England and Wales High Court (Chancery Division) Decisions

IN THE HIGH COURT OF JUSTICE
CHANCERY DIVISION

		Royal Courts of Justice Strand, London, WC2A 2LL
		11/06/2009

	Para
INTRODUCTION	1 - 24
THE WITNESSES
Witnesses of Fact
The Banks' Witnesses	25 - 45
Ferrero's Witnesses	46 - 64
Mr Abidali's Witnesses	65 - 75
Expert Witnesses
The Banking Experts	76 - 81
The Hazelnut Market Experts	82 - 96
The Turkish Law Experts	97 - 105
THE FACTS
Introduction	106
Background - The Period Ending in 2000
The Turkish Hazelnut Market	107 - 125
Baskan Gida	126 - 132
Ferrero	133 - 138
Ferrero Industrial's Hazelnut Purchasing Strategy	139 - 144
Ferrero's Trading Relationship with Baskan Gida Prior to 2001	145 - 184
Indo-Med and its Relationship with Baskan Gida Prior to 2001	185 - 189
The 2000 Hazelnut Crop	190 - 191
2001
January to July	192 - 194
The Turkish Economic Crisis and its Effect on Baskan Gida	195 - 201
The FKB Contracts	202 - 226
The Dahod Agreement	227 - 242
Baskan Gida Approaches the Banks	243 - 249
July 2001 - The Banks' Meeting with Mr Casale, and the Ferrero July Trade Reference
The July meetings	250 - 261
The July Letter	262 - 272
August to December 2001	273 - 275
The Dahod Investment	276 - 284
The Banks' Continuing Due diligence	285 - 291
The 2001 Crop - August to Mid-October 2001	292 - 320
The Framework Contract	321 - 338
The Baskan Embark upon their Fraud	339 - 351
The 45 Day Lie	352 - 371
Ferrero/Baskan Gida trading - Mid-October to 13th December 2001	372 - 393
The Re-Grade Contracts	394 - 403
Negotiating the Acknowledgment of Assignment - Round One	404 - 431
The Early December Contracts	432 - 456
The December Letter	457 - 489
Negotiating the Acknowledgment of Assignment - Round Two	490 - 494
Completion of the Facility	495 - 509
Notifying Ferrero of the Assignments	510 - 518
The Re-numbered December Contracts	519 - 530
2002
January - Dealings between Baskan Gida and Ferrero
The 3rd January Bolowich Email	531 - 536
Contracts 5088 and 5089	537 - 545
January and February - The Implementation of the Fraud	546 - 559
Introducing the Scheme to Mr Abidali	560 - 585
Selling the Scheme to Ferrero	586 - 627
Ferrero Decide to Cooperate	628 - 630
The 28th January Contract Renewals	631 - 641
The Early February Contracts	642 - 667
The Split Invoices - Round 1	668 - 680
The 13th February 2002 Framework Contract Amendment and Aksu/ Baskan Gida Takeover	681 - 699
The Split Invoices - Round 2	700 - 708
The Position of the Parties in Late February 2002	709 - 748
The Aftermath
The Banks Discover the Baskan ' Fraud	749 - 750
March 2002 - Mr Byles speaks to Mr Casale	751 - 753
Investigation by the Banks	754 - 760
Ferrero's reaction	761 - 767
Contacts between Ferrero and the Baskans	768 - 773
The Freezing injunction on Aksu Gida	774
Late April - The Banks enquire further of Ferrero	775 - 782
Baskan Yuksel	783 - 787
Internal enquiry within Ferrero	788 - 795
Litigation Begins	796 - 798
Ferrero's Conduct from March 2002	799 - 808
Dealings between Mr Abidali, Mr Dahod and the Baskans from March 2002 onwards	809 - 823
LAW AND ANALYSIS
Conspiracy to injure by unlawful means	824 - 848
The Conspiracy Claim against Ferrero	849 - 927
The Conspiracy Claim against Mr Abidali	928 - 954
The Banks' Proprietary Claim against Ferrero	955 - 957
Conversion - Turkish Law	958 - 988
Knowing receipt	989 - 992
The underlying factual question	993 - 997
Misrepresentation - Deceit and Negligence	998 - 1012
The Trade References
The July Letter	1013 - 1015
The December Letter	1016
The Framework Contract
The Framework Contract as signed on 15th October 2001	1017 - 1021
The Framework Contract as amended on 13th February 2002	1022
The Ferrero Industrial Acknowledgements of Assignment	1023 - 1028
The Countersigned Contracts	1029 - 1038
Contracts 5094 -7	1039 - 1044
Contracts 5088 and 5089	1045 - 1050
Contracts 5100 - 1	1051 - 1053
Contract 5104	1054 - 1055
The Negligent Representation as to Timely Delivery in the December Letter
Reliance	1056 - 1064
The BBL Principle	1065 - 1068
Causation - Novus Actus Interveniens	1069
Remoteness	1070
Contributory Negligence
Audited Accounts	1071 - 1090
The 30%point	1091 - 1103
Failure to monitor deliveries and payments	1104 - 1109
Failure to warn Ferrero of the Banks' Rights	1110 - 1111
Contributory negligence conclusion	1112 - 1114
CONCLUSIONS	1115

Mr Justice Briggs:

