if it had succeeded in full at trial, should it then launch at risk even though an appeal would be inevitable. An email from Mr Millichamp's superior, Mr Kay, dated 12 August 2008, noted by the trial judge at J [306], suggested that it was not Mr Kay's view that a launch should occur: As discussed last week my view is that in the event of our success and should [Sanofi] decide to appeal we should in some way allow the injunction to continue, and seek damages should any appeal fail to go [Sanofi's] way. 223 Mr Millichamp was cross-examined about this and sought to explain it away on the basis that the situation was different where an appeal was pending. But the trial judge was unimpressed by this since his Honour was unpersuaded that the business case for a launch at risk before a trial at first instance was any different to a launch at risk pending an appeal from a successful trial outcome: J [317]. 224 The Commonwealth criticised this conclusion. Its argument was that it was inappropriate to rely on that analysis because circumstances had changed by the time it came to be done. As we understood this point, it was that in the counterfactual, Apotex would have launched at risk in September 2007 whereas the subsequent analysis was not premised on Apotex having entered the market at that time or, to use the language of the Commonwealth's submission, the subsequent analysis related to 'an entirely different circumstance when Apotex had already been held off the market for a year by reason of the interlocutory injunction': AS [77]. 225 We would reject this contention for two reasons. First, it is not an argument that the trial judge overlooked evidence or submissions and therefore lies outside Ground 2(b). Secondly, even if it were, it is without substance. 226 As to the second point, the trial judge dealt in fine detail with three risk/reward analyses which were conducted by Apotex. The first was prepared prior to the hearing of the interlocutory injunction application, the second prior to the delivery of the final judgment by Gyles J, and the third following the delivery of the Full Court's judgment. Each was directed at the same issue which was whether, in the event that Apotex were successful in each impending judgment, it made economic sense to launch at risk given the possibility of subsequent defeat and concomitant exposure to damages. 227 As the trial judge noted at J [324], the second analysis of July 2008 dealt with three scenarios. One of these was that there would be no other authorised generic, that there would be a 12.5% price drop on PBS listing, and that Apotex would obtain a 35% market share. However, the trial judge found at J [324] that that scenario was mirrored in the June 2007 analysis. This matters because his Honour also found that under that scenario, Apotex's exposure to damages greatly exceeded the profits it expected to make: J [324]. Once that is brought to account, it can be seen that there is nothing in the Commonwealth's submission. The risk/reward analysis conducted in July 2008 included within it the same assumption that Apotex would achieve the 35% market share that its earlier analysis prior to the interlocutory injunction had contained. As such the contention that circumstances had changed because Apotex had been kept out of the market for a year falls away, for the 35% market share assumption remained. 228 Consequently, even if it were open to this Court to entertain this submission, which it is not, it is incorrect. Further, it is equally clear that the trial judge was aware that the situation in July 2008 could not be precisely the same as it had been in June 2007: J [322] ('I accept that the July 2008 and September 2009 risk/reward analyses addressed different circumstances from those that existed in August 2007'). Correctly, with respect, his Honour tied this back to the actual evidence with which he had to deal. This was Mr Millichamp's oral evidence that Apotex tried to get to market as soon as possible. His Honour felt at J [322] that that evidence was difficult to reconcile with the analyses that Apotex had itself done in July 2008 and September 2009: Nevertheless, those analyses and Apotex Canada's ultimate response to them tends to contradict the assertion made in Mr Millichamp's oral evidence that Apotex tries to get on the market as soon as possible. 229 We agree. In light of these matters the trial judge was unpersuaded by Mr Millichamp's attempts to explain how that evidence could be reconciled with his evidence that Apotex tried to get to market as soon as possible on the basis that the business case had changed: J [317]. The trial judge's point was that the fact that both the June 2007 and July 2008 risk/reward analyses were premised on, inter alia, a 35% market share and that this showed that Apotex stood to lose much more than it stood to gain by launching, meant that Mr Millichamp's evidence that the business case had changed could not be correct. 230 Returning then to the contention under discussion (i.e. whether the only inference open on the evidence was that Dr Sherman would have launched at risk on 1 April 2008), we do not conclude that this could possibly be so. There was abundant evidence from which it could be inferred that Apotex would not launch at risk on 1 April 2008. This included the facts that it had not launched at risk in the US, had not been proposing to launch at risk if it had succeeded before Gyles J and did not launch at risk once it fully succeeded in the Full Court. 231 Once that position is reached, the problem the Commonwealth confronted in its endeavours to have the trial judge conclude that Apotex would have launched at risk on 1 April 2008 even though the matter was listed for final trial is precisely as the trial judge described. An inference about what Dr Sherman would, or would not have not done, simply could not be drawn without evidence from Dr Sherman. Once this is grasped, the submission that the trial judge had engaged in 'unsubstantiated speculation' must be rejected. The trial judge, properly in our view, declined to engage in speculation to fill the evidentiary vacuum resulting from the absence of Dr Sherman. The Commonwealth's second line of attack therefore fails.

Third line of attack: The trial judge's findings were contradicted by certain evidence and submissions 232 The Commonwealth submitted that the trial judge's conclusion that Dr Sherman might have changed his position once it was known that launch date would be 1 April 2008 and the trial date was 28 April 2008 was also 'directly contradicted by the contemporaneous records and actions': AS [76]. To this end, the Commonwealth identified five sources of material: (a) those referred to at AS [60]-[67]; (b) a document described as B1.294.CB5.0228; (c) a document described as B1.299.CB11.0223; (d) a document described as B1.313.CB11.0229 at .0230; and (e) Dr Sherman's email of 14 September 2007 sent at 12.25 pm (raised by Mr Gleeson at T83.6). 233 We would reject this submission on the basis that the premise on which it rests - that the trial judge engaged in speculation about what Dr Sherman would have done - is unfounded. Turning then to the identified matters:

(a) The Commonwealth's written submissions at [60]-[67] 234 These paragraphs consist of a recitation of what the Commonwealth's case at trial had been. At AS [57], the Commonwealth's submission says: 'The essence of the Commonwealth's case at trial was as follows'. Paragraph [58] sets out Dr Sherman's email of 20 February 2007 and [59] outlines what the Commonwealth says were powerful considerations supporting its interpretation of that email. Paragraph [60] is a broad statement of the Commonwealth's contention that Apotex would have launched at risk but for the injunction. It concludes with these words: 'The critical elements of that submission and the evidence relied upon in support can be categorised as follows'. It is likely therefore that the Commonwealth's submission is really that the materials upon which it relies on appeal are contained in AS [61]-[67] rather than AS [60]-[67]. 235 These paragraphs of its closing submissions contain references both to documents to which the trial judge did refer and those to which his Honour did not. The submission made under this third strand is that the trial judge's speculation was contradicted by materials contained in these paragraphs of its closing submissions. To the extent that AS [61]-[67] contain references to documents to which the trial judge did refer, it lies outside the scope of Ground 2(b) which is only about evidence and submissions which the trial judge did not advert to. It is necessary to consider each paragraph:

Paragraph [61] 236 Starting then with AS [61], the Commonwealth first refers to Mr Kay's 25 June 2007 email to Dr Sherman asking him to reconfirm his instructions. As already noted, this email was dealt with by the trial judge at J [241]. The email resulted, according to the Commonwealth, in a 'flurry of emails'. Two such flurries are nominated by the Commonwealth using this nomenclature: B1.200.CTB9.0129 and B1.201.CTB9.0131. The former is an email chain which appears in items 20A, 20B, 20C, 20D, 21 and 22 of the Chronology. The trial judge referred to these emails at J [242]-[244]. As such, there can be no suggestion that the trial judge did not consider this evidence and, indeed, to read J [242]-[244] is to see that his Honour gave the emails close consideration. In any event, the point for present purposes is that this email chain cannot be called in aid of this third line of attack. It is not enough to say that the trial judge's conclusion was contradicted by evidence to the contrary. The Commonwealth's case is that it was contradicted by evidence which the trial judge did not consider and which was material. This email chain does not therefore assist the Commonwealth. 237 The second document - B1.201.CTB9.0131 - appears in the Chronology as items 21A and 23. It is part of the same email exchange. The trial judge referred to this exchange at J [245]. Consequently, it does not assist the Commonwealth since it is evidence which the trial judge considered. In any event, the email is entirely redacted as the trial judge noted at J [245]. 238 Nothing in AS [61] therefore can sustain the submission that the trial judge overlooked evidence material to his conclusion that the Commonwealth had not shown that Dr Sherman might not have changed his mind.

Paragraph [62] 239 This paragraph refers to a number of documents, although the submission largely does not distinguish between those that the trial judge referred to and those which his Honour did not. A comparison between the document codes in this paragraph and the Chronology reveals that four documents referred to in this paragraph were not referred to by the trial judge. These are: (a) the email chain between Mr Millichamp's and Ms McTavish dated 28 June 2007 (items 27 and 28); (b) Mr Chalk's redacted email to Mr Hughes sent at 6.59 pm on 28 June 2007 (item 28A); (c) Mr Millichamp's email forwarding Mr Chalk's email to Mr Haas at 10.28 pm on 28 June 2007 (item 29); and (d) the email chain dated 8-10 August 2007 (items 38 and 39). 240 We have dealt with each of these above and explained why none were material. Consequently, there is no evidence referred to in AS [62] which assists the Commonwealth in the submission it advances at AS [76].

Paragraph [63] 241 We have listed above at [124] the documents referred to in AS [63] which were not referred to by the trial judge. We have already explained above why the trial judge was not required to refer to these documents in his reasons. We would, in any event, infer that the trial judge did not refer to these documents because he considered them immaterial. 242 It follows that there is nothing in the documents referred to at AS [63] to support the submission at AS [76] that this was contemporaneous evidence which contradicted the trial judge's conclusion.

Paragraph [64] 243 There are no documents referred to in this paragraph which were not referred to by the trial judge.

Paragraph [65] 244 The only document referred to in this paragraph which was not considered by the trial judge was the email from the General Manager of Sanofi, Mr Moulding, to other Sanofi employees on 21 August 2007: item 48. We have explained above why this email is of no relevance.

Paragraph [66] 245 There are no documents referred to in this paragraph which were not considered by the trial judge except for an email dated 5 September 2007: item 56. The initiating email from Mr Millichamp (item 54) to which it responds was considered by the trial judge at J [279] however the response from Mr Haas (item 56) was not. We have explained above why Mr Haas' response was not material.

Paragraph [67] 246 There are no documents in this paragraph which were not considered by the trial judge. 247 Having considered the evidence referred to in the Commonwealth's written submissions at [61]-[67] which were not considered by the trial judge, we do not think that any of them contradicted the trial judge's conclusion that the Commonwealth had not proven that Dr Sherman might not have changed his mind once apprised of the 1 April 2008 launch date and the 28 April 2008 trial date.

(b) Document B1.294.CB5.0228 248 This is Mr Millichamp's email to Mr Kay dated 4 September 2007: item 54. The trial judge referred to this email at J [279]. It therefore is not within the ambit of the Commonwealth's appeal.

(c) Document B1.299.CB11.0223 249 This is Mr Haas' email to Mr Millichamp dated 5 September 2007: item 56. Although the trial judge did not refer to this email, we have already concluded above that this email was not material which the trial judge was obliged to consider in his reasons. We would infer that his Honour correctly regarded this email as immaterial.

(d) Document B1.313.CB11.0229 at .0230 250 This is an email from Mr Millichamp to Mr Kay dated 14 September 2007: item 60C. However, this email was referred to by the trial judge at J [268] and therefore is not within Ground 2(b).