INTRODUCTION

1. At the end of December 2001 Baskan Gida Sanayi Ve Pazarlama A.S. ("Baskan Gida") was, and had for many years been, the largest single exporter of hazelnuts in Turkey and, indeed, in the world. On 14^th December 2001 Baskan Gida made a loan agreement ("the Facility") with the Bank of Tokyo-Mitsubishi UFJ Ltd and KBC Bank N.V. (respectively "BTM" and "KBC" and collectively "the Banks"), the stated purpose of which was to enable Baskan Gida to borrow, on a rolling basis, up to US$35 million in order to finance the purchase and processing of hazelnuts prior to their export, it being envisaged that repayment would be achieved by the assignment by Baskan Gida to the Banks of its rights to receive the proceeds of the sale of its hazelnuts to its largest foreign customer, the Ferrero group of companies (collectively "Ferrero").
2. By the end of February 2002 Baskan Gida had drawn down €22,821,566 pursuant to the Facility, and repaid through Ferrero only €1,402,022.68. During January and February 2002, unknown to the Banks at the time, Baskan Gida transferred substantially the whole of its business, assets, trading premises, hazelnut stocks and employees to Aksu Gida Sanayi Ve Ticaret Limited ("Aksu Gida"). Baskan Gida ceased trading at the end of February and thereupon defaulted on its repayment obligations to the Banks. In May 2002 there was a further purported transfer of the hazelnut business and assets to Baskan Yuksel.
3. The Banks therefore suffered a net shortfall of € 21,419,543.32, against which they received a payment of $75,000 from Baskan Gida in June 2002, and a further $720,000 from the sale of pledged hazelnuts later that year.
4. The Banks claim, and this is not seriously in dispute between the active participants in this litigation, that from start to finish they were the victims of a premeditated fraud in which the prime movers were those members of the Baskan family who owned Baskan Gida and who (although this is in dispute) were also the beneficial owners of Aksu Gida and Baskan Yuksel. The Banks' case is that the Baskans' real purpose in negotiating the Facility was to obtain by deception the funds necessary to rescue their potentially profitable hazelnut export business from the state of hopeless insolvency into which Baskan Gida had fallen, using the Banks' money for the purchase and transfer to Aksu Gida (and later Baskan Yuksel) of the hazelnut stock necessary to continue that business into the future, leaving the Banks with a worthless recourse against Baskan Gida.
5. With the benefit of hindsight and detailed subsequent research, the Banks say that the Baskans negotiated the structure of the Facility in such a way as to enable them to obtain the Banks' money, use it for the purchase of hazelnuts and their transfer to Aksu Gida, together with the ongoing hazelnut business and associated assets, within a time-frame which would enable the fraud to be completed before any default in re-payment to the Banks could put them on notice of what had been planned and implemented. For that purpose the Baskans stated falsely to the Banks during the negotiation of the Facility that Baskan Gida's terms of trading with Ferrero gave rise to a 45 day period between Baskan Gida's requirement to pay its suppliers and the receipt of payment from Ferrero for exported hazelnuts, a falsehood which has come to be known in the proceedings as "the 45 Day Lie".
6. Although the Banks anticipated incurring a risk of some loss in the event that Baskan Gida ceased trading during the period of the revolving Facility, they claim that the extent of their loss was primarily attributable to two specific frauds in the manner in which Baskan Gida operated the Facility. The first consisted of the presentation to the Banks of false invoices suggesting that Ferrero was instructed to make payments in respect of hazelnut purchases to the Banks' collection agent, while at the same time instructing Ferrero to make those payments elsewhere in the invoices actually delivered. The second consisted of the obtaining of hazelnuts purchased with the Banks' money and pledged by way of security by false statements that they were to be used to satisfy Ferrero's export orders, whereas in truth they were then transferred to Aksu Gida. More generally the Banks say that they were the victims of fraudulent trading, since the Baskans did not intend when drawing down money under the facility, that it would ever be repaid.
7. Although the Baskans succeeded in defrauding the Banks, they were largely denied the benefit of that fraud, primarily because of the prompt intervention of one of their Turkish banks, Yapi Kredi, into the affairs of Aksu Gida. Despite an attempt at a further transfer of the hazelnut business from Aksu Gida to Baskan Yuksel later in 2002, no long-term rescue of the family business was thereby achieved. Although Baskan Gida, Aksu Gida and four members of the Baskan family are joined as the first, sixth and second to fifth defendants respectively, and have all formally denied the claim, they have taken no active part in these proceedings. Even if a judgment in these proceedings could be enforced against them in Turkey, it is unlikely that any of them are worth powder and shot.
8. The Banks have however sought to recover their losses from two groups of defendants. The first ("the Abidali Defendants") consists of Mr Shabbir Abidali, his company Indo Mediterranean Commodities Limited ("Indo-Med") and two entities in which, or in the assets of which, the Banks allege that Mr Abidali has a beneficial interest, namely Alanvar Establishment and Ridgebeach Limited. Indo-Med went into voluntary liquidation in March 2006, and its only participation since then has been the provision of documents and a waiver of legal professional privilege, by its liquidator on its behalf.
9. The Banks claim that Mr Abidali and Indo-Med played a central part in both the planning and implementation of the fraud. They are therefore sued as participants in an alleged unlawful means conspiracy. Mr Abidali denies this. He accepts that, at his direction, Indo-Med played a central role in the transactions between Baskan Gida and Aksu Gida in early 2002, but claims that Indo-Med thereby did nothing unlawful, and did not knowingly or intentionally harm the interests of the Banks.
10. The second group of defendants are four Ferrero companies. They are Ferrero Industrial Services G.E.I.E. ("Ferrero Industrial"), Ferrero S.p.A. ("Ferrero Italy"), Ferrero OHGMBH ("Ferrero Germany") and Ferrero France S.A. ("Ferrero France"). I shall refer to them collectively as "the Ferrero Defendants". The Ferrero group included companies incorporated in other countries, but Ferrero Italy, Germany and France together made the overwhelming bulk of hazelnut purchases from Baskan Gida at the material time. I shall refer to them as "the Ferrero Operating Defendants". Ferrero Industrial did not itself purchase hazelnuts, but was formed to manage, on a collective basis, all the raw material purchases required by the operating companies in the Ferrero group, including the Ferrero Operating Defendants.
11. The Banks put their claim against the Ferrero Defendants on a number of alternative grounds. Originally, they claimed that by reason of assignments by Baskan Gida of its rights to the payment by Ferrero of the price for the exported hazelnuts, duly notified to the Ferrero Operating Defendants, those defendants are liable in contract, notwithstanding having paid Baskan Gida in full for the specific purchases. Before these proceedings were issued, that claim was made the subject of negative declaration proceedings in Italy, with the result that the contractual claim in these proceedings was stayed by Lawrence Collins J. pursuant to the Judgments Regulation. That claim remains pending in Italy, and its determination is unlikely to occur in the near future.