(e) Dr Sherman's email of 14 September 2007 sent at 12.25 pm 251 This document was mentioned by Mr Gleeson at T83. It is item 61 of the Chronology. The parties are in agreement that parts of Dr Sherman's email were referred to by the trial judge at J [270]. Mr Gleeson accepted this but suggested, without identifying them, that parts of it had not been: T83.8. So far as we can see, the relevant portion of the email is at B1.323.STB3.0257 and is as follows: We may be launching in Australia soon. Also, it is possible that we will relaunch in the US within months. Questions as follows: 1. Our US tablets are darker in colour than Plavix, and I know we did trials to lighten. What did we submit in Australia? Identical to US tablets, or lighter in colour, or different in any other way? 2. Is the inventory of tablets that we have on hand for US saleable in Australia. i.e. can we repackage for Australian launch? 3. If we want to lighten the colour for the US on launch, what is needed? Does it require pre-approval to reduce colour in filmcoat? CBE0, CBE30, annual report?? What is needed to do? Who can answer these questions? 252 The trial judge's treatment was at J [270] and was, relevantly, in these terms: The same day, Dr Sherman sent an email to people in Apotex Canada including Dr Desai, Mr Caccamo and Mr Haas, stating "[w]e may be launching in Australia soon" and asking whether "the inventory of tablets that we have on hand for US [is] saleable in Australia. i.e. can we repackage for Australian launch?" 253 The complaint therefore is that the trial judge did not advert, first, to the statement that it was possible that Apotex would launch in the US within months or, secondly, to the first or third of the three questions he set out. The first statement is not material and no error is disclosed by not referring to it - that a launch was possible does not approach the height at which the Commonwealth's submission is pitched. Dr Sherman's first and third questions were, in our opinion, clearly considered by the trial judge. As the balance of J [270] shows, the trial judge was aware of the answers that Dr Sherman received to these questions because he set out Mr Haas' emails responding to Dr Sherman in the same paragraph: Mr Haas informed Dr Sherman that he was looking into this and that Apotex Australia had been "unable to get listed in the PBS book for December". He also told Dr Sherman that "[t]he formulation submitted to Australia is identical to that of the USA", and that approved shelf life was two years. Dr Sherman responded, asking "When were the tablets made? What is expiry dating? I expect that the tablets are in fact good for years. What is required to extend expiry dating for Australia to 3 years?" Mr Haas provided a response that appears to suggest that the US product would have had an expiry date of 29 August 2008 which (I infer) would not be sufficient for retail sale in Australia after 1 April 2008. On 17 September 2007, Ms Keast advised Rekha Panchal, who was part of the regulatory affairs team of Apotex Canada, of the stability data that would be required to support a three year expiry date. 254 We do not accept therefore that the trial judge did not consider the questions in Dr Sherman's email. Even if that were not so, we struggle to see how Dr Sherman's first and third questions could be characterised as evidence which the trial judge was bound to consider. 255 Having taken into account the answers to the questions, his Honour's omission to set the questions out is an empty point. As such, this point goes nowhere. 256 Having considered the five sets of material said to support the Commonwealth's third line of attack on the trial judge's finding that the Commonwealth had not proven that Dr Sherman might not have changed his mind, we conclude that none of them substantiate that submission.

3. Impermissible speculation as to whether Dr Sherman would have changed his mind if he had known that the hearing would be in April 2008 with a judgment by August 2008 257 Largely we have dealt with this in the preceding section. However, the Commonwealth advanced an additional submission that the trial judge ought not to have engaged in speculation about what Dr Sherman might do if apprised of the trial date of April 2008 without considering (AS [77]): (a) the 'compelling reasons' why Dr Sherman had made the decision to launch in the first place; (b) the interlocking steps which had been taken since then to launch at risk; (c) the actual evidence before his Honour about what had been done to justify a launch at risk; and (d) the fact that immediately before the hearing of the interlocutory injunction Mr Kay's only concern was whether a competitor could beat Apotex to a 1 April 2008 launch. 258 The main problem with this submission is that the trial judge did not speculate that Dr Sherman might have changed his mind if apprised of the trial dates. What his Honour concluded was that he was not minded to draw any inference as to what Dr Sherman might have done on the evidence before him. However, even if that problem were surmounted, we would not accept any of (a)-(d).

(a) The 'compelling reasons' submission 259 The Commonwealth did not explicitly identify the 'compelling reasons' to which it was referring but it is likely, and we are prepared to assume, that this is a reference to the 'powerful considerations' set out in AS [59]. Having discussed Dr Sherman's 2007 decision to launch, AS [59] begins with 'Powerful reasons supported that decision'. Then follows references to three considerations.

First compelling reason 260 The first reason was that Apotex Canada was an aggressive generic drug company that had a strategy of identifying weak patents and launching generic medicines at risk as long as it was not restrained by an interlocutory injunction. We would reject this for three reasons. 261 First, the trial judge made no finding to this effect. Indeed, as we have explained above, the evidence before his Honour as to how Apotex approached launching at risk did not persuade his Honour to accept Mr Millichamp's evidence that Apotex was always pushing the boundaries and trying to get to market as soon as possible. In particular, Apotex had not truly launched at risk with clopidogrel in the US or in a comparable way with either perindopril erbumine or carvedilol or even with clopidogrel once it had succeeded in the Full Court. Without disturbing those findings, which the Commonwealth did not suggest that we should, we do not see how it would be open to this Court now to find as the Commonwealth suggests. 262 Secondly, even if it were now open to this Court to do so, we would not. To make good this contention, the Commonwealth now referred the Court to the agreement by which Apotex Canada had indemnified Apotex Australia for patent infringement and its closing submissions before the trial judge at [83], [88], [93], [104]-[105], [241], [224]-[252]. 263 The indemnity agreement provides no basis for the making of such a finding. It merely shows that the parent company, Apotex Canada, had indemnified the Australian subsidiary, Apotex Australia, for patent infringement and obtained, inter alia, rights to control any patent infringement litigation. 264 As for [83], [88], [93] and [104]-[105] of the Commonwealth's closing submissions at trial, none advance an argument that such a finding should have been made. 265 Paragraphs [224]-[252] are a long section of the Commonwealth's closing submissions in which it developed its contention that objective factors favoured the view that Apotex would launch at risk. This view the trial judge rejected. Amongst [224]-[252], however, [241] contains this statement: 'Mr Millichamp's evidence that Apotex is aggressive and has an appetite to challenge patents was not challenged. Nor was his evidence about Apotex's "attitude to risk" and willingness to launch at risk'. However, as we have explained above, the trial judge gave extensive reasons for not accepting this evidence. Unless those findings be overturned, this submission and the finding it would necessarily entail, is not open to this Court. 266 Thirdly, even if all of that were wrong, none of this material has been demonstrated to be the reason why Dr Sherman decided to launch clopidogrel in the first place which is, after all, the use to which the Commonwealth seeks to put it. Put shortly, it is not evidence about Dr Sherman's reasons. At best, it is evidence about Mr Millichamp's understanding of Apotex's practices. 267 In those circumstances, the first 'compelling reason' is not open to this Court to find, ought not to be found even if it were and, even if it should be found, does not demonstrate anything about Dr Sherman's original reasons for deciding to launch clopidogrel. We therefore do not accept it.

Second compelling reason 268 Here the Commonwealth submitted that Australia was a new market for the Apotex group, Apotex was in an investment phase, and that clopidogrel was a blockbuster drug with enormous sales potential. The first is without substance since it is obvious that the case is concerned with the fact that Apotex was attempting to enter the Australian market as the first generic. The third is also without substance as the very first paragraph of the trial judge's reasons for judgment refer to the 'blockbuster' nature of the drug and the fact that it had generated worldwide annual sales of US$1 billion for the Sanofi group (and the BMS Group). 269 As to the second point (i.e. that Apotex was in an investment phase), the material said to support this contention was to be found at [83] and [225]-[237] of the Commonwealth's closing submissions at trial. Paragraphs [83], [225]-[227], [229] and [231]-[237] do not deal in any way with the fact that Apotex was in 'investment phase' and may be put aside. However, [228] and [230] do contain material about this. Paragraph [228] contains an en passant reference to Apotex being in 'investment phase' as part of a submission that Apotex stood to benefit from being the only generic on the market. The point therefore is that Apotex would not only benefit from getting clopidogrel to market since it would be the only generic but also because this would benefit a company in investment phase. The reference at [230] is merely to Mr Millichamp's evidence about this being unchallenged and was accompanied by a corresponding submission to the same effect in this Court. 270 The trial judge did not advert to the fact that Apotex would benefit from being the first generic on the market not only because of the sales it would make but also because it was in investment phase. At this point, it is useful to bring this back to what the Commonwealth seeks to get from it. This is the proposition that the trial judge ought not to have speculated about what Dr Sherman would have done had he known of the 1 April 2008 launch date and the trial date of 28 April 2008 because his Honour's alleged speculation was contradicted by the evidence of the compelling reasons Dr Sherman had decided to launch clopidogrel in the first place. Once that is recalled, this material lacks the quality which the Commonwealth's submission ascribes to it at AS [77], namely, being one of the compelling reasons why Dr Sherman had decided to launch in the first place. It was a reason identified by Mr Millichamp. As such, this material does not establish the existence of this as one of Dr Sherman's reasons.

Third compelling reason 271 Here the Commonwealth referred to the fact that Apotex Canada had launched clopidogrel at risk in the US and had made over US$1 billion in sales in one week before the grant of an injunction. There are two problems with this. 272 First, the trial judge found at J [217]-[218] that he was not satisfied this was a launch truly at risk due to the cl 14(ii) of the amended settlement agreement which capped Sanofi's damages at 50% of Apotex Canada's net sales. The force of this will be apparent. Apotex was always going to keep half of the sales it made. Once that is brought to account we do not accept that it constituted a compelling reason why Dr Sherman would have decided to launch at risk in Australia given the absence of an equivalent arrangement in this jurisdiction. Secondly, the proposition that the trial judge failed to consider this matter cannot be maintained given that he did consider in some detail what had happened when clopidogrel was launched in the US including the revenues generated: J [206]-[218].

Conclusions of compelling reasons submission 273 In those circumstances, we do not accept that the three 'compelling reasons' put forward by the Commonwealth provide any warrant for a finding of error on the part of the trial judge.

(b) The interlocking steps submission 274 The Commonwealth's submission did not identify the interlocking steps which it said Apotex had taken to launch at risk. One possible candidate for what these interlocking steps might be is that they are the same as the 'detailed arrangements for supply' the Commonwealth pointed to at AS [72]. If that is what was intended by the submission, we have already explained why the arrangements for supply which had been made do not lend weight to the idea that they demonstrated that Apotex was committed to launch at risk. 275 If something else was intended by the submission, it is not obvious to us what it was and, in the absence of explanation, we are disinclined to entertain it.

(c) The overlooked evidence submission 276 The Commonwealth submitted that the trial judge had overlooked six items of evidence pertaining to the thinking within Apotex at September 2007 and the calculations which had been done to justify a launch at risk. The overlooked evidence was identified at AS [77] and was as follows:

An email from Mr Millichamp to Mr Weingarten and Ms McTavish dated 3 May 2006 (item 6) 277 This email was not referred to by the trial judge. In this email which is long before Dr Sherman's email of 20 February 2007, Mr Millichamp informed Mr Weingarten, the Vice-President of sales at Apotex Canada, and Ms McTavish that IMS data showed that clopidogrel sales in Australia were worth AU$139.6 million and were growing at a rate of 15%. He also provided a proposed pricing scenario for Apotex's clopidogrel product. However, what this does not deal with (but which the trial judge did deal with) was the other half of the picture. No doubt, Mr Millichamp as the executive in charge of Australia was focussed on the profits to be made by a launch, but it was Apotex Canada which was going to wear the cost of any unsuccessful patent litigation as the indemnity agreement between the parent and its subsidiary made clear. It is of little use, therefore, to point to evidence that there were good sales to be made if there was a launch without at the same time assaying what the possible consequences would be if Apotex were successfully sued for patent infringement. As we have explained above, the trial judge dealt in detail with the various risk/reward analyses which looked at both sides of the picture. As such we do not think there was any reason for the trial judge to have referred to this email, or in the case that he ought to have, that it would have made a material difference to his decision in circumstances where he considered in detail the risk/reward scenarios.