12. Secondly, the Banks claim that the Ferrero Defendants are liable in unlawful means conspiracy for having participated with knowledge in Baskan Gida's fraud, in particular through the activities and knowledge of three Ferrero employees, Mr Casale, Mr Rosa Brunet and Mr Bolowich. As against the Ferrero Defendants, the claim is that they participated in the implementation rather than in the design of the fraud.
13. Next, the Banks claim against the Ferrero Defendants in deceit, upon the basis that they were induced to lend to Baskan Gida by dishonest misrepresentations made, principally by Mr Casale, on their behalf. Alternatively the Banks claim that the same misrepresentations were made negligently.
14. Finally, the Banks claim against the Ferrero Operating Defendants an account upon the basis of their receipt of hazelnuts subject to a continuing security interest in favour of the Banks, or because of receipt of hazelnuts purchased with the proceeds of the sale of hazelnuts subject to that security interest. I will refer to this final type of claim as the Banks' Proprietary Claim.
15. The Ferrero Defendants deny all these claims. As to the conspiracy claim, they admit the Baskans' fraud on the Banks, but deny either knowledge, participation or intent to harm the Banks. As to the claim in deceit, they deny both misrepresentation and dishonesty. As to the claim in negligence they deny negligence, and reliance by the Banks, assert contributory negligence on the part of the Banks, and advance both "scope of duty" and remoteness defences. In relation to both the deceit and negligence claims Ferrero deny causation by advancing a novus actus defence. As to the Banks' Proprietary Claims they say that the Banks cannot prove that Ferrero received any hazelnuts subject to any continuing security interest of the Banks, let alone any interest of which they had knowledge when making payments for them to Aksu Gida and Baskan Yuksel.
16. The Banks do not claim that they can prove the fraud and deceit which they allege against the Ferrero Defendants by reference to 'smoking gun' direct evidence, although to some extent they do as against Mr Abidali. Rather they say that, when the whole of the facts about the defendants' participation in the relevant events is examined in detail, their conduct is explicable only in terms of an inference of dishonesty. For this purpose, and for the active defendants' purpose of demonstrating the exact opposite, no stone has been left unturned, both in preparation and during the conduct of the trial. Every aspect of the parties' dealings during two hazelnut crop seasons has been examined in minute detail. Lengthy evidence from hazelnut market experts has been deployed and tested in relation to the central allegation that the Ferrero Defendants' purchases from Baskan Gida were not on normal arms-length market terms. Enormous efforts were devoted by forensic accountants into establishing (happily in the end by agreement) relevant statistics regarding Ferrero's stock levels and hazelnut requirements, in relation to an allegation that Ferrero dishonestly assisted Baskan Gida by ordering large quantities of hazelnuts which were not needed. The entirety of Ferrero's hazelnut purchasing programmes from all sources over two years has been precisely recreated for the purpose of testing allegations of unusual, and therefore suspicious, trends during the period of the fraud.
17. In addition, the Ferrero defendants' allegations of contributory negligence by the Banks in the negotiation and operation of the Facility have led to the most detailed examination of the minutiae of the Banks' conduct during the relevant period, by reference to comprehensive surviving documentation, many days of cross-examination of the Banks' employees, and the deployment and testing of the evidence of banking experts.
18. The overall result has been a trial of spectacularly large proportions, even by reference to the substantial sums at stake. The trial bundles run to more than 190 lever arch files. There have been 56 days of oral evidence, 30 days of submissions, and 35 days for judicial reading and preparation of this judgment.
19. In the end, as is typical of relatively well documented commercial transactions mainly between sophisticated entities, the primary facts are, though very detailed, not the subject of large dispute. What the parties did, wrote and said to each other is either established by surviving documents and notes of meetings, or largely forgotten beyond practical recall during the six years since the relevant events occurred. The critical factual disputes concern the matters which they knew, suspected, or to which it is alleged that they deliberately turned a blind eye. In addition, (as often in a case where fraud is alleged), there is the major question of motive.
20. In the absence of admissions, and there are none, these issues turn on inferences properly to be derived from the primary facts viewed in their entirety, paying due regard to the demeanour and apparent credibility of the witnesses, but remembering "always to test their veracity by reference to the objective facts proved independently of their testimony, in particular by reference to the documents in the case, and also to pay particular regard to their motives and to the overall probabilities." (Armagas Ltd v Mundogas SA [1985] 1 Lloyd's Rep. 1 at 57, per Robert Goff LJ).
21. As for motive, both groups of defendants put this question in the forefront of their respective arguments. For the Ferrero Defendants it is said first, that nothing has emerged to displace the prima facie improbability that a large, respectable group of soundly based companies should wish to participate in a fraud on their Turkish supplier's bankers, and secondly, that despite exhaustive enquiry there is not a shred of evidence that the individual employees alleged to have been dishonest had any personal motive (such as corrupt inducement from the Baskans) to involve their employer in any such impropriety.
22. For Mr Abidali it is submitted that while Indo-Med's participation in the transactions complained of is readily explicable as a legitimate means of assisting its investing partners in obtaining repayment of substantial loans from Baskan Gida, no motive for participation in a dishonest conspiracy to injure the Banks has been identified.
23. In the present case my conclusions on all those difficult issues have been informed by my ever-increasing immersion in the detailed commercial realities of the international trading and financing relationships between the parties, gained during the course of a very long trial, at the invitation of all them. Although mindful of the advantages of economy in the expression of reasons for my conclusions, I have found it impossible to provide a full and fair expression of them without reference to a detailed and mainly chronological account of the facts, and an explanation of the relevant commercial background.
24. Before doing so, I must state my conclusions about the general demeanour and credibility of the witnesses, which I do by reference to the order in which they were called.

THE WITNESSES

The Banks' Witnesses

25. The Banks first called Mr Philip Lewis, now the Head of Credit and Risk Management within the London Branch of KBC. In 2001-2 he was a member of KBC's Marketing Support Group ("the MSG") in London, which was itself part of KBC's Global Trade Finance Group ("GTFG"), also based at its London branch.
26. The role of the staff of the MSG was to provide support and assistance to relationship managers within the GTFG. KBC's relationship with Baskan Gida was managed by Mr Kris van Broekhoven, and Mr Lewis was, in effect, his right hand man during the negotiation and operation of the Facility. Although he was in that sense subordinate to Mr van Broekhoven, he remained with KBC after Mr van Broekhoven's departure in May 2002, with the result that he was deployed by KBC as its main factual witness and, for the same reason, endured the longest cross-examination.
27. Mr Lewis made an immediate initial impression as a helpful, straightforward and relaxed witness, whose ability to assist the court with factual evidence derived from a good memory was fortified by his own realistic appreciation of its inevitable limitations, and a readiness to explain and distinguish between the relatively few occasions when he could recall relevant events, and the more frequent occasions when he was obliged to reconstruct from subsequent work on surviving documents. With the assistance of the Banks' legal team he had evidently taken considerable care to piece together his detailed account of events in his long witness statement from contemporaneous documents. That care was if anything exemplified by his considerable disappointment at having his reconstruction revealed to be erroneous on one significant point (namely that he had attended a particular meeting in March 2002, when in fact documents which he had missed showed that he had not), and his dismay at being asked in the middle of his cross-examination to check again whether he had made any other errors.
28. He was however a witness with a mission, namely to educate the court and in particular his cross-examiner Mr Strauss QC in what he regarded as the fundamental differences between structured trade finance as a niche form of bank lending and what he described as traditional or balance sheet based bank lending, and thereby to emphasise what he regarded as the large differences in the principles properly to be applied to each. At the heart of his perception of the differences between the two types of lending lay his frequently repeated assertion that analysis of the creditworthiness of the intending borrower plays a much less significant part of the necessary due diligence to be applied to a structured trade finance proposal, than it does in relation to a proposed traditional loan.
29. Since a central thrust of the Ferrero Defendants' case on contributory negligence was that the Banks failed to carry out a proper analysis of Baskan Gida's creditworthiness, this was a mission designed to assist the Banks in repelling that allegation. It was also a self-serving mission, since Mr Lewis' admitted mistake in mis-understanding Baskan Gida's audited accounts lies at the heart of that part of the Ferrero Defendants' case.
30. While I have no doubt that Mr Lewis' views on the differences between structured trade finance and traditional bank lending were sincerely held, the fact that they served the interests both of his continuing employer and himself inevitably detracted to some extent from his independence, and therefore from his reliability as a witness. As will appear when I state my findings of fact, the evidence as a whole shows that Mr Lewis considerably exaggerated those differences. The result is that I found his opinions about those matters less than wholly persuasive, despite his considerable experience in the field of structured trade finance. His evidence about factual matters which did not relate to the alleged shortcomings of his part in the Banks' due diligence was nonetheless generally reliable.
31. The Banks then interposed Mr Izzet Gokdeniz, a Turkish businessman with considerable experience in international trade, who had given evidence in a criminal trial in Turkey in which members of the Baskan family were accused of defrauding one of Baskan Gida's Turkish banks. He was (I think to his considerable surprise) called only as a witness of fact, rather than as an expert, and found it hard to contain his firm opinion that, despite their acquittal, the Baskans were guilty as charged. Nonetheless his factual evidence, which he gave through an excellent interpreter, was careful, helpful and reliable.
32. Next to be called was Mr van Broekhoven. I have already identified him as the relationship manager for the purposes of KBC's relationship with Baskan Gida. He left KBC in May 2002. As a result, although his two witness statements were made after careful consideration of the underlying documents, he lacked that ongoing relationship with KBC which contributed both to the strengths and weaknesses of Mr Lewis's evidence, being less well prepared than Mr Lewis and therefore having his recollection less well refreshed, but on the other hand being inherently less likely to have his independence undermined by loyalty to a continuing employer.
33. That said, Mr van Broekhoven was, as the person within KBC responsible for the day to day conduct of its relationship with Baskan Gida, inevitably the target of allegations of negligence during cross examination, and therefore in the position of someone having to defend his reputation for competence and skill.
34. He came across as a fluent and intelligent witness, ready to engage openly and with good humour with the subject matter of his cross examination. In my judgment he believed that he could recall from memory rather more than that which he really remembered about events taking place 7 years ago, in particular in relation to an important meeting he had with Mr Casale in July 2001, where his apparent recollection in cross-examination went further either than the meeting note made at the time, or in his first witness statement, made in July 2008. Apart from that tendency to over- remember, he was a generally reliable witness of fact.
35. As for his response to allegations of negligence, it seemed to me that, initially, he adopted a rather unsatisfactory refusal to accept responsibility for matters of criticism, although as his lengthy cross examination continued, this tended to wear off.
36. Next to be called was Mr Anthony Parello, BTM's first witness. He was at the material time in charge of the Import Section within the Trade Finance department of BTM, which was itself part of the European Operating Service Centre of the bank, providing back office services for BTM as a whole. His job was to supervise that part of the operation of the Facility which consisted of BTM'S response to utilisation requests from Baskan Gida pursuant to the facility agreement, an important but essentially mechanical function which required him and his team to satisfy themselves that upon every request for a draw-down under the Facility, Baskan Gida had provided appropriate documentation to justify the issue of a letter of credit in its favour, pursuant to the detailed terms and conditions of the agreement.
37. Mr Parello was an obviously honest and, on matters of fact, wholly reliable witness. He saw his job as consisting of following orders and procedures without troubling himself as to the underlying reasons for them. As a disciplined foot soldier, rather than general, he was unable to respond to cross examination as to the underlying purpose of various of the detailed conditions with which Baskan Gida had to comply.
38. A little more insight into the thinking behind BTM's operational processes was later provided by his manager Lance Wenden, then the Head of Trade Finance Products in BTM's European Operations Department. He was another generally reliable witness on matters of fact, but was understandably reluctant to engage in any in depth analysis of the alleged shortcomings in BTM's part of the operational management of the Facility. To the extent that he was prepared to do so, I found his explanations for alleged departures from the strict letter of the Facility Agreement generally convincing.
39. Of more importance to the issues of contributory negligence was the evidence of Mr Philip Graham, who was at the material time the Deputy General Manager of the European Credit Division of BTM, who therefore played an important executive role in BTM'S decision to participate in the Facility. He was, in every respect, a most impressive witness, demonstrating a thorough appreciation of the issues, gained after a careful refreshment of his memory from the surviving documents, and a careful reading of the criticisms of his bank's conduct set out at length in the report of Mr Rex, the Ferrero defendants' banking expert. Although still employed by BTM, and the target of a relatively small part of the criticisms of its conduct pursued in cross-examination, I detected no defensiveness or partiality in his evidence. On the contrary, he made sensible admissions after evidently careful thought about matters which might have been handled better, and provided helpful explanations of aspects of the detail of the conduct of due diligence in relation to the proposed Facility.
40. Mr Graham was followed by Mrs Soskin-West, at the material time head of the International Banking division of BTM. She was the immediate superior in the BTM chain of command to Mr Byles, the relationship manager with immediate day to day responsibility for BTM's negotiation of the Facility, and continued to be employed by BTM at the time of the trial. She had less immediate contact with the negotiation and operation of the facility than BTM's other witnesses, and therefore less to contribute. She was however directly involved in BTM's response to Baskan Gida's default in early 2002, and provided a useful insight into the thinking behind its conduct at that stage. She was a generally reliable and straightforward witness, although not as precise in her recollection or as penetrating in her analysis as Mr Graham.
41. Mr Byles was an important and valuable witness, both because of his central involvement in BTM's part in the story, and because, having ceased to work for BTM in 2006, his evidence was unaffected by loyalty to a continuing employer. He joined BTM in late 1998 with considerable previous experience in factoring transactions, and he was BTM's main point of contact both with Baskan Gida and with Ferrero throughout. I found him to be a helpful witness, the reliability of whose recollection was fortified by his ready acknowledgment of the gaps in his memory, and a determination not to guess or reconstruct. Although personally criticised for aspects of BTM's alleged negligence, he did not generally allow his evidence to be marred by defensiveness or prevarication, and made sensible admissions in relation to relevant shortcomings in his and BTM's due diligence. My only reservation as to the independence of his approach to the matters of criticism of the Banks' conduct was that his overall conclusion was that their loss was primarily the fault of KBC (in the operational monitoring of the Facility) rather than BTM. In this respect he succumbed to the inevitable temptation to pass the buck.
42. Mr Byles' evidence was interrupted by a pre-arranged video conference, over two days, for the examination of Mrs Gulay Pay, through an interpreter, in Giresun. She was engaged successively by Baskan Gida, Aksu Gida and by Baskan Yuksel as an export supervisor at the Giresun factory, before being made redundant in 2004. Subject to the combined difficulties imposed by the language and locational barriers, she provided helpful and honest evidence from her experience in that capacity, both in explaining procedures within the factory, where she personally prepared Baskan Gida's export documents and invoices, and in describing from her perception as a middle ranking member of staff Baskan Gida's descent into insolvency, and the manner in which its business was from February 2002 taken over by Aksu Gida and then by Basken Yuksel. Apart from a moment of unease at questioning which implied that she had participated in conduct contrary to a Turkish court order, I found her approach to giving evidence to be a careful and dispassionate process of assisting the court, without any apparent agenda of her own.
43. BTM's final employee witness was David Philbin, at the material time a credit analyst in its European Business Division. He performed much the same supporting role for Mr Byles that Mr Lewis performed at KBC for Mr van Broekhoven, until he was re-assigned to other credit analysis tasks in early October 2001. As a result he played no part in the later stages of the Banks' due diligence, and no part in the operational or post-default stages of the Facility, and was for that reason subjected to a much shorter and less searching cross-examination than either Mr Lewis or Mr Byles. He remained employed by BTM at the time of the trial.
44. Mr Philbin's shorter cross-examination and more limited role in the relevant events meant that he had less opportunity than the Banks' main witnesses to settle in as a witness, and I had less opportunity to form a detailed view of his reliability. Nonetheless his factual evidence appeared to be reasonably reliable, and his response to criticisms of the Banks' conduct much less a matter of mission that was that of Mr Lewis, despite his continuing loyalty to BTM as his employer. He made some sensible admissions, for example as to the importance of studying Baskan Gida's audited accounts (which became available only after he had been re-assigned to other work). In certain respects however I found his attempts to respond to alleged failures to identify weaknesses in the financial information about Baskan Gida which he did study rather unconvincing. He did at the end of his evidence provide a useful interpretation of his manuscript notes of his and Mr Byles' visit to Giresun in July 2001, when they both met Mr Casale.
45. The Banks' final witness of fact was Mr Rahmi Aslan (usually mis-spelt Arslan in the documents, including his own witness statement). He was the store keeper employed by SGS Turkey to manage the warehouse ("Warehouse 2") at Baskan Gida's Giresun factory premises which was used for the storage of hazelnuts while pledged to the Banks pursuant to the Facility. He gave his evidence through an interpreter, since he spoke no English at all. He was a solid and reliable witness with no axe to grind who gave a clear explanation of his limited role at Giresun, and of the warehouse receipts and other documents which he created while employed there.