An email from Ms Peterson to Mr Millichamp dated 16 May 2006 (item 7) 278 The Commonwealth's submission that the trial judge did not consider this email is wrong. His Honour dealt with it at J [222].

An affidavit of Mr Millichamp dated 7 July 2017 at ss38-39 279 These two paragraphs are as follows: On 22 June 2007, I emailed Mr Andrew Kay my financial forecasts and my recommendation regarding the proposed launch. My email to Mr Kay assumed that the launch would have involved the listing of the Apotex Registered Clopidogrel Products on the PBS. This is clear because I have factored into the financial forecast the 12.5% price drop which would have occurred upon PBS listing. Annexed to my First Affidavit and marked CONFIDENTIAL RM-86 is a copy of my email to Mr Kay, part of which has been redacted because it contains information which Apotex Australia considers is subject to legal professional privilege. Mr Kay responded to my email of 22 June 2007 twice that same day, with further questions. Annexed to this affidavit and marked "CONFIDENTIAL RM-196" is a copy of that chain of emails. 280 The Commonwealth relies in writing on these two paragraphs to make good the proposition that the trial judge had failed to have regard to what the actual thinking was within Apotex in September 2007 as to a launch at risk and the calculations which had been done to justify such a launch. This submission should be rejected. All that ss38-39 do is summarise the content of two emails which are then exhibited at RM-86 and RM-196. The exhibited emails are part of the Chronology as item 18. As the Chronology makes clear, these were in fact referred to by the trial judge. His Honour did so at J [234] where he set Mr Millichamp's email out and refers to Mr Kay's response. As such it is clear that the trial judge was aware of the 'actual thinking' within Apotex (as it happens in June 2007, not as the Commonwealth's submissions suggest at AS [77], September 2007). There is, therefore, nothing in this submission.

An email from Mr Millichamp to Mr Kay dated 22 June 2007 (item 18) 281 This is an email from Mr Millichamp to Mr Kay setting out Apotex's recommendation and attaching financial scenarios relating to launching at risk. This email is therefore both evidence of Apotex's internal thinking in June 2007 and of calculations about the recommended launch, as the Commonwealth submits. The difficulty with the Commonwealth's submission is that the trial judge dealt with this email and the forecasts contained within it. He set out Mr Millichamp's recommendation that there should be a launch at risk at J [234]. The email contained profit forecasts based on two different scenarios which the trial judge set out. The difference between the scenarios was that one proceeded on an assumption that no other generic entered the market and Apotex obtained a 35% market share. The other scenario assumed that another generic did enter the market and that Apotex obtained a 17% market share. Under the first scenario, Mr Millichamp forecast year one profits of AU$20,077,236 whilst under the second he forecast year one profits of AU$9,751,800. 282 However, as the trial judge noted at J [236], Mr Millichamp's forecasts included no consideration of the financial impact on Apotex were Sanofi's patent to be held valid and for Apotex to have infringed it. Put another way, Mr Millichamp's forecasts were all upside without any consideration of the downside. 283 Nevertheless, Mr Millichamp did consider Apotex's potential exposure to Sanofi were the patent to be held valid a couple of days later in an email he sent to his superior, Mr Kay. The trial judge dealt with this email at J [237]. In this email Mr Millichamp adverted to the difficulty of assessing what Sanofi's losses would be but came up with an estimate of AU$50-70 million per year. 284 Pausing there, it will be at once apparent that when one combines Mr Millichamp's estimate of the potential damages to the potential profits to be made, the former dramatically outstripped the latter. This was a point the trial judge returned to at J [324] where he observed of these two scenarios (and a third not presently material), that 'the potential exposure arising out of a claim for damages by Sanofi greatly exceeded the forecast profit'. 285 The Commonwealth relied on this email (item 18) to argue that the trial judge had failed to have regard to what the actual thinking was within Apotex in September 2007 as to a launch at risk and the calculations which had been done to justify a launch at risk. However, in light of the foregoing discussion, this submission must be rejected. The trial judge did have regard to the email and the financial calculations it contained.

Emails dated 13 September 2007 between Mr Fahner and Mr Smith (items 60 and 60A) 286 These emails concern efforts to put in place the amount of AU$50 million in security in the event that the injunction were not granted which we have mentioned above. The Commonwealth relies on a passage in it in which Mr Smith emailed Mr Fahner, copying in persons including Mr Kay and Mr Millichamp to say (item 60): The amount is AUD50 million. Senior management of Apotex are aware of and have previously approved this amount (see emails below and attached). 287 It also relies on Mr Fahner's response to Mr Smith's email which states (item 60A): Ok. Keep me informed. $50 million AUD is a lot of money and we will start the ball rolling to ensure the availability for an L/C in case it happens. 288 There is nothing in this submission. The trial judge set out this part of Mr Smith'semail in item 60 at J [269] and Mr Fahner's response email in item 60A at J [270].

Emails sent between 19-21 September 2007 (item 71) 289 This is a series of emails between Mr Smith and Mr Fahner concerning the bank guarantee. The trial judge set Mr Smith's email out at J [274]. As such there is no substance in the submission that the trial judge overlooked it.

Conclusions on the overlooked evidence 290 Most of the evidence pointed to by the Commonwealth as not having been considered by the trial judge was, in fact, considered by his Honour, and in some cases, in considerable detail. The matters which were not referred to by the trial judge were not material and we would infer that this is why his Honour did not refer to them.

(d) The fact that immediately before the hearing of the interlocutory injunction Mr Kay's only concern was whether a competitor could beat Apotex to a 1 April 2008 launch 291 The Commonwealth referred to two emails passing between Mr Millichamp and Mr Kay on 14 September 2007. The relevant text of Mr Kay's email is this: Does anyone else have an MA or PBS listing? My concern of a pre-emptive competitor launch is obvious given we are 7 months from launch. 292 Of this the trial judge said at J [268]: Mr Kay questioned whether anyone else had a TGA approval or PBS listing and stated that the only concern was "a pre-emptive competitor launch … given we are 7 months from launch". Mr Millichamp replied that no-one else had a product on the ARTG and, "[a]s a result no company, apart from us, can get a PBS listing yet". He also stated that "[o]ur strategy will be to launch ASAP, assuming no injunction, and ensure we sell in stock and lock in customers to optimize our revenues and block any potential competition for as long as we can". 293 The word 'only' is therefore something which was added by the trial judge. Mr Kay did not say that his only concern was a pre-emptive competitor launch. The Commonwealth's submission at AS [77] was that 'immediately before the hearing, Mr Kay's only concern was whether a competitor could beat Apotex to a 1 April 2008 launch'. It will be seen that this is not what the evidence shows. More importantly, even if it did, it cannot be said that the trial judge failed to take account of this matter.

Conclusions on suggested impermissible speculation as to whether Dr Sherman would have changed his mind if he had known that the hearing would be in April 2008 with a judgment by August 2008 294 The trial judge did not engage in speculation. In any event, we would reject the submission that the trial judge had failed to consider: (a) the 'compelling reasons' why Dr Sherman had made the decision to launch in the first place; (b) the interlocking steps which had been taken since then to launch at risk; (c) the actual evidence before his Honour about what had been done to justify a launch at risk; or (d) the fact that Mr Kay's only concern was whether a competitor could beat Apotex to a 1 April 2008 launch. The matters said to have been not considered were, in the main, considered by his Honour. To the extent that some were not, they were not material to the result.

The failure to suggest to Mr Millichamp that ss38-39 of his affidavit of 17 September 2007 was false and related materials 295 This issue arose during Mr Gleeson's address. Mr Millichamp swore another affidavit dated 17 September 2007. This was the affidavit put before Gyles J on the application for the interlocutory injunction. Perhaps confusingly, the paragraph numbers of this affidavit that are relevant are the same as were those of his affidavit of 7 July 2017, i.e. ss38-39. In Mr Millichamp's 17 September 2017 affidavit, these paragraphs are as follows: GenRx has provided sanofi-aventis with an undertaking not to have its clopidogrel 75mg (as hydrogen sulfate) tablets listed on the PBS until the outcome of sanofi-aventis's application for interlocutory relief has been determined, provided that any such application is commenced and determined by 11 October 2007. However, it is GenRx's intention to apply for listing of its clopidogrel 75mg (as hydrogen sulfate) tablets at the next available opportunity, which is by 1 December 2007. Other undertakings provided to Sanofi-Aventis Annexed to this affidavit marked RM-8 is a copy of a letter from Blake Dawson Waldron to Allens Arthur Robinson dated 14 September 2007, in which GenRx provides sanofi-aventis with certain undertakings in respect of any sales or other distribution of clopidogrel 75mg (as hydrogen sulfate) tablets. In paragraphs (2) and (3) of this letter, GenRx undertakes to refrain from providing any discount or material benefit to purchasers of clopidogrel tablets with specific discounts or material benefits for other medicaments which are also supplied by sanofi-aventis in certain circumstances. However, medicaments which are marketed under the Winthrop brand have been expressly excluded from these undertakings. 296 Mr Gleeson's submission went as follows: the case before Gyles J set the legal and practical context for the question of whether Apotex would launch at risk. In that context, Apotex was duty bound to be candid with Gyles J. Gyles J was solemnly informed by Apotex (through Mr Catterns) that unless restrained from supplying clopidogrel to the Australian market, Apotex would apply for a PBS listing on 1 December 2007: T24.10-15. Further, Mr Millichamp had sworn this to be so at s38 of his affidavit read before Gyles J. The solemnity of this statement was further underscored by the fact that Apotex had also proffered an undertaking to provide security of $50 million: T24.16-17. The thesis that Apotex had not yet decided to seek a PBS listing required for its acceptance the conclusion that Apotex, through Mr Millichamp, had wilfully misled Gyles J by telling his Honour that Apotex was going to seek the listing when, in fact, no such decision had been made. The idea that this is what had happened had not been put to Mr Millichamp. In particular, it was not put to him that he had sworn false evidence or had given false instructions to Mr Catterns. Since this had not been put to Mr Millichamp the trial judge was obliged to proceed on the basis that s39 of his affidavit was, in fact, true. Once that was accepted, the only inference which was open to the trial judge was that Dr Sherman had decided that if the injunction was not granted Apotex would seek PBS listing. 297 We do not accept this submission. It is important to be clear that the submission entails that the trial judge made a factual finding whose nature required it to be put to Mr Millichamp that s38 of his affidavit was knowingly false. The trial judge did not anywhere in his reasons make a finding that when Mr Millichamp swore his affidavit (or when the affidavit was subsequently read before Gyles J) he knew that Dr Sherman had already decided that Apotex would not seek a PBS listing even if the injunction were refused. In terms, therefore, there was no direct finding that the trial judge made that was required to be put to Mr Millichamp. 298 Inevitably therefore the Commonwealth's contention must be seen to rest on an implication which it says must be drawn from the trial judge's conclusions. In this case, what the trial judge found was that had the injunction not been granted, Mr Millichamp understood that he could not proceed to have Apotex seek PBS listing without first obtaining approval from Dr Sherman: J [249]-[251]. That then raised the question of whether Dr Sherman would have given that approval. On that question, the Commonwealth failed to satisfy the trial judge that Dr Sherman would have decided to seek PBS listing but for the injunction. Importantly, the trial judge did not find that Dr Sherman had already decided that Apotex would not seek PBS listing were the injunction refused. Rather, his Honour found only that the Commonwealth had failed to prove that Dr Sherman would have directed Apotex to seek a PBS listing if the injunction were refused. In this regard, it is to be noted that a finding that a party has failed to prove fact A does not entail, without more, a finding that fact 'not A' is the case. Although a trial judge may reason that a party has failed to prove fact A because fact not A is the case, this is not what happened in this case. In that regard, as will be seen, the reasoning of the trial judge was precise and astute. 299 The relevant matters found by the trial judge were therefore that: (a) Mr Millichamp knew that he could not seek a PBS listing without the approval of Dr Sherman; and (b) the Commonwealth had not proved that Dr Sherman would have decided to seek a PBS listing if the injunction had been refused. 300 One might perhaps infer from this that s38 of Mr Millichamp's affidavit was knowingly false to the extent that it omitted any reference to the fact that the decision maker was Dr Sherman. However, Sanofi made that submission at trial and the trial judge rejected it declining to find that Mr Millichamp's evidence in this affidavit was knowingly false on account of that omission. Obviously enough, the Commonwealth does not contest that conclusion. 301 The question which then arises is whether it must be inferred from (a) and (b) that s38 was knowingly false because Mr Millichamp knew that Dr Sherman had already decided that Apotex would not seek PBS listing if the injunction were refused. The trial judge rejected Mr Millichamp's evidence that Dr Sherman had already decided that Apotex would seek PBS listing if the injunction was refused. However, in the course of doing that, his Honour did not find that Mr Millichamp knew that Dr Sherman had decided this. In the absence of such a finding, it cannot be a necessary implication from his Honour's reasons that s38 of Mr Millichamp's affidavit was knowingly false and, as we have already observed, the trial judge made no express finding to that effect either. 302 In that circumstance, nothing the trial judge found required that it be put to Mr Millichamp that s38 of his affidavit was false. 303 Turning then to s39 of Mr Millichamp's affidavit, the Commonwealth did not really develop this point in writing. However, from Mr Gleeson's address it appears that the significance to be ascribed to it is that it refers to the letter sent by Apotex's solicitors, Blake Dawson Waldron, to Sanofi's solicitors, Allens Arthur Robinson, on 14 September 2007. It is convenient to assume that the Commonwealth is entitled to rely upon the letter of 14 September 2007 as being evidence which the trial judge overlooked (although the Chronology suggests that it was is not one of the documents upon which the Commonwealth relies since it is not shaded blue). 304 Mr Gleeson submitted that the letter's significance was that it referred to the undertaking Apotex had proffered to Sanofi to provide the security of $50 million and to undertakings it had also proffered to keep track of its sales if the injunction were refused: T28.36-41. As we understood it, the point was these were solemn inter partes undertakings and their solemnity was inconsistent with any conclusion but that Apotex had decided to seek PBS listing and launch if the injunction were refused. There are two problems with this submission. First, as the Chronology makes clear, the Commonwealth made no submission at trial about this letter. Secondly, even if it had, the trial judge considered both of these matters for at J [272] he recorded that at the hearing of the interlocutory injunction 'Apotex offered security of A$50 million, to be increased as appropriate "as the proceeding progresses". Apotex also offered to maintain comprehensive accounting records of all clopidogrel products sold'. 305 It may be said against this that the trial judge did not advert to the fact that these offers took the form of an inter partes undertaking which had been made in the letter of 14 September 2007. However, we do not accept that the fact that the offers took the form of an inter partes undertaking in any way undermines the trial judge's conclusions. In particular, we do not accept that the giving of such undertakings was any more inconsistent with Dr Sherman not yet having made up his mind than the offers to which the trial judge did refer.