Ferrero's Witnesses

46. Ferrero's first witness was Mr Antonio Do, the Chief Operating Officer of Ferrero at the time, Chairman of Ferrero Industrial, and a member of the main board of Ferrero International SA, the group's holding company. He remained employed by Ferrero at the time of the trial, as head of Soremartec, the unit which provides the design and testing of Ferrero's new products. As might be expected from such a senior employee, Mr Do's evidence was carefully prepared, consistent and usually authoritative, but inevitably limited in its recollection of the detail of the Ferrero relationship with Baskan Gida and the Banks, being matters mainly handled by his juniors Mr Rosa Brunet and Mr Casale.
47. Mr Do made no secret of the fact that he was a witness with a mission, namely to uphold the good name of Ferrero against allegations of fraud, dishonesty and reprehensible business dealings. He sought to do so not by any display of indignation, but with a well-controlled, patient and fully engaged response to two days of cross-examination, conducted initially in English but latterly mainly in Italian through an excellent interpreter, as his perceived need for precision in his replies came to tax the considerable skill with which he spoke English.
48. Mr Do did not stand personally accused of fraud or dishonesty in any part of the Banks' meticulously pleaded case. Nonetheless, two related aspects of his evidence gave me reason to doubt whether his evidence in Ferrero's defence was entirely honest. The first was Mr Do's denial of any knowledge at the time that the Baskans remained the controlling force behind both Aksu Gida and Baskan Yuksel, when Ferrero continued to trade with those companies in and after February 2002, a trade which continued after it became apparent to Ferrero that the Baskans could no longer be trusted. The second related to Mr Do's reason for the failure of Ferrero to respond more constructively to requests by the Banks for information after Baskan Gida's default in February 2002. Mr Do put this down to the Banks' refusal to provide a requested agenda for proposed face to face discussions, but there was no documentary or other evidence that this request had been communicated to the Banks, and I reject that explanation.
49. I consider that Mr Do and (as will appear) Mr Rosa Brunet and Mr Casale all decided to lie about the first of those matters, with the result that Mr Do, who was in charge of Ferrero's response to the Banks' claims, and who therefore faced the main cross-examination about it, was also driven to lie about that as well. Where a witness is prepared to lie when he perceives it to be in his or his company's interests to do so, it deprives his evidence of inherent reliability. Nonetheless, much of Mr Do's evidence was either unchallenged or corroborated, and it was, much more often than not, reliable for that reason.
50. Mr Alessio Casale was at the material time first the Deputy Purchasing Manager and, from January 2002, a "Direttore" (i.e. manager) of Ferrero Industrial, with particular responsibility for the purchase of hazelnuts. He had as a result the day to day responsibility for the conduct of Ferrero's relationship with Baskan Gida, which he conducted on an almost daily basis on the telephone and by faxes and emails, mainly with Melih Baskan, who was in charge of Baskan Gida's export sales. Mr Casale has continued to be employed by Ferrero, and is now a manager within Ferrero Trading SA. In addition to his native Italian, he is a fluent speaker of English, with a working knowledge of German and French as well. He and the Baskans used English as their common language. He also gave his evidence in English, with very occasional help from an interpreter.
51. Mr Casale came to the witness box quite extraordinarily well prepared to give evidence, extending to the minutest detail, of matters which occurred several years ago. He had a very good but not absolutely infallible memory for detail, occasionally appearing genuinely to recall points of detail which proved to be inaccurate by reference to contemporary documents. These occasional inaccuracies in no sense detracted from the generally comprehensive quality of his evidence, which he gave at great length in a written witness statement and, in cross-examination, over ten days. Generally, he displayed an articulate determination to address head on the questions he was asked and, despite a cross-examination aimed at challenging his honesty in numerous respects, remained calm, courteous and cooperative throughout.
52. The care and thoroughness which he brought to giving of evidence was matched, in my judgment, by similar qualities which he applied to his work for Ferrero at the time. Although it was put to him that he had participated in the Baskans' fraud on the Banks out of a corrupt motive, I have no hesitation in concluding that, throughout his dealings with the Baskans on Ferrero's behalf, he acted in pursuit of what he genuinely believed to be Ferrero's best interests.
53. I have unfortunately been driven to the conclusion that in certain specific respects Mr Casale persuaded himself that Ferrero's interests required him to lie in the evidence given at this trial. I shall describe those aspects of his evidence in due course, together with my reasons for rejecting both his evidence, and his professed honesty, on those occasions. In summary, they comprise first, his denial that he knew in substance the meaning of the English word "assignment" when negotiating in careful detail the English language versions of documents which involved Ferrero in the structure of the Banks' Facility to Baskan Gida, in the later part of 2001. Secondly, and this is a falsehood into which Mr Do and Mr Rosa Brunet were also drawn, he untruthfully denied knowing that the Baskan family were at all times in de facto control of Aksu Gida, and later Baskan Yuksel, while they were simultaneously pretending to have withdrawn from the hazelnut market. Thirdly, he was not frank with the court about the extent to which, and the time at which, he suspected that the Baskans were dealing otherwise than honestly with the Banks.
54. I consider it possible that the shortcomings in Mr Casale's evidence about the first and third of those matters may have been caused or at least substantially contributed to by his convincing himself into a state of denial about the extent of his knowledge at the material time, faced with the unpalatable consequences for Ferrero and for himself personally of facing up to the reality now that the matter has become the subject of major litigation. But I have been unable to persuade myself to any such sympathetic conclusion in relation to his denial that he and his colleagues within Ferrero Industrial knew that the Baskans were, throughout, in de facto control of Aksu Gida and Baskan Yuksel. That is something impossible for him to have forgotten, or even to have deceived himself about.
55. The result of having come to those conclusions about Mr Casale's preparedness to lie by no means leads to a conclusion that his evidence as a whole is to be rejected. On the contrary, large parts of it accord both with contemporaneous documents, other corroborative evidence, and with the balance of probabilities. Nonetheless, his readiness to depart from the truth in accordance with his perception of Ferrero's interests leads to the result that I have been unable to place unquestioning reliance on Mr Casale as a witness of truth, but have been forced to address all of his detailed evidence with an open mind as to its truth and accuracy, as dictated by a consideration of the evidence as a whole.
56. I have reached the same overall conclusion about Mr Rosa Brunet. He was throughout the material period Mr Casale's immediate superior at Ferrero Industrial, as one of its two capital Managing Directors, his responsibility being for raw materials procurement, including, but by no means limited to, hazelnuts. As Mr Casale's superior he was involved in the important decisions about Ferrero's relationship with Baskan Gida, but much less with the day to day detail. He retired from Ferrero in 2005.
57. As a witness, Mr Rosa Brunet faced difficulties considerably greater than those of Mr Casale, in particular because first, his English was less good, so that he needed much more assistance from an interpreter. Secondly he was experiencing real difficulties with his eyesight, which made it hard for him to read important documents put to him in cross-examination. More generally, and not least because of his detachment from Ferrero for the last few years, his recollection of events, and in particular dates and times, was much less precise than that of Mr Casale. He was less focused in his evidence and occasionally a little muddled. Nonetheless he showed no less enthusiasm than Mr Casale to engage with his cross-examiner, in providing full and usually direct answers to questions.
58. Unfortunately for the reliability of his evidence, Mr Rosa Brunet associated himself with the general Ferrero denial of knowledge that the Baskans were in de facto control of Aksu Gida and Baskan Yuksel, a reality which, however deniable in the sense of there being no surviving documents to show that he knew otherwise, he cannot in my judgment have forgotten. For reasons which I shall in due course explain, an understanding of who in 2002 really controlled Aksu Gida and Baskan Yuksel was essential to Ferrero's commercial decision-making at the time, and must have formed a lasting impression on those, including Mr Rosa Brunet, who took those decisions.
59. The result is that, as with Mr Casale, I have been obliged to test Mr Rosa Brunet's evidence, step by step, against the evidence as a whole, and the general probabilities, without being able to attribute inherent reliability to it. But as with Mr Casale I have no doubt that Mr Rosa Brunet's conduct, both at the time and at the trial, has been motivated by his genuine perception of Ferrero's best interests, rather than by any corrupt motive of his own.
60. Mr Michael Bolowich was, on the Banks' pleaded case, the third member of the fraudulent triumvirate within Ferrero, but at the end of his cross-examination Mr Wardell promptly acknowledged, in response to my enquiry, that in the light of the evidence as it then stood, he could no longer pursue any allegation of dishonest participation by Mr Bolowich in the matters complained of. That sensible concession accorded with the impression which I had by then formed of him, namely that of a relatively lowly but wholly honest fringe participant in the relevant events.
61. His official role at the material time was that of deputy purchasing manager within Ferrero Industrial, with particular responsibility for Ferrero Germany's requirements for certain raw materials, including hazelnuts. Prior to the centralisation of Ferrero's purchasing he had been a buyer of certain raw materials, not including hazelnuts, for Ferrero Germany as its direct employee, with executive responsibility for buying decisions. After the establishment of Ferrero Industrial he remained in Germany in effect in a branch office of Ferrero Industrial there, but since all executive buying decisions were made by Mr Rosa Brunet's team in Italy, his function was reduced to the purely administrative task of placing purchase orders for Ferrero Germany, on strict written instructions, usually from Mr Casale in relation to hazelnuts, with no executive responsibility or discretion of his own. Having thus been marginalised, he eventually retired from Ferrero in June 2004, aged 63.
62. Four years' separation from Ferrero did nothing to assist his long term memory of relevant events, and even his short term memory proved less than reliable under cross-examination, giving opposite answers to the same question put in cross and then re-examination on two consecutive days without any apparent appreciation that he was contradicting himself. Nonetheless he clearly did his best honestly to assist the court to the limited extent that he could, and was generally careful to distinguish between reconstruction and recollection. He was open and frank in his answers, readily agreeing with the logic of questions where appropriate, and disclosed no partiality either for or against Ferrero and its senior staff, while disclosing that he and Mr Rosa Brunet had a less than entirely harmonious working relationship. He was therefore a helpful and reliable witness, to the limited extent that his recollection permitted.
63. Ferrero's last witness was Mrs Mariapia Di Matteo who, both at the material time and when she gave evidence, was a senior member of the Ferrero Italy treasury department, reporting at the material time to Mr Salomone. Her responsibilities included making sure that payments the subject matter of Baskan Gida invoices to Ferrero Italy were, once approved, paid efficiently and on time by Ferrero's Italian bankers to the appropriate Baskan Gida recipient bank. In that role, she played an important part in Ferrero's contribution to the finally agreed wording upon which Ferrero was asked by Baskan Gida to acknowledge the assignment to the Banks of receivables arising from its sales to the Ferrero operating companies.
64. Subject to one or two small points where I doubted the precise accuracy of her recollection, Mrs Matteo proved to be a thoroughly good and reliable witness. The combination of a good memory and careful preparation enabled her to provide useful evidence as to the functioning of Ferrero Italy's treasury department, and its particular role and responsibilities in relation to the making of payments for deliveries by Baskan Gida. She gave her evidence with conspicuous care and caution, and her evident desire to assist the court was in no way affected by any partiality towards her continuing employer.