Submissions concerning overlooked submissions 306 Although the Commonwealth did not identify in its written submissions the detail of its case in relation to submissions which it alleged had been overlooked by the trial judge, Mr Gleeson spent some time on this topic in his address: T66.44-69.44. He submitted that the trial judge had failed to deal with a number of paragraphs of the Commonwealth's closing submissions in writing. The paragraphs identified were [129]-[134], [140]-[144], [158]-[166], and [174]: T67.2-8. He also submitted at T67.16-18 that the trial judge had overlooked oral submissions made about emails of late-June extracted at document B2.720.T0957 at pp 995-997. 307 Mr Gleeson also developed further a submission during oral argument that the trial judge had clearly considered that it might be commercially humiliating for Apotex successfully to resist the injunction and then not launch: T67.42-45. He also submitted that comments made by the trial judge at that time indicated that he was alive to the inconsistency between no decision having been made to launch and the fact that Mr Catterns hold told the Court that Apotex would launch unless restrained with Mr Millichamp sitting in Court: T68.5-16. 308 The end point of this suite of submissions by Mr Gleeson was a contention that it was apparent that the trial judge had overlooked material documents and submissions, yet at the time of the hearing it appeared that the trial judge was making observations which were supportive of the Commonwealth's position: T69.37-43. Although Mr Gleeson did not say this, the sub silentio suggestion was that the change in the judge's views of the matter between the time of the hearing and the production of his judgment was to be explained by the delay in the production of the reasons, that point being buttressed by the failure to advert to material evidence and submissions. 309 In this section we therefore deal with: (a) the suggested failure of the trial judge to advert to certain paragraphs of the Commonwealth's closing written submissions and its oral submissions at trial about the late-June emails; (b) the suggested failure of the trial judge to deal with: (i) the commercial humiliation that Apotex would face were it successfully to resist the injunction application and then not launch; (ii) the inconsistency between Mr Catterns' statement to the Court that Apotex would launch unless restrained (and Mr Millichamp's evidence to that effect) and Apotex not having made such a decision; and (c) the significance of the trial judge's observations during argument.

(a) Failure to advert to submissions 310 Five sets of submissions are said to have been overlooked. Dealing with each in turn:

Paragraphs [129]-[134] of the Commonwealth's closing submissions 311 Paragraph [129] is as follows: Secondly, on 28 June 2007 at 6.58pm, Mr Millichamp forwarded the 5.12am 27 June email from Dr Sherman, which is redacted for privilege, to his Australian colleagues, Karen McTavish, Paddy Smith, Bryan Moore, Marnie Peterson, Sylvie Laroche and Ms Keast (a copy recipient of Dr Sherman's emails of 27 June 2007), with the words "FYI - game on !!!" Ms McTavish responded, asking "So have they now been notified?", to which Mr Millichamp replied "Karen. Not yet." 312 The submission that [129] was not considered by the trial judge is wrong. The submission refers to Mr Millichamp's email of 28 June 2007 sent at 6.58 pm in which he said 'Game on !!!'. The trial judge dealt with this email at J [245]. 313 Paragraph [130] is in these terms: Thirdly, on 28 June 2007, at 10.28pm, Mr Millichamp sent an email to Mr Haas, copied to Mr Smith, Ms Keast, Ms Peterson and Ms McTavish, which said: [redacted] as per instructions from Barry the plan (in outline) for clopidogrel is as follows: 1) Seek to revoke the Sanofi patents at time of TGA approval - circa end July start August this year. 2) Sanofi will apply for interlocutory relief. If they get it and we are enjoined we will not be able to launch until after final trial (if we are successful) which could be 18 months away from now or more. 3) If we are successful in defending our position vs Sanofi and an injunction is not granted by the courts then we will launch. 4) We will not be able to get PBS listing until earliest December 07. 5) Therefore if we are successful and are able to launch we will need product (blister packed by Apotex) in Australia ideally end October but latest Mid November. 6) Initial quantities and follow up / replenishment orders etc can all be discussed and arranged in a way that is best for you bearing in mind timings and blister constraints. Could you please continue to work with Marnie, Paddy and Karen etc on the packaging options. Please confirm when you would like purchase orders raised. We are not sure that we can launch yet so need to know if we should place orders in anticipation or should wait. Barry has made it clear that he does not want to waste money on launch activities until we are clear on what we can do. At our next demand meeting we should talk though the various options please. (emphasis in original) 314 This sets out Mr Millichamp's email of 28 June 2007 sent at 10.28 pm. This email was not referred to by the trial judge. However, as we have explained above at [141] we do not think this email was material once the full scope of the trial judge's reasoning is grasped. Consequently, we do not think that anything turns on the fact that the trial judge did not deal with [130]. 315 Paragraph [131] is as follows: Although we have not seen Dr Sherman's email, Mr Millichamp's emails only sensibly permit one conclusion: Dr Sherman had instructed Apotex to launch at risk, by commencing proceedings to revoke Apotex's patent at or close to the time that ARTG registration was achieved, putting Sanofi on notice that Apotex intended to launch unless restrained by an injunction and to launch at risk if not restrained. 316 This contains a recitation of the Commonwealth's case that Dr Sherman had instructed Apotex to launch at risk. However, the trial judge was aware that this was the Commonwealth's case. Indeed, that was the central proposition which occupied much of his Honour's reasons and it was that case which the trial judge rejected. 317 Paragraph [132] is in these terms: Further confirmation can be seen in an email from Mr Millichamp to Mr Haas on 29 June 2007, which was copied to Dr Desai and Mr Kay among others. They were recipients of Dr Sherman's emails of 27 June 2007. The context was that Dr Desai had sent an email to certain officers of Apotex India on 28 June 2007, which stated "we will not be making any launch stocks of the bisulfate salt even though we expect approval in the next couple of months" (presumably because it appeared to be inconsistent with Mr Millichamp's stated understanding of Dr Sherman's instructions). Mr Haas emailed Mr Millichamp on 29 June 2007 in order to draw his (Mr Millichamp's) attention to Dr Desai's comment. Mr Haas asked Mr Millichamp to "clarify to ensure everyone (R&D, Execs, Demand etc) is on the same page". 318 This refers to Mr Millichamp's email to Mr Haas of 29 June 2007 copied, inter alia, to Dr Desai and Mr Kay (i.e. item 31) which responded to Mr Haas' request for clarification upon receiving Dr Desai's email concerning the bisulfate salt (i.e. item 30A). We have explained above why both of these emails were not materials which the trial judge needed to deal with. For the same reasons, we do not think anything turns on the fact that the trial judge did not advert to the submission at [132]. 319 Paragraphs [133]-[134] are in these terms: Mr Millichamp responded: I was probably not clear enough in my mail to you. We are not requesting launch stocks to be made before we commence litigation (in the next month or so) and start selling to customers as we have done with previous launches e.g. Perindopril and Carvedilol. [redacted] The scenario I referred to was in the event that we avoid the imposition of an injunction granted in favour of Sanofi. We will be free to sell until final trial, and after, if we are successful. In the instance that we are successful we will need stock ready for a December PBS listing. [redacted] Thus if we are successful we will go ahead and launch (the bisulphate salt). If we are not successful in avoiding an injunction then the matter will go to final trial which could be a year or so (or longer) away. That's why Barry does not want us to make stock right now and incur unnecessary expense. [emphasis added] Again, that email is consistent only with an intention to launch clopidogrel products, while delaying production of stock for a 1 December 2007 launch until the outcome of the interlocutory injunction was known. (emphasis in original) 320 They extract Mr Millichamp's email referred to in the previous paragraph (i.e. item 31) and therefore add nothing to the submission in [132]. As we have explained above, we do not accept that this was material which the trial judge was bound to examine in his reasons.