Mr Abidali's Witnesses

65. Mr Abidali went first, and was cross-examined at great length, and in minute detail by reference to the surviving documents. His behaviour in the witness box had many of the trappings of a reliable witness. For example, he gave his evidence firmly and with clarity. He was well prepared, and had an evidently good memory of particular events. He was articulate, generally good humoured, and endured a lengthy and persistent cross-examination challenging his honesty and integrity with unfailing courtesy and restraint.
66. That said, he proved to be a thoroughly unreliable witness, essentially because of the hopeless conflict between both the gist and the detail of his written and oral evidence, and the usually clear contrary picture presented by the contemporaneous documents, many of which had either been written to or by him. Those conflicts, almost all of which I had found myself compelled to resolve against Mr Abidali, will become apparent from my findings of fact about the matters in which he was personally involved.
67. My conclusion that Mr Abidali is not an honest man is reinforced by three aspects of his conduct which occurred before he began to give evidence at trial. The first was that he took an active part in the formulation, execution and backdating of apparently contractual documents in 2002 which were plainly designed to give a false and misleading impression to outsiders (i.e. they were sham documents) purporting to record the circumstances and terms of the transfer of assets between Baskan Gida and Aksu Gida. The second was that he authorised false explanations of those documents by Indo-Med in its pleadings in this litigation before Mr Abidali himself was joined as a party, supporting them by statements of truth and witness statements, including an obviously false witness statement in support of Aksu Gida's case in Turkish proceedings in 2002. The third was that while Mr Abidali was responsible on behalf of Indo-Med for its compliance with its disclosure obligations in these proceedings, Indo-Med failed to disclose the overwhelming bulk of the documents which, when subsequently disclosed by its liquidator, revealed the truth of the matters about which Mr Abidali had previously provided false evidence. When cross-examined about all these matters at trial, Mr Abidali provided explanations and excuses which I found wholly unpersuasive. My reasons for so doing will emerge in detail in the next section of this judgment.
68. Mr Ashraf Dahod was called to corroborate much of Mr Abidali's evidence. In terms of demeanour, he was just as, if not more, persuasive than Mr Abidali, giving his evidence with an apparently convincing air of quiet restraint. Furthermore, Mr Wardell for the Banks had much less by way of inconsistent contemporary documents with which to undermine his account of events, since Mr Dahod had allowed Mr Abidali to act as his agent for almost all purposes in relation to his substantial investment in Baskan Gida and Aksu Gida.
69. His credibility was however fatally undermined towards the end of his cross-examination by his resolute refusal to face the realities displayed in an email sent to him by Mr Abidali on 14^th October 2002, and in his reply to it, both of which I shall have to describe later in some detail. The gist of them was a description of Mr Dahod's interest in the hazelnut business by then being run by Aksu Gida and Baskan Yuksel as that of a pure financier of a business belonging to the Baskans, with a right to a profit share, rather than that of an owner of the business. This was flatly contradictory not merely to Mr Dahod's evidence about the emails, but to the whole of his and Mr Abidali's case about their participation in Aksu Gida. Unless satisfactorily explained, which it was not, it revealed Mr Dahod as a liar, not merely about a particular incident, but in the central aspects of his attempted corroboration of Mr Abidali.
70. It may be that Mr Dahod's untruthfulness can be put down in part to his having been deceived throughout by Mr Abidali, and having since then been in denial about the misconduct of a fellow member of a tightly knit religious community. Unlike Mr Abidali, Mr Dahod does not stand accused of active participation in a conspiracy against the Banks, and I make it clear that nothing in my description of his misconduct as a witness should be taken to imply a conclusion that he was indeed a fellow conspirator at the time. Nonetheless, he has had plenty of time before giving evidence in these proceedings to choose between telling the truth and lying in support of Mr Abidali, and my clear conclusion is that, whether for reasons of loyalty or otherwise, he chose to do the latter.
71. Mr Zoeb Raniwala was a chartered accountant who provided part-time accountancy assistance to Indo-Med during the relevant period. He was called to corroborate Mr Abidali's account of certain mainly peripheral aspects of his evidence, having no close involvement in the central aspects of the case affecting Mr Abidali, and in particular of the role of Aksu Gida. He came across as a fair-minded, relatively independent and therefore generally reliable witness, whose evidence I found easy to accept, albeit that, save in relation to one matter concerning the use made of an account of Safe Holding Limited at Habib Bank in Zurich, it was not of central materiality.
72. The last witness to give oral evidence for Mr Abidali was his English solicitor, Mr John Trotter, who was at the material time the senior partner of the London solicitors firm of Bates Wells and Braithwaite. He was called to provide a character reference for Mr Abidali, and to corroborate certain aspects of the non-critical detail of Mr Abidali's account of the facts in issue.
73. For most of his time in the witness box, Mr Trotter gave the impression of a man in real fear and trepidation, as well he might, because the important part of his participation in the matters in issue was as the English legal contributor to the drafting of the sham agreements by which the Baskans and Mr Abidali sought to present to the outside world a thoroughly misleading picture of the circumstances of the transfer of the hazelnut business from Baskan Gida to Aksu Gida. I have no doubt that Mr Trotter's apparent trepidation stemmed from a fear that it would be suggested to him in cross-examination that he had played a part in that process thoroughly unbecoming of an English solicitor. In the event, Mr Wardell chose to save his ammunition for more important targets and, as this became apparent to Mr Trotter, his confidence as a witness visibly increased.
74. The result of Mr Trotter being spared a more searching examination of his role was that the limited ambit of the matters about which he gave evidence did not include much about which, in the event, I found it necessary to disbelieve his evidence. To the limited extent that he therefore provided corroboration for parts of Mr Abidali's evidence, it did him no good on the central matters in issue. Mr Trotter did provide Mr Abidali with a character reference, but coming from a solicitor who assisted in the drafting of documents which, in my judgment, he ought to have appreciated were being improperly backdated, the reference was of no assistance to Mr Abidali. Indeed, Mr Trotter's role in the matter can only have been honest if he was thoroughly deceived by Mr Abidali.
75. Finally, Mr Abidali tendered under the Civil Evidence Act a statement of a Mr Hatim Husain, an employee of Habib Bank Zurich, a member of Mr Abidali's and Mr Dahod's religious community and a director of Safe Holding Limited. His very brief evidence was not accepted by the Banks in the sense of going unchallenged, but Mr Husain was proffered for cross-examination only upon condition that the Banks paid his expenses, which they declined to do. The result of this rather childish standoff, (bearing in mind the vast costs employed generally in this litigation), was that Mr Husain was not tendered for cross-examination, and I was left to give such weight to his witness statement, tendered under the Civil Evidence Act, as I thought fit. In the event, Mr Husain's evidence corroborated Mr Abidali's case only in relation to the issues affecting Safe Holdings Limited, as to which, I did not in the end disbelieve Mr Abidali.

Expert Witnesses

The Banking Experts

76. The lengthy and detailed allegations of contributory negligence on the part of the Banks contained in Ferrero's defence inevitably led to both parties calling banking experts. Both of them brought to bear long experience in banking, in consulting and advisory work relating to banking transactions, and in the preparation and giving of expert evidence on banking issues. The independence of each expert went wholly unchallenged.
77. Ferrero called Mr Paul Rex, now the managing director of the banking and securities consultancy company GBRW Limited. His banking career began as a graduate trainee at Chemical Bank in London in 1974. By the time he left in 1988 he was vice-president and head of ship finance and commodities finance within Europe. From 1988 to 1992, as assistant general manager at the London branch of Crédit Agricole, he was head of its Corporate Banking and Finance Division with responsibility, among many other things, for its commodity and trade finance units. Since 1993 Mr Rex has been engaged full time in consulting.
78. In cross-examination Mr Rex acknowledged that he had not at any time in his career led an emerging country structured trade finance transaction from start to finish. Nonetheless his responsibilities had included the risk analysis of such transactions from the perspective of a participating bank in a syndicate, the management of the department regularly taking a wide range of country risk, and the preparation of expert evidence on pre-export facilities with characteristics similar to the Facility in the present case. In my judgment the range and depth of Mr Rex's banking experience qualified him well for his task in this litigation, even if he had not sat precisely in the hot-seats occupied by the Banks' main witnesses. Both in his reports (including the joint statement of the two banking experts) and in his response to cross-examination, Mr Rex proved to be measured, articulate, reasonable and persuasive, keeping an open mind to the consideration of alternative points of view, and giving his opinions with care and precision. All in all I found him to be a most impressive witness.
79. The Banks' expert Mr Robin Bryant gained his direct banking experience from 1960 through to 1991 at Westminster Bank Limited, County Bank Limited, Standard Chartered Bank plc, NM Rothschilds & Sons Limited and finally at Riggs AP Bank Limited, dealing with a wide range of lending transactions, as well as sitting on credit committees. He had even less hands-on experience of structured trade finance for emerging market exporters than Mr Rex, although his time at Riggs AP Bank Limited (between 1989 and 1991) necessarily involved international trade finance, 'AP' standing for Anglo-Portuguese. Since 1992 Mr Bryant has been engaged in the provision of consultancy services to the banking industry, including the preparation and giving of expert evidence in a large number of cases. Again, I am satisfied that his broad and long experience well qualified him to provide expert evidence in the present case, even though his career had not been focused upon the particular type of lending with which it is concerned.
80. As will appear from my conclusions in due course, to the effect that the Banks were indeed contributorily negligent in a number of respects, Mr Bryant had a difficult task in acting as an expert witness for the parties defending those allegations. Nonetheless, both at his meeting with Mr Rex, in his written reports and in particular in cross-examination, Mr Bryant sensibly acknowledged, without partiality or undue defensiveness, that the Banks had indeed fallen short of the standards to be expected of reasonably competent bankers at various points in an otherwise lengthy and thorough due diligence, and he managed to keep well in mind the distinction which needed to be made between the acknowledgement of shortcomings, his evident sympathy for the individuals concerned, and his humane understanding of the reasons why they had fallen into error.
81. All in all, Mr Bryant was another impressive expert witness, who provided substantial assistance to the court.