Paragraphs [140]-[144] of the Commonwealth's closing submissions 321 Paragraphs [139]-[140] are in these terms: Later that day, Dr Sherman replied, saying "We should put Sanofi on notice ASAP. Who is taking care of it?" Mr Millichamp's evidence that that email reflected a desire expressed about 10 days prior by Dr Sherman was not challenged. Nor was his evidence that he: had no doubts that Dr Sherman was supportive of the plan to launch the Apotex Registered Clopidogrel Products at risk in Australia in the event that Apotex Australia was not subject to an interlocutory injunction, and understood that he was asking someone in the business to put Sanofi on notice of Apotex Australia's intention to launch its generic clopidogrel products as soon as Apotex Australia obtained registration of them on the ARTG. 322 The Commonwealth did not submit that the trial judge had not dealt with [139] but it is necessary to consider it in order to understand the submission at [140]. Dr Sherman's email saying that 'We should put Sanofi on notice ASAP. Who is taking care of it?' was dealt with by the trial judge at J [254]. The Commonwealth did not develop its submission about [140] beyond the fact that it was not considered. The submission relates to two aspects of Mr Millichamp's evidence. We were not taken to this evidence and it does not form part of the Commonwealth's case on overlooked evidence. Assuming Mr Millichamp's evidence was that Dr Sherman had said to Mr Millichamp 10 days earlier that Apotex should put Sanofi on notice, we fail to see what this adds to the trial judge's treatment of the Dr Sherman's email at J [254]. As we have explained above, what really matters about Dr Sherman's email is that it was a response to Mr Kay's earlier email in which he told Dr Sherman that Apotex expected to lose the injunction application. The absence of any response to that proposition in Dr Sherman's email led the trial judge to conclude at J [255] that Dr Sherman was expecting to lose the application. The fact that Dr Sherman may have said something to Mr Millichamp 10 days earlier is irrelevant to that observation. 323 As for the quoted portion of Mr Millichamp's evidence set out in [140], we fail to see how this undermines any aspect of the trial judge's conclusion that Dr Sherman was expecting to lose the application. 324 Paragraph [141] is as follows: Contrary to Mr Kay's view, Mr Millichamp's expressed view at the time contemplated that Sanofi might not obtain an interlocutory injunction. On 9 August 2007, Mr Millichamp responded to an email from Mr Kay, which is redacted and which responded to Dr Sherman's 8 August email asking "Who is taking care of it?" Mr Millichamp sent his email to recipients including Dr Sherman, Dr Desai, Mr Haas, Mr Caccamo and Mr Kay. The unredacted portion reads: If Sanofi are successful and get an injunction then we will, of course, not launch and the legal process will run its course in court. If we are successful and defend against an application for interlocutory relief from Sanofi we would like to launch as soon as possible. The earliest date that we can get PBS (Government reimbursement) listing is Dec 07. We asked Stephen Hass to prepare us a launch plan so that in the event we are successful we can launch as soon as possible. We were not expecting any stock to be made or packed until we know the outcome of the hearing for interlocutory relief which should be approximately 2 to 6 weeks after we file for revocation. 325 However, the text of the email set out does not show that it was Mr Millichamp's view that Sanofi might fail to get an injunction; he was charting different courses of action depending on an unknown outcome. In any event, the critical issue is what Dr Sherman thought and this the trial judge dealt with at J [255]. 326 Paragraphs [142]-[143] are in these terms: Mr Millichamp's email closed by asking Mr Kay to advise if the plan articulated in his email was "an acceptable course of action". Mr Haas' response was "thank you for providing further clarity". Mr Kay's response (copied to all recipients, including Dr Sherman) was "Thanks for the comprehensive update. OK by me", while Mr Caccamo's (again copied to all recipients including Dr Sherman) was "We have aligned our launch plans accordingly". It was not put to Mr Millichamp that Dr Sherman had countermanded or corrected Mr Millichamp's understanding as expressed in the email. Rather, Mr Millichamp's evidence that he did not recall any later instructions from Dr Sherman modifying his instructions should be accepted. 327 We have dealt with each of the emails mentioned in these paragraphs and explained why they were not material. In that circumstance, we detect no error in the trial judge not adverting to these paragraphs. 328 Paragraph [144] is as follows: As at this point, it is clear that: (a) Apotex Canada (Dr Sherman) had instructed Apotex (Mr Millichamp) to commence proceedings against Sanofi to revoke its patent, notify Sanofi that Apotex intended to launch at risk if not restrained, and to launch at risk if not restrained; (b) Dr Sherman was pressing Apotex to implement those instructions ("We should put Sanofi on notice ASAP. Who is taking care of it?"); (c) Mr Millichamp had no doubt about Dr Sherman's instructions and intended to implement them; (d) consistently with those instructions, Apotex had requested that Apotex Canada adjust supply plans to ensure that Apotex Australia had sufficient stock to launch at risk, if not restrained; and (e) Apotex Canada had agreed to adjust those plans accordingly. 329 We do not accept that the trial judge did not deal with this case. He rejected proposition (a) on the facts. He dealt with the email in (b) but observed that the email from Mr Kay which preceded it showed that Dr Sherman was expecting to lose the injunction application. His Honour was aware that it was Mr Millichamp's evidence that he had been instructed to launch at risk as suggested in (c) but rejected that case. As to (d) and (e), as we have explained above, the 'detailed arrangements' for supply were in fact meagre.

Paragraphs [158]-[166] of the Commonwealth's closing submissions 330 Paragraphs [158]-[159] are in these terms: Mr Millichamp's view of the document was exactly aligned with that taken by Mr Moulding, the Managing Director of Sanofi at the time. On 21 August 2007, Mr Martin (Sanofi's "State Sales Manager/KAM - Pharmacy Qld") forwarded Apotex's undated "team" letter to others in Sanofi. It was forwarded by Mr Dick to Mr Moulding, who responded to Mr Dick, Mr Gracie, Ms Allsopp and Mr Brindell: Dear all, This seems to be very clear evidence of intention to launch and the urgency for an interim injunction - Rebecca do you have any more updates on this? It should be noted that Mr Martin's original email, sent just days after Sanofi received the letter, also stated that "You may get asked questions at pharmacy level re this announcement, at this stage make your accounts aware that we are in the process of putting an injunction on this pending launch, until the patent challenge is heard". That corroborated Mr Millichamp's belief that Sanofi would move for such relief. (Mr Millichamp expressed that view on at least two other contemporaneous occasions, in emails on 9 August and 29 August.) (emphasis in original) 331 We have already explained why we do not think that the views of the Managing Director of Sanofi throw any light on what Dr Sherman thought. As such, the fact that the trial judge did not deal with these two paragraphs discloses no error. 332 Paragraph [160] is as follows: Mr Millichamp was questioned about linguistic subtleties of the short 20 February 2007 email of Dr Sherman's, and in particular the sentence "If they do not give an undertaking for our damages and do not get an injunction, we will launch." Mr Millichamp's evidence that he understood that "If the injunction was granted, we couldn't launch. If an injunction wasn't granted, we would launch at risk" and "if an interlocutory injunction wasn't applied for, or was applied for and we were free to launch, we would have launched" should be accepted. That evidence was not challenged, and is the natural reading of the message. It comports with Mr Millichamp's evidence that he expected, at all times, that Sanofi would seek interlocutory relief and that "The assumptions were basically no injunction applied for, so we launch at risk or injunction applied for successful, so we can't launch. Unsuccessful and we launch. That was the initial assumptions from Dr Sherman and that basically played through." 333 The trial judge concluded that the February email had been overtaken by events. Most importantly, the email in which Mr Millichamp said that any launch was subject to Dr Sherman's further approval rendered debate about the meaning of the February email beside the point. There was no need for the trial judge to deal with this paragraph. 334 Paragraphs [161]-[164] are in these terms: Mr Millichamp's evidence that he understood the purpose of writing to Sanofi was "at the earliest possible stage to make them aware that we were launching, such that if litigation were to commence, it would commence at the earliest possible time and that we were to be completely honest and open and transparent with our intentions" should be accepted, as should his evidence that the reason he wanted litigation to commence at the earliest possible time was because "we wanted to get on the market as early as we possibly could". Mr Millichamp's reference to "the alendronate litigation" where "the judge was making it very clear that we should be very transparent and open and clear on our intentions" appears to be a reference to Moore J's comments in Merck & Co Inc v GenRx Pty Ltd [2006] FCA 1407 at [18]-[19]. Moore J stated, inter alia, that the balance of convenience warranted the grant of interlocutory relief and that GenRx had "deliberately" set out about making commercial arrangements to import and sell alendronate products subject to Merck's patent but had not brought "revocation proceedings", and had placed itself in this position "quite deliberately and knowingly. To use the language of some of the authorities, its 'eyes were wide open'." That oral evidence was consistent with Mr Millichamp's own statements in an email of 25 June 2007, where he said: * We propose notifying Sanofi of our planned launch by applying to revoke their patent either on the same day that we launch or a few days earlier. The idea is to notify Sanofi as soon as we know for sure that we can launch, i.e. when we know that TGA approval will be granted but not give Sanofi time to try and block our initial launch efforts. * We will submit our application for PBS listing once we are sure that no injunction will be imposed by the courts (PBS listing is granted roughly three months after initial application). We will, of course, have to get to court before we know this outcome. If we apply for PBS listing as soon as we launch we will trigger the 12.5% price drop and Sanofi will argue that pricing has been dropped and therefore their damages cannot be readily measured, will cause irreparable harm and the decision cannot be reversed etc. When we get to court we will need to let the judge know that we plan to PBS list. In court the idea, in relation to the 12.5 % drop and further launch activities, is that we will argue that damages are an adequate remedy, give undertakings that we will record all sales, and provide a bank guarantee in order to pay costs and damages in the event that we are not successful at final trial. We will have to give a bank guarantee that is effective in Australia, but supported by Apotex, as GenRx have no assets or profits at this stage in this jurisdiction to prove that we can pay out in the event of a Sanofi success. * It is difficult to forecast the amount that we would have to put up as a bank guarantee but we estimate that it could be in the range of 50-70 Million AUD (for one year's damages) depending on what Sanofi argues in court. [emphasis added] By launch efforts, he meant taking orders. (emphasis in original) 335 Once the trial judge concluded that Dr Sherman had indicated that Sanofi should be put on notice and that Dr Sherman expected to lose the injunction application, these submissions become beside the point. What Mr Millichamp thought did not really matter once the central decision maker was identified as Dr Sherman. The trial judge did not err in not dealing with these submissions. 336 Paragraph [165]-[166] are as follows: On 24 August 2007, Allens (for Sanofi) wrote to Blake Dawson Waldron (for Apotex) asserting that Mr Millichamp's 17 September 2007 letter to Sanofi and his 17 August 2007 letter distributed to pharmacists constituted "a threat to engage in conduct in breach of" Sanofi's rights under the Patent. Allens demanded three undertakings. Blake Dawson Waldron's response of 28 August 2007 declined to give the first of the undertakings and said that Apotex could not give the other two. It included the statement: "For the purpose of any interlocutory injunction hearing only, our client concedes that GenRx Clopidogrel 75mg tablets would fall within claims 1, 3, 10 and 11 of the patent in suit". 337 That there was correspondence between the solicitors immediately prior to the litigation does not appear to us to impact on the correctness of the trial judge's analysis. We do not think that his Honour erred in not dealing with this submission.

Paragraph [174] of the Commonwealth's closing submissions 338 This paragraph is as follows: On 3 September 2007, Allens (acting for Sanofi) replied to Blake Dawson's letter of 28 August 2007. Relevantly, Allens stated: [Apotex] has already engaged in conduct in breach of Sanofi-Aventis' exclusive rights under section 13 of the Patents Act 1990 (Cth). GenRx's undated letter to pharmacists (a copy of which was attached to our letter of 24 August 2007 marked 'A') constitutes an offer to either sell, or otherwise dispose of, GenRx Clopidogrel. In any event, it is clear that your client has threatened, unless restrained by the Court, to import GenRx Clopidogrel and to sell it in Australia. That threat was clear from both the letter of 17 August 2007 from GenRx Pty Ltd (GenRx) to Sanofi-Aventis' Australian subsidiary, and GenRx's second undated letter to pharmacists. 339 As with the August 2007 letters above, we do not think that the correspondence passing between the solicitors in September 2007 was material with which the trial judge had to deal.

The oral submission at T995-997 [B2.720.T0957] 340 As Mr Gleeson explained it, in this portion of the transcript, counsel for the Commonwealth at trial dealt with the late-June emails: T67.16-18. The relevant portion of the trial transcript is T995-997. Counsel for the Commonwealth did indeed take the trial judge to a number of documents at T995-997. Mr Gleeson did not however identify any particular document at T995-997 which was additional to the Commonwealth's case on the documents it says were overlooked by the trial judge. As such, we have already dealt with these emails and concluded either that the trial judge did refer to them or that they were not material. To the extent that there is anything additional at T995-997, it was not drawn to our attention and we do not propose to consider it.

(b) Failure to deal with commercial humiliation and the significance of the statements made by Mr Catterns and the evidence of Mr Millichamp 341 The trial judge dealt with Mr Catterns' statement that Apotex would launch at risk unless restrained at J [84] and J [272]. Above we have dealt with the Commonwealth's contention that it needed to be put to Mr Millichamp that his affidavit in which he said that Apotex proposed to launch at risk was false. The present submission is a variant of this contention. The point now is not the seriousness of the finding inherent in concluding that it was not Apotex's intention to launch at risk when both Mr Catterns had said (on instructions) and Mr Millichamp had sworn that it was Apotex's intention. Rather, the point is the commercial humiliation that would have arisen in the same situation. As we understood the point, the contention was that to inform the market and Gyles J that Apotex was going to launch and then, having successfully resisted Sanofi's injunction, not do so would have resulted in commercial humiliation. As Mr Gleeson put it this way during the appeal (T65.2-5): You just don't see a square confrontation with the question: how can [Apotex] win consistently with the commercial humiliation and the impugning of their conduct in the court that their argument connotes. 342 It is apparent that the trial judge did not, in terms, deal with this contention. The first question is whether such an argument was put to him. The Commonwealth's submission that the argument had been put was based on a portion of the transcript of the closing addresses in which a discussion about commercial humiliation took place between the trial judge and Mr Sheahan QC (who appeared for Sanofi). To be clear, this discussion was not with the Commonwealth's counsel. At T1111.1-.6 of the trial transcript, the trial judge said this: Well, the fact that they proceeded to argue the application, didn't merely proffer undertakings and take the early hearing date, I strongly suggest that they had turned their mind to it. I mean, it would be commercially humiliating to have the judge refuse the injunction and then go out in the marketplace and say we're not launching until some time after this litigation is resolved or late the following year. 343 Subsequently, Mr Brereton SC (the Commonwealth's senior counsel at trial) said this at T1215.11-34 of the trial transcript: And your Honour was asking Mr Sheahan some questions about this and about - your Honour pointed out: … be commercially humiliating to have the judge refuse the injunction and then not go out into the marketplace - That was your Honour's question at line 4 and 5. And then Mr Sheahan, about line 15 and following, gave some reasons to fight the injunction. And for the most part, they seem pretty plausible. You - he put a value on being seen to be aggressive, for example: … you want to improve your negotiating position for the patentee - Etcetera. They are all good reasons why you would proceed to fight the injunction but they're good reasons why you would proceed to fight the injunction and, if you were successful, to proceed to launch. They're not good reasons to fight the injunction, win and then not launch. How does that play out in terms of being seen to be aggressive in challenging patentees? You fight it, you lose and then you don't launch. That would be the worst message. And it would also involve, again, Mr Catterns - and … obviously … would - saying something to the court that's not true. If it was at all qualified, they - they thought to themselves, there's a new piece of information, we're not going to get listed till 1 January, we're not sure what we're going to do - 344 The first question is whether it is to be inferred from what Mr Brereton said that the Commonwealth had in fact advanced a case based on commercial humiliation. We do not think that it can. The submission is about the reasons advanced by Mr Sheahan about why one might resist the injunction. The endpoint of the submission was that these reasons were 'not good reasons to fight the injunction, win and then not launch'. The only relevance of Mr Brereton's reference to the trial judge's earlier comment about commercial humiliation was as a pinpoint to where Mr Sheahan set out his reasons. As such, we do not accept that the Commonwealth advanced a case based upon commercial humiliation. Consequently, we do not accept that the trial judge was required to deal with such a contention.

(c) The significance of the trial judge's comments during argument 345 Having taken the Court to a variety of comments made by the trial judge which appeared to favour the Commonwealth during final submissions, Mr Gleeson then took the Court to a number of remarks made by counsel for Sanofi at the same time which he described as 'on their face not strong': T69.42. The submission was then this at T69.37-44: MR GLEESON: … So, your Honours, what we seem to have, at least we submit, is we've got material documents not considered, even though they have been squarely addressed upon. We have in the final submissions at a time when the matter was acutely before the judge's mind a square focus on some of the key matters which suggest that the Commonwealth's case was very probable. We have some answers to that case by Sanofi, which on their face not strong. And yet these ultimate issues have, I put respectfully, disappeared from square focus in the judgment. In that sense, we're asking the court to intervene. 346 The difficulty is that the 'ultimate issue' insofar as the trial judge's remarks are concerned were not advanced by the Commonwealth. The reference to the 'material documents not considered' has been dealt with above. In short, the trial judge considered all the documents which were material. To the extent that there is anything else in this submission, it consists only of observations made by the trial judge and Mr Brereton's fleeting reference to those observations. We do not think the trial judge was required to deal with his own musings during argument unless they were made its own by the Commonwealth. As we have explained above, we do not think that they were made the subject of a submission by the Commonwealth.

Conclusions on Ground 2(b) 347 The Commonwealth fails to make good this ground.