The Hazelnut Market Experts

82. The Banks called Mr Murat Solak who is currently the foreign trade manager of Basaran, a large Turkish producer and exporter of hazelnuts. His twenty years' experience in the hazelnut business included working full time for another large exporter Sezginler from when he left university in 1992 until called to do military service in 1998. In 1999 he formed a joint venture with a Dutch company with a view to setting up a new hazelnut producing, exporting and trading business, called Arasco. This did small-scale business (by comparison with Baskan Gida) in exporting hazelnuts during the material period, not exceeding about 2,000 metric tons ("mt") of kernels per annum and doing little business with the chocolate making industry. Arasco closed its processing and exporting business after a bad crop in 2003.
83. Notwithstanding that his experience during the material period was not directly comparable with a business like Baskan Gida's, Mr Solak demonstrated an impressive knowledge and understanding of the Turkish hazelnut market (including the export market) both before, during and after the relevant period, so that he was well qualified from the perspective of experience to provide the expert opinions requested of him.
84. As a Turkish resident and speaker throughout, this no doubt added to the depth of Mr Solak's knowledge of the market, but his relatively limited English meant that he delivered his written reports and most of his oral evidence in Turkish, using English only where he encountered difficulty in getting across a technical point through a non-technical interpreter. His English was, for most of the time, amply sufficient for him to understand the gist and usually the nuances of questions asked of him in cross-examination, with the need for only limited assistance from interpreters for that purpose.
85. Notwithstanding his long and broad experience, Mr Solak had no prior dealings with Ferrero or with Baskan Gida. In one sense this was an advantage, since it provided for him a standpoint of complete independence. His absence of experience in dealing with Ferrero and comparable foreign customers on the scale manifested by the transactions in issue, particularly during the relevant period, did constitute a disadvantage in terms of his attempt to assess the commerciality or otherwise of the transactions in issue from Ferrero's point of view as customer.
86. Mr Solak had no previous experience as an expert witness, or more generally of participating in the litigation process. To some extent this led to his offering enthusiastic participation in the debate about certain contested issues of fact which went well beyond the ordinary role of a market expert. Because they were often based upon hearsay and rumour, it was necessary for me to treat those contributions with some caution.
87. More generally however, Mr Solak presented a lively, intelligible and usually vivid picture of the Turkish hazelnut market at the material time, which I found to be of considerable assistance. My main reservation with his evidence was that he demonstrated a certain rigidity of thought in relation to the complicated issues about Baskan Gida's likely processing costs, which appeared to stem from an a priori assumption that if the businesses with which he was connected could not process particular standards and specifications of hazelnuts at less than a certain price, then none of his competitors, including Baskan Gida, could do so either. The consequence was that Mr Solak's opinions as to Baskan Gida's likely processing costs constantly came out at the top end of the range of comparables made available to me, being invariably significantly in excess of the estimates provided by Ferrero's market expert, and very substantially in excess of those cost estimates provided by Baskan Gida to Ferrero at the material time. This led to the inevitable question in cross-examination why Baskan Gida might be thought likely to have understated its costs to its principal customer when negotiating prices on a cost-plus basis, to which Mr Solak had no ready or persuasive answer. Similarly, he had no answer to the question why, from the limited documentation available, it appeared that Baskan Gida's principal competitor for Ferrero's business, Oltan Gida, appeared to quote similar costs to those of Baskan Gida, and to sell at similar prices, when dealing with Ferrero. His main reason for equating his own costs with the likely costs of Baskan Gida was that, in his view, there were no real economies of scale to be achieved when processing hazelnuts. I did not find that evidence particularly convincing, however understandable coming from a much smaller participant in the market at the material time than Baskan Gida.
88. Mr Solak's opinions about the commerciality or otherwise of the transactions in issue from Baskan Gida's perspective appeared to be heavily influenced by his assumption that Baskan Gida had gone out of business in 2002 due to commercial misjudgements made in the conduct of its hazelnut business. As he put it graphically in cross-examination, if Baskan Gida had got it right, the litigation would not be taking place. It is in fact impossible to know precisely what caused Baskan Gida's failure, not least because in their own explanations to Ferrero and Indo-Med members of the Baskan family attributed their problems to a wholly distinct biscuit manufacturing business. It was therefore necessary for me to approach this assumption, and the consequently dim view which Mr Solak had of Baskan Gida's business decision-making, with some caution.
89. Ferrero's hazelnut expert was Mr Thomas Haas-Rickertsen, whose working life since 1973 has been closely involved in the Turkish hazelnut trade, and who from 1982 to date has been the managing partner of Pisani & Rickertsen, a hazelnut trader founded in Hamburg in 1905. In 1988 Pisani & Rickertsen founded an Istanbul based hazelnut cracker and processor known as Progida AS in partnership with a Mr Ufuk Ozongun. By 2007 Pisani & Rickertsen working together with Progida constituted together the second largest Turkish exporter of hazelnut kernels.
90. Mr Haas-Rickertsen's qualification as a hazelnut expert was enhanced by his having sat part-time as a judge to the Court of Arbitration of the Waren-Verein Der Hamburger Borse e.V, in connection with quality and contract disputes relating to commodities including hazelnuts. Although mainly based in Hamburg rather than Turkey, Mr Haas-Rickertsen made numerous visits to Turkey as part of his responsibilities. Both he and Mr Solak were able to have recourse for their opinions about market prices in 2001-2 to detailed manuscript notes taken by them at the time. Whereas Mr Solak obtained the market information for his notes directly, Mr Haas-Rickertsen acknowledged that he derived a substantial amount of his market information from his partner Mr Ozongun, who was based full-time in Turkey and, unlike Mr Haas-Rickertsen, spoke fluent Turkish. Mr Haas-Rickertsen's first language is German, but he gave his evidence almost exclusively in very accomplished English, with occasional assistance from a German interpreter.
91. Mr Haas-Rickertsen was subjected to a sustained attack by Mr Wardell in cross-examination (and in oral and written submissions) upon the basis of an alleged lack of independence. The foundation for this attack was that Ferrero was a substantial customer of Mr Haas-Rickertsen's firms and that, in April 2007, he had declined an invitation to act as the Banks' expert, partly upon that ground. How, Mr Wardell asked him, could he give independent expert evidence if he considered himself disabled from doing so on the instruction of a party opposed to Ferrero? The gist of his response was that his experience as an arbitrator, not infrequently appointed as one of two or three, by a customer of his firm, had in the past enabled him to develop a practice of acting independently in relation to disputes involving customers and that, with his long experience of the hazelnut market, he believed that he could do so in the present case.
92. Mr Wardell also attacked Mr Haas-Rickertsen's written evidence as containing speculation which, generally, favoured Ferrero as against the Banks. He also accused him of deliberately ignoring his own large firm's processing costs, when providing comparables based on much smaller businesses.
93. Although an apparent lack of independence no longer goes to the admissibility of an expert's evidence (see R (Factortame Ltd) v. Secretary of State for Transport, Local Government and the Regions (No 8) [2003] QB 381), such that expert evidence is admissible even if given by a full-time employee of the party calling the expert, it plainly goes to the weight of the evidence given. In my judgment, taking Mr Haas-Rickertsen's written and oral evidence as a whole, the fact that Ferrero remained a substantial customer of his firm during the whole of the period when he was preparing and giving expert evidence did not significantly detract from the independence with which he gave his opinions on matters of market expertise. By contrast, there was some force in Mr Wardell's submission that in areas where he was forced to speculate, Mr Haas-Rickertsen appeared to dwell upon speculations favourable to Ferrero's rather the Banks' case. In most instances this occurred in relation to the frequently arising question whether the conduct of Baskan Gida in entering into allegedly disadvantageous contracts with Ferrero could have had any ordinary commercial motive, the Banks' case being that it was transparently the result of terminal financial difficulties. In those respects, Mr Haas-Rickertsen's starting point usually was an assumption that Baskan Gida had a commercial reason for acting as it did, other than pure financial desperation, and much of his written evidence about that issue consisted of the identification of possible explanations why that might be so. Mr Haas-Rickertsen made no secret of the fact that in those respects he was forced to speculate in the absence of hard evidence, and I consider that his approach derived not so much from any lack of independence, as between the Banks and Ferrero, but from an assumption borne of long experience in dealing in a market in which Baskan Gida was the largest player, that it was a hard-nosed commercial operator, inherently likely to have good reason for doing what it did. In this respect he started from a diametrically opposed assumption to that of Mr Solak, and their opposite starting points explained much of the large differences in their opinions on this issue. I have no doubt that both opinions were genuinely held.
94. In that respect, both Mr Solak and Mr Haas-Rickertsen understandably placed reliance, in their assessments of the rationale for Baskan Gida's conduct from time to time, upon their common perception (I infer shared generally in the Turkish hazelnut market) that Baskan Gida had a long established tendency to engage in large scale short speculation; that is, accepting commitments to sell in the future without covering those obligations with stocks in hand, hoping that a fall in the market between contract and delivery would enable it to make a profit by buying the necessary raw materials at a lower price.
95. In cross-examination, when faced with alternative analyses of financial extremity and ordinary commercial rationale, I found that Mr Haas-Rickertsen approached these questions with an open and independent mind, illustrated by the fact that, by the end of his cross-examination, he readily assented to the proposition that some of Baskan Gida's last contracts, before ceasing business in early 2002, looked like the last desperate throw of the dice by a business facing financial ruin. My conclusion on the issue of independence is therefore that, while it has been necessary for me to treat Mr Haas-Rickertsen's evidence with more caution that that of Mr Solak because of the Ferrero customer connection with his firm, in general Mr Haas-Rickertsen demonstrated a properly independent approach to the giving of his opinions, as might be expected from his long experience as a commercial arbitrator.
96. More generally Mr Haas-Rickertsen made up for his relative lack of direct day-to-day involvement in the Turkish market at the material time (as against that of Mr Solak) by bringing a thorough and rigorous analytical approach to his task, and by his readiness to explain in detail the underlying reasons for his opinions. Furthermore, although initially reluctant during cross-examination to reassess opinions given earlier in writing, he became increasingly disposed to do so during a long cross-examination (including undertaking re-calculations during short adjournments) which displayed a commendable determination to assist the court.