GROUND 2(C): WAS THE JONES V DUNKEL INFERENCE WRONGLY DRAWN? 348 Particular (c) was as follows: (c) The primary judge erred in: (i) concluding that the Commonwealth's case suffered an evidentiary deficiency which could only be made good by calling Dr Sherman (at [349]); (ii) concluding that a Jones v Dunkel inference was open to be drawn, and should be drawn, against the Commonwealth in respect of Dr Sherman (at [344]-[347]); (iii) failing to address, either adequately or at all, submissions of the Commonwealth in relation thereto. 349 At trial, Sanofi submitted that a Jones v Dunkel inference should be drawn against the Commonwealth for its failure to call a number of witnesses including Dr Sherman: J [164]. At J [165]-[173], the trial judge set out the general principles applicable and summarised them at J [170] in these terms: The three conditions that must be fulfilled before the rule can operate concern, first, the relationship of the party to the missing witness and (inter alia) whether the missing witness would be expected to be available to one party rather than the other; second, whether it appears that the missing witness is likely to have knowledge of the relevant facts; and, third, whether any satisfactory explanation has been given by the relevant party in respect of its failure to call the missing witness. 350 His Honour then dealt with the position of Dr Sherman at J [343]-[347] which was in these terms: Sanofi BMS submitted that the Commonwealth's failure to call Dr Sherman should lead the Court to infer that his evidence would not have assisted the Commonwealth's case. This brings me to the question whether, in respect of Dr Sherman, the three conditions referred to by Glass JA in Payne v Parker are satisfied. In this case I would have expected Dr Sherman to have been available to the Commonwealth rather than to Sanofi BMS. Contrary to a submission made by the Commonwealth, I think it would be quite unrealistic to have expected Sanofi BMS to have called Dr Sherman in its defence of the Commonwealth's claim. Apotex Canada and Sanofi BMS were at relevant times trade rivals engaged in significant litigation in at least three jurisdictions. The US litigation culminated in a judgment in favour of Sanofi BMS against members of the Apotex group (including Apotex Canada) for in excess of US$442 million. In Australia, the Apotex companies made claims against Sanofi BMS in excess of A$138 million under the same undertakings as to damages upon which the Commonwealth's case is based. And in Canada, Apotex Canada was found to have engaged in infringement of Sanofi's Canadian Patent on what was likely to have been a significant scale by manufacturing and exporting clopidogrel products for the US market. In my view it would be unreasonable and contrary to common sense to have expected Sanofi BMS to call Dr Sherman. The fact that Apotex Canada made Mr Fahner available to make an affidavit for the Commonwealth and to travel to Australia for cross-examination raises the obvious question of why the Commonwealth did not seek to call evidence from Dr Sherman on an issue of critical importance to its case. In circumstances where Mr Fahner, a senior executive at Apotex Canada, was called by the Commonwealth, it seems to me it would be reasonable to expect it to have also called Dr Sherman to give direct evidence of his state of mind at relevant times by affidavit and, if required, orally either in person or perhaps, by video link if he was unwilling or unable to travel to Australia. There was no evidence called by the Commonwealth that would provide any explanation as to why it did not call Dr Sherman or what attempts, if any, it made to call him. I am mindful that Dr Sherman was, at the time of the trial, based in Canada. However, as I have mentioned, this did not prevent the Commonwealth calling Mr Fahner or other Apotex group executives resident overseas to give evidence in support of its case. I conclude that Dr Sherman was a witness who I would have expected to have been available to the Commonwealth and who would have had a close knowledge of relevant facts. In circumstances where the Commonwealth's decision not to call Dr Sherman was wholly unexplained, I infer that the Commonwealth chose not to call him because it considered that his evidence would not have assisted its case. 351 Of significance is his Honour's observation that the Commonwealth had called an Apotex witness from Canada, Mr Fahner. It is evident that the absence of any explanation for why Dr Sherman was not called impressed the trial judge. While it is easy to imagine that Dr Sherman might have told the Commonwealth that he was not willing to travel to Australia to give evidence, the absence of any evidence to that effect provides material from which, on its face at least, it was open to the trial judge to conclude that the Commonwealth decided not to call Dr Sherman because it did not think that it was to its advantage to do so. 352 Having laid that groundwork, the trial judge then reasoned this way at J [348]-[349]: I am not prepared to infer, based on the 20 February 2007 email, or any of the subsequent correspondence in evidence which was said to justify the drawing of such an inference, that Dr Sherman was likely to have instructed Mr Millichamp to procure the listing of Apotex's clopidogrel products with effect from 1 April 2008. In my opinion, the Commonwealth's case suffers from an evidentiary deficiency which cannot be made good by drawing inferences from correspondence written by Dr Sherman in the lead up to the hearing of the interlocutory application. In particular, I do not think it can be inferred that if Dr Sherman had known that the trial of the patent proceeding would commence in the same month that Apotex Australia obtained a PBS listing of its clopidogrel products (triggering a 12.5% statutory price reduction), that he would have, in those circumstances, authorised Apotex Australia to obtain such a listing before judgment was delivered or, at least, until the trial had concluded (by which time he and his colleagues and his legal advisers may have had a clearer view of the strength of Sanofi's case). 353 In its written submissions, the Commonwealth identified four errors in this reasoning: AS [81]-[90]. The first and second contentions were effectively the same and were that the trial judge had erred in thinking that there was a situation in which Jones v Dunkel could apply. The Commonwealth submitted that the trial judge was required to consider whether it should be inferred from the documents it relied upon that Dr Sherman would have decided to launch at risk if the injunction had been refused. The trial judge had erred in thinking that he could decline to draw an inference from the documents just because Dr Sherman had not been called. 354 We do not accept this submission. The trial judge did not decline to draw inferences because Dr Sherman was not called. Rather, his Honour felt that the documents upon which the Commonwealth relied could not, in themselves, sustain any inference about what Dr Sherman would have done if the injunction were refused. The reason this was so was because by the time the injunction was granted, it was known that there would be an early trial in April 2008 and that this was the same month that Apotex would have obtained its PBS listing. These two matters were not known at the time of any of the correspondence upon which the Commonwealth relied. 355 Assuming for the sake of argument that these pre-injunction documents, viewed in pure isolation, could support an inference that Dr Sherman would have launched at risk, the problem is that the question the trial judge had to decide did not arise in such a circumstance of pure isolation. Rather, it arose in a factual milieu which necessarily now included additional facts that Dr Sherman would have known, i.e. that there was to be a trial and PBS listing in April 2008. The trial judge's point is that no amount of scrutiny of the prior correspondence can provide any assistance in determining what Dr Sherman's reaction to these unanticipated developments would have been. It is this problem which is the evidentiary deficiency to which his Honour referred at J [349]. 356 As such, his Honour was correct to identify this as an evidential deficiency from the Commonwealth's perspective. Once that is accepted, it follows that the Commonwealth's first and second submissions must be rejected. The trial judge did not utilise Jones v Dunkel to decline to draw an inference from the pre-injunction documents. Rather, he thought those documents were insufficient to draw any inference from in light of the developments which had occurred. 357 The Commonwealth's third contention was that the trial judge had failed to deal with the question of which party bore the onus of demonstrating that Dr Sherman would be expected to be called: AS [84]-[85]. It submitted that the party seeking the drawing of the Jones v Dunkel inference bore the onus of proving this matter. That party was Sanofi. 358 It may, we think, be assumed that as the party seeking the drawing of the Jones v Dunkel inference, it was for Sanofi, rather than the Commonwealth, to demonstrate that circumstances enlivening the rule were present and that the inference should be drawn. This necessitated that Sanofi demonstrate both that the evidence of Dr Sherman was likely to be relevant to the determination of a fact in issue, and that Dr Sherman was relevantly in the camp of the Commonwealth. 359 The trial judge was evidently persuaded of both matters. On appeal, the Commonwealth took issue with both conclusions. As to the relevance of Dr Sherman, the Commonwealth submitted (as part of its fourth contention at AS [86]-[87]) that Apotex had successfully claimed legal professional privilege over substantial parts of its internal correspondence. This mattered because whilst the Commonwealth accepted in its written submissions that Dr Sherman could have been asked whether he would have launched at risk in the counterfactual, he would not have been able to have been asked any questions about the material over which privilege had been claimed. His answer would therefore have been a hindsight answer made 10 years after the event and divorced from the contemporaneous documentary material (which was privileged): AS [88]. 360 Taking this submission at its highest, it is not a submission that Dr Sherman's evidence about whether he would have launched at risk in the counterfactual was irrelevant. Rather, it is a submission that, despite being relevant, its significance was to be downplayed. As such, it cannot be said to have been erroneous for the trial judge to have approached the matter on the basis that the evidence of Dr Sherman would have been relevant. 361 In any event, the Commonwealth's submission greatly overstates the effect of Apotex's claims for privilege. Whilst much of the correspondence was masked for privilege, much of it was not. It is difficult to see how the Commonwealth's submission - that the available material which was not masked for privilege could lead only to the inference that Dr Sherman would have launched at risk - can be reconciled with its present submission that Dr Sherman could have been asked nothing useful about that same material. 362 In his address, Mr Gleeson developed this submission by drawing an analogy between Dr Sherman and the position of Mr Robb in Australian Securities and Investments Commission v Hellicar [2012] HCA 17; 247 CLR 345 at [168] ('Hellicar'). In Hellicar, ASIC had not called Mr Robb and the Court of Appeal had drawn a Jones v Dunkel inference against ASIC. At [162] the High Court explained the evidence he might have given in this way: The respondents submitted that it should not be inferred from the rendering of this bill that Mr Robb looked again at the minutes after the February meeting. The better view of the evidence would require rejection of that submission. But even if it were accepted, the evidence that Mr Robb could be expected to give of what was said or done at the February meeting would have to be assessed in light of the part he had played in preparation of the minutes that were ultimately adopted by the board as a correct record of what was decided at the February meeting. It could be expected that Mr Robb could have given evidence to the effect (a) that he recalled what was said and done at the meeting and it accorded with the minutes which had been drafted and settled under his supervision, or (b) that he had no independent recollection of some or all of what transpired at that meeting, or (c) some mixture of the two. It seems unlikely that he would say that he had not seen, in any of the successive drafts, the reference to tabling and adoption of a draft ASX announcement, a copy of which, as the respondents emphasised for other reasons, he had been given and he had annotated. It seems very unlikely that he would say, contrary to his own interests and the interests of the firm of which he was a member at the relevant times, that he positively remembered that what was recorded in the minutes was untrue. 363 It then reasoned this way at [168] and [169]: The Court of Appeal concluded that ASIC's not calling Mr Robb founded an inference that his evidence "would not have assisted the ASIC case". There was no basis for drawing any inference that Mr Robb would have given evidence adverse to ASIC's case. More particularly, there was no basis for concluding that it was more likely than not that he would say that the minutes whose preparation he had supervised were false. The most that could be inferred from Mr Robb not being called by ASIC was that he could not give evidence, from his own independent recollection, of what had happened at the February board meeting. He was not "a person presumably able to put the true complexion on the facts relied on [by ASIC] as the ground" for any inference that ASIC sought to have drawn from the evidence. And contrary to the conclusions reached by the Court of Appeal, what Lord Mansfield said in Blatch v Archer permitted no larger conclusion. ASIC had tendered admissible evidence that, if accepted, showed that the draft ASX announcement had been tabled and approved. This was not a case where ASIC's case depended on inference, let alone on "uncertain inferences", or where there was a question about whether "limited material is an appropriate basis on which to reach a reasonable decision". It was not a case where "the missing witness would be expected to be called by one party rather than the other" or where it was known that "his evidence would elucidate a particular matter" (emphasis added). It was the respondents' case that depended upon inference. The inference they sought to have drawn was that the minutes were not accurate evidence of tabling and approval. ASIC not calling Mr Robb did not entail that the inference which the respondents sought to have drawn could be more safely drawn. It was not to be expected that ASIC would call him (and the respondents not call him) if Mr Robb would have given any evidence that cast doubt on whether the minutes were accurate. 364 We do not think the positions of Dr Sherman and Mr Robb are analogous. Here, Dr Sherman's evidence would have undoubtedly elucidated the central issue in this case: what would Dr Sherman have done if the injunction had been refused? 365 Turning then to the fourth contention i.e. the question of which party should have called Dr Sherman, the Commonwealth developed two contentions. The first was that Dr Sherman was equally available to both the Commonwealth and Sanofi. The second was that Dr Sherman was not in the Commonwealth's camp in light of Apotex's extensive claims for privilege. 366 The first contention is a false issue. The question is not whether both parties stood in the same position when it came to the practicalities of calling Dr Sherman. It is whether he was more in the camp of one party rather than the other. His Honour considered and, in our view, correctly rejected the Commonwealth's contention that Sanofi ought to have called Dr Sherman. The trial judge did this on the basis that Sanofi and Apotex were trade rivals engaged in litigation across three jurisdictions: J [344]. To tease that out, Sanofi ran the real risk if it called Dr Sherman that he would seek to injure it by giving unhelpful evidence. Dr Sherman was not therefore in Sanofi's camp. 367 As to the second contention that Dr Sherman was not in the Commonwealth's camp, the question of what can be inferred from Apotex's extensive claims for privilege requires a broader conspectus of the evidence than the Commonwealth's submissions allow. As we have explained above (at [361]), we do not accept the Commonwealth's submission in relation to privilege. Further to this, it might, in principle, be possible to infer from the claims made for privilege that Apotex was unwilling to co-operate with the Commonwealth. The problem with that inference, however, is that it cannot be made to co-exist with the fact that the Commonwealth did, in fact, call several witnesses from Apotex including two from overseas. More precisely, the Commonwealth's submission must confront the fact that it had called Mr Fahner who was an employee of Apotex usually resident in Canada. The trial judge was alive to this and to the fact that the Commonwealth had also called two other witnesses from Apotex, Mr Millichamp (from Australia) and Mr Goel (from India): J [342]. 368 We do not, therefore, see how an inference can be drawn that the claims for privilege justified the drawing of an inference that Apotex was not co-operating with the Commonwealth for it is clear that it was. The only way around this conundrum is to redraw the inference so that instead of suggesting that the claims for privilege implied that Apotex was not co-operating with the Commonwealth it would instead be suggested that the claims for privilege implied only that Apotex was not co-operating to the extent of making Dr Sherman available. It is true that this would overcome the difficulty for the argument that the other Apotex witnesses constituted, but it comes at the price of being entirely implausible. There is no basis upon which the claims for privilege may be used to draw different inferences about the degree of Apotex's co-operation depending on who the relevant witness was. 369 In our view, the fact that the Commonwealth called three witnesses from Apotex, including one from Canada and another from India, was evidence from which it could be inferred that Apotex was willing to cooperate with the Commonwealth and Apotex's claims for privilege in no way impacted upon this. 370 Leaving aside the question of Apotex's claims for privilege, the Commonwealth then submitted that an inference of co-operation could not be drawn about Dr Sherman himself. The reason for this was that it was 'obvious' that Dr Sherman would not have been willing to give evidence: AS [87]. In this context, the implications of the word 'obvious' should be explicitly stated. The implication is that no evidence was necessary in order to establish that Dr Sherman was not willing to give evidence. What were the facts, matters and circumstances advanced to establish the necessity for this implication? 371 They were: first, that he would have had no interest in exposing himself to cross-examination by his trade rival, Sanofi; secondly, Apotex had not called Dr Sherman in its own case against Sanofi on the undertaking before reaching settlement which gave rise to an inference that he was unwilling to give evidence; thirdly, as Mr Gleeson observed in his address at T102, the Commonwealth did not have access to any affidavit evidence prepared by Dr Sherman from the Apotex proceedings; and fourthly, because the trial judge had rejected questions asked by Sanofi's counsel to Apotex witnesses directed to establishing that Dr Sherman was available to give evidence. 372 We would accept that it would be open to infer from the first three of these matters that Dr Sherman would be unwilling to give evidence. However, the question is whether that inference should be drawn. The Commonwealth's submission that the inference should be drawn must dwell in the long shadow cast by the fact that at trial it led no evidence that Dr Sherman had declined to give evidence on its behalf. As his Honour observed at J [346], there 'was no evidence called by the Commonwealth that would provide any explanation as to why it did not call Dr Sherman or what attempts, if any, it made to call him'. The absence of any evidence of that kind powerfully now tells against drawing any inference that it was obvious that Dr Sherman would not agree to give evidence. We decline to draw such an inference. 373 Once that is accepted, the reasoning of the trial judge cannot be faulted. His Honour did not err in declining to draw an inference that Dr Sherman would have instructed Mr Millichamp to procure the PBS listing of Apotex's clopidogrel products from 1 April 2008. 374 For completeness, we should note the Commonwealth's submission under this heading that the approach of Mason J in Air Express necessitated that it was Sanofi which ought to have called Dr Sherman. Here the argument was that the onus lay on the defendant to a claim under an undertaking as to damages to demonstrate the existence of countervailing circumstances why compensation ought not to be ordered. It is doubtful this submission relates to Jones v Dunkel at all. The submission seems to be an effort to put on Sanofi a particular legal burden to have called Dr Sherman. Assuming the submission is correct, it would, however, render the Jones v Dunkel question irrelevant. There would be no issue as to the significance for the Commonwealth of its failure to call Dr Sherman. The effect of the Air Express submission would be that Sanofi was required to call Dr Sherman and having not done so, was therefore bound to fail. As such, we doubt that this submission meaningfully relates to Ground 2(c) which takes as its point of departure that Jones v Dunkel is an applicable part of the conceptual framework. Even if the Commonwealth's submission can be meaningfully integrated into such a framework, Mason J was in dissent in Air Express and we see nothing in the other reasons for judgment which suggests that Sanofi was bound to call Dr Sherman.

GROUND 2(D): DELAY 375 Ground 2(d) is as follows: (d) The extreme delay by the primary judge in delivering judgment allows an appellate court more readily to conclude that arguments and evidence not addressed were overlooked, that the advantages otherwise usually held by a trial judge were lost, and that the above errors can be more readily established. 376 There were 31 months between the conclusion of the case and the delivery of the judgment by the trial judge. The Commonwealth made two submissions. First, whilst ordinarily the fact that a trial judge had not dealt with every item of evidence might not lead to an inference that that evidence had been overlooked, this was not so where such a significant delay was involved: AS [92]. Secondly, the trial judge had rejected certain aspects of Mr Millichamp's evidence and found him to be in some ways an unsatisfactory witness. Because of the delay, the deference shown by appellate courts to credit findings was therefore lost: AS [93]. 377 We would accept, in principle, the point made by the Commonwealth. As this Court said in Expectation Pty Ltd v PRD Realty Pty Ltd [2004] FCAFC 189; 140 FCR 17 at [68]-[73] ('PRD Realty'): Where there are relevant contemporaneous materials, such as file notes and correspondence, and there is significant delay between the hearing of evidence and the giving of reasons for conclusions, being reasons that do not advert to the contemporaneous materials and do not give specific reasoning for accepting or rejecting the evidence of particular witnesses, the conclusions reached should be given careful scrutiny and consideration by an appellate court where the findings are challenged on appeal. Delay between the taking of evidence and the making of a decision is not, of itself, a ground of appeal, unless the judge could no longer produce a proper judgment or the parties are unable to obtain from the decision the benefit which they should (cf Boodhoo v Attorney-General of Trinidad and Tobago [2004] 1 WLR 1689 at [11]-[12]). Nor does such delay of itself indicate that a trial has miscarried or that a verdict is in any manner unsafe. However, where there is significant delay in giving judgment, it is incumbent upon an appellate court to look with special care at any finding of fact challenged on appeal. In ordinary circumstances, where there is a conflict of evidence, the trial judge who has seen and heard the witnesses, has an advantage. That advantage includes seeing the oral and documentary evidence unfold in a coherent manner, which cannot be replicated on appeal (State Rail Authority (NSW) v Earthline Constructions Pty Ltd (in liq) (1999) 73 ALJR 306 per Kirby J at [90]; Bartlem Pty Ltd v Cox Industries (Australia) Pty Ltd (2002) 55 IPR 449 at [87]). That advantage will ordinarily prove decisive on appeal unless it can be shown that the trial judge failed to use or misused such an advantage. The mere fact of a long delay itself weakens a trial judge's advantage. Thus, delay must be taken into account when reviewing findings made by a trial judge after a significant delay from the time when the relevant evidence was given. In the normal course, statements made by a trial judge of a general assertive character can be accepted as encompassing a detailed consideration of the evidence. However, where there is significant delay, such statements should be treated with some reserve. After a significant delay, a more comprehensive statement of the relevant evidence than would normally be required should be provided by the trial judge in order to make manifest, to the parties and the public, that the delay has not affected the decision. In cases not affected by delay, an appellate court is entitled to assume that the mere failure to refer to evidence does not mean that it has been overlooked or that other forms of error have occurred. However, where there is significant delay, no favourable assumptions can be made. In such circumstances, it is up to the trial judge to put beyond question any suggestion that he or she has lost an understanding of the issues. Where there is significant delay, it is incumbent upon a trial judge to inform the parties of the reasons why the evidence of a particular witness has been rejected. It is necessary for the trial judge to say why he or she prefers the evidence of one witness over the evidence of other witnesses (Hadid v Redpath (2001) 35 MVR 152 at [34] and [53]). Of course, where the trial judge, notwithstanding significant delay, demonstrates by his or her reasons that full consideration has been given to all of the evidence, the parties and the public may be satisfied that the delay has not affected the decision. More specifically, if the reasons demonstrate that the delay has not weakened the trial judge's advantage, confidence will be maintained in the decision. For example, it would be open to a trial judge to explain in the course of giving reasons that contemporaneous notes were made of impressions formed as evidence was given by witnesses of importance (see R v Maxwell (unreported, Court of Criminal Appeal, NSW, Spigelman CJ, Sperling and Hidden JJ, No 60282 of 1998, 23 December 1998)). 378 In the course of dealing with Ground 2(b), we have engaged with each of the matters which were said to have been overlooked by the trial judge. In every case, we have concluded that the evidence was either not material or was in fact examined by the trial judge in his reasons. Thus the first limb of the Commonwealth's argument does not result in the demonstration of error. Even assuming that the trial judge's failure to advert to the evidence was the result of the delay, the evidence was not material to the outcome. 379 That reasoning does not, however, apply to the second limb which relates to his Honour's assessment of the evidence of Mr Millichamp. One begins by reminding oneself about the metes and bounds of the context with which the trial judge was confronted. Although the earlier affidavits prepared by Mr Millichamp had suggested that he was the key decision maker in relation to the question of whether Apotex would launch at risk if the injunction was not granted, by the time of the trial, the role of Dr Sherman had been dislodged from the undergrowth. Sanofi launched an attack on the credit of Mr Millichamp on the basis that his earlier evidence had been false (because of his failure to disclose the role of Dr Sherman), but the trial judge did not accept this. Instead, his Honour was inclined to read Mr Millichamp's evidence in light of the context in which it had been given. That context had been Apotex's efforts to resist the interlocutory injunction and the trial judge reasoned, correctly in our opinion, that not too much was to be read into Mr Millichamp's omission of any mention of Dr Sherman: J [332]. 380 By the time of the trial, however, it had become apparent that the critical decision maker was Dr Sherman. This fact was recognised on both sides although the forensic postures then adopted differed. The Commonwealth's case became that Dr Sherman had decided in February 2007 that Apotex would launch at risk. Once that decision was made, no further discretion remained in Mr Millichamp to decide otherwise. The die had been cast and all that awaited was for Mr Millichamp to find out whether Gyles J refused or granted the injunction. If the injunction were refused, then Mr Millichamp already had his instructions and they were that he should proceed to launch at risk. 381 One difficulty with this case was the absence of Dr Sherman to give any evidence as to what he would have done. The Commonwealth's answer to this problem was that it could prove that Dr Sherman had already made that decision on 20 February 2007. The evidence for this consisted of his email of that date to which reference had already been made so that no further reference to Dr Sherman was necessary. If this defensive line could be maintained, it entailed that no inquiry into what Dr Sherman would have done was necessary. On this view, Mr Millichamp had his instructions; all that remained was for him to act upon them. 382 A further problem with that case was the large amount of evidence which was only consistent with Dr Sherman having a continuing role in the decision making process. These included the emails passing between Mr Kay and Dr Sherman on 25 June 2007 and 27 June 2007 and the cascade of emails between Dr Sherman, Mr Kay, Mr Millichamp and others on 26 and 27 June 2007. 383 It also included Mr Millichamp's email of 27 June 2007 in which he referred to the fact that any launch at risk was subject to Dr Sherman's 'further advice/approval'. Mr Millichamp was cross-examined about this email. He endeavoured to say that he thought it was unlikely that Dr Sherman would have changed his mind. We will return to this cross-examination in a moment, but it will be at once apparent that this answer involved a breach of the defensive line which the Commonwealth had sought to erect. Whether Mr Millichamp thought it likely that Dr Sherman would or would not change his mind did not detract from the fact that his email showed that it was not the case that Mr Millichamp was mechanically to carry into effect Dr Sherman's instructions contained in his email of February 2007. Mr Millichamp's own email showed that Dr Sherman, as might naturally be expected of the CEO and Chairman of the entire Apotex group, continued to be the decision maker. 384 Turning then to the cross-examination itself, this was set out in the reasons of the primary judge at J [249] in these terms: Mr Millichamp was cross-examined by Mr Sheahan QC on the email to Mr Haas as follows: MR SHEAHAN: Now, what you were indicating to Mr Hass, I suggest, was that the final decision to launch, even if you were successful in avoiding an injunction, would be a decision made by Dr Sherman at that time; correct? --- And I think the thinking was that Dr Sherman gave the initial instructions. Andrew Kay had then clarified things with Dr Sherman. Again, we need to show him respect. If there was anything we needed to go to him with, we would. But, if there was no change on the instructions he had given, we would just move ahead. So do you say to his Honour that what - the only possibility you're addressing here is that after having informed Canada of your success on the interlocutory injunction, then, out of the blue, as it were, they might indicate to you that they had had a change of heart? --- Very unlikely. … MR SHEAHAN: You were saying that it's very unlikely that out of the blue Canada would - having been told of your success - just send you a message saying, "We've changed our mind." You agree? --- I agree with that. … And because that scenario was unlikely, what I want to suggest to you is that that is not the scenario you were addressing in this email by saying: We plan to launch subject to Barry's further advice/approval You were addressing another possibility, weren't you? --- It's - in - in retrospect I - or in - in hindsight or in - I can't actually remember what scenario I was discussing here. It's to Steven Hass [sic]. I guess I'm just covering all angles and just saying look, subject to Barry's further advice/approval - it's a - it's almost a throwaway line. But you know, there could be a scenario, I guess, where Barry could change his mind. I mean, I ..... think he would. But - it would be very unlikely. What you were addressing here, I suggest, was your intention to confirm the instructions that you had after a successful interlocutory proceeding, to make sure that Canada still wanted to proceed at risk in Australia? --- It's an email to the - I already had the instructions on what to do and had already explained to my boss - this is another email to somebody in production, and I've just put that comment in there. I never at any stage would - I think it would be very, very unlikely indeed that Barry would ever change his mind. And it's only to Steven Hass [sic], who actually is a personal friend of mine. And I put that - put that line in there - I didn't think that I would have to go through another round of approval with - with - with Barry. 385 It will be seen that the last answer suggested, contrary to the terms of his own email, that he did not think that he would need further to consult with Dr Sherman. It was this aspect of Mr Millichamp's evidence that the trial judge found to be unsatisfactory (having acquitted him of the larger charge of giving deliberately false evidence by failing in his earlier affidavits to disclose the role of Dr Sherman). His Honour dealt with this at J [250]-[251]: I found this evidence most unconvincing. In my view it was largely non-responsive and evasive. Mr Millichamp was not asked whether he believed that Dr Sherman was unlikely to change his mind, but he was at pains to emphasise how unlikely it was that Dr Sherman would do so ("very, very unlikely indeed"). The suggestion that Mr Haas was his personal friend, or that he was only "Stephen Haas" or "somebody in production" does not explain why the launch was said by Mr Millichamp to be subject to Dr Sherman's "further advice / approval". I do not accept the explanation given by Mr Millichamp concerning his email to Mr Haas. In particular I do not accept that the relevant statement was a "throw away" line or that Mr Millichamp was merely seeking to "cover all angles". Nor do I accept that it was unlikely that Dr Sherman would change his mind. On Mr Millichamp's own evidence, Dr Sherman had already changed his mind in that the revocation/launch strategy for Australia was originally predicated on a successful outcome of the litigation in the US and Canada. In my view, the terms of Mr Millichamp's email are more likely than not to be correct in so far as they suggest that Dr Sherman had not finally committed to a launch at risk in Australia and that it would be necessary for Mr Millichamp to obtain final approval to launch at risk in the event that no interlocutory injunction was granted. 386 It will be apparent that the trial judge's criticisms were not related to the demeanour of Mr Millichamp. Rather, they were directed at the content of what he had said. The trial judge examined that content and found it wanting. This is not a case, therefore, where a trial judge has said that he did not accept a witness's evidence because he found it unconvincing. Rather, it is a case where the trial judge has explained precisely why he found the response given orally to be unsatisfactory and why he preferred the contemporaneous written record. The basis for the credit finding is therefore entirely before this Court. 387 The delay in the production of the trial judge's reasons required, as this Court held in PRD Realty at [72], that the trial judge inform the parties the reasons why the evidence of Mr Millichamp was rejected. As the Court observed at [73], where the reasons of the trial judge demonstrate that, notwithstanding the delay, full consideration has been given to all of the evidence, the public may be satisfied that the delay has not affected the decision. In our view, this is such a case. The trial judge's reasons for finding this aspect of the evidence of Mr Millichamp unsatisfactory demonstrate, clearly in our view, that the trial judge was very much alive to the detail of the evidence and its significance. Indeed, if we may say with respect, the trial judge's reasons as a whole represent a most thorough and searching excavation of the very complicated factual questions which the case generated. 388 In that circumstance, we are not satisfied that the delay provides any reason for reviewing his Honour's conclusion that this aspect of Mr Millichamp's evidence was unsatisfactory. Ground 2(d) should be rejected.

GROUND 2(E): FINDINGS THE TRIAL JUDGE OUGHT TO HAVE MADE 389 This ground does not arise where we have not detected any error in the trial judge's approach to the issues otherwise raised in Ground 2.

CONCLUSION ON GROUND 2 390 In light of the above, it follows that Ground 2 must be rejected.

CONCLUSION 391 As have explained above, once Ground 2 is rejected the appeal must be dismissed. Whilst the Commonwealth would have succeeded on Ground 1 had it succeeded on Ground 2, we do not think, in the circumstance, that there should be a departure from the ordinary position that the unsuccessful party should pay the costs of the successful party. 392 We would make the following orders: (1) The appeal be dismissed. (2) The Appellant pay the costs of the Respondent to the appeal as taxed, assessed or agreed. I certify that the preceding three hundred and ninety-two (392) numbered paragraphs are a true copy of the Reasons for Judgment of the Honourable Justices Besanko, Perram and Yates.