The Turkish Law Experts

97. Both the Banks and Ferrero called senior retired lawyers with distinguished academic backgrounds. The Banks called Professor Unal Tekinalp, who until 2002 was the chairman of the Commercial Law Department of the Istanbul University Law Faculty. Ferrero called Professor Tugrul Ansay, who until his retirement in September 2008 had been the Dean of the Law School at Koc University, Istanbul. Both had academic careers stretching back for very many years, and both were prolific writers of articles and textbooks on Turkish law.
98. Neither spoke English as their first language, and none of the numerous academic texts or, for that matter, the small number of reported decisions of the Turkish Court of Cassation to which they made reference in support of their opinions, were originally written in English. Professor Tekinalp prepared an English language report, with assistance from colleagues in his law firm, but preferred to give evidence through interpreters. Professor Ansay had a better command of English and, no doubt to assist and save time, did not require the assistance of an interpreter.
99. In the event, neither of those different approaches proved entirely satisfactory. Professor Tekinalp found it necessary from time to time to correct what he regarded as mistranslations of technical legal language by his lay interpreters (and had to change interpreters midway through his evidence though not, so far as I understood, due to any inadequacy on the part of the first). I felt that Professor Ansay's response to a fast moving cross-examination in what was neither his first nor his second language, without interpreters, suffered a little from his occasional misinterpretation of the English questions asked. Nonetheless both experts were of considerable assistance in dealing with occasional infelicities in the English translations of the Turkish and German language texts put to them in cross-examination.
100. Professor Tekinalp was taken to task for an alleged lack of independence. Having in paragraph 5.1 of his first report reminded himself that it was no part of his function to express any opinion on the correctness or otherwise of the factual allegations to which he needed to make occasional reference, he was accused of making a one-sided selection of facts relevant to the question whether Ferrero had, when purchasing hazelnuts from Baskan Gida, Aksu Gida and Baskan Yuksel exercised the degree of care required in order to avail itself of presumed good faith under Article 3 of the Turkish Civil Code, before concluding in the negative, in paragraph 7.1.26 and 27 in his first report.
101. More generally, Mr Strauss relied upon Professor Tekinalp's long career as a legal adviser to, and in due course board director of, a major Turkish bank, in support of the submission that Professor Tekinalp's independence was undermined by an inevitable sympathy for the Banks' position as pledgee of the hazelnuts.
102. No criticism was made of Professor Ansay's independence, but Mr Wardell did with some force point to his relative lack of specialisation in relation to the law of pledge, by comparison with that of Professor Tekinalp.
103. There was in my judgment some justification for both these competing contentions. Professor Tekinalp's career had undoubtedly involved him more frequently in matters relating to pledges than Professor Ansay. In particular, his involvement was not merely that of an academic, but of a practitioner, in advising upon and settling pledges, although neither he nor Professor Ansay had any experience of litigating disputes about pledges in court. Against that, Professor Tekinalp did approach the matter instinctively (and understandably) more from the banker's perspective of a secured lender than from the position of a purchaser of originally pledged goods from a person with apparent authority to sell them, which formed the starting point of Professor Ansay's analysis. Furthermore, Professor Tekinalp undoubtedly did fall into the error of making a one-sided selection of the facts relevant to the question of Ferrero's good faith, albeit that this did not in my judgment undermine anything more than that relatively brief and, in the event, non-critical part of his report.
104. More generally, I found that, despite his need for the use of interpreters, Professor Tekinalp was slightly the more articulate presenter of his thesis, but with the corresponding disadvantage that from time to time he erred towards dogmatism when under cross-examination. By contrast, Professor Ansay was initially less clear or forceful in the presentation of his opinion, but correspondingly more helpful in his analytical response to questions in cross-examination.
105. In the end, as will appear, I did not find there to be such a disparity in the respective qualities of these two distinguished experts as to lead or enable me to conclude that the opinions of the one were in all respects to be preferred over the opinions of the other. My conclusions as to the issues of Turkish law have necessarily turned upon the detailed examination of the extent to which the alternative opinions of each of them on any particular issue were better supported by the codal texts, and academic writings cited in support.

THE FACTS

106. Almost all the events about which there are disputes of fact occurred during 2001-2. In order to understand them in their context it is necessary to set the scene. For that purpose I have chosen to take, as it were, a snapshot as at the end of 2000. I have done so by reference to the latest practical date before the events which have led to this litigation began to distort the parties' relationships, which (leaving aside the Banks which had yet to come on to the scene) had until then been cordial and mutually profitable. Since so much of the Banks' case depends upon portraying the other parties' conduct as a suspicious and uncommercial departure from the established norm, it is necessary to describe the parties' normal relationships, in the period ending at the end of 2000, in considerable detail, so as to provide the backdrop of apparent normality against which the events in dispute can be appraised.

Background – The Period Ending in 2000

The Turkish Hazelnut Market

107. Turkey is the world's largest producer of hazelnuts, and they constitute one of the country's major exports. They are farmed mainly along the southern shores of the Black Sea, usually in modest sized orchards. The trees come into blossom in February and, subject to regional variations, harvesting takes place in August and September.
108. Generally speaking, hazelnuts are classified by reference to variety, quality, calibre and specification. Each of those types of classification needs some elaboration for present purposes.
109. Generally, a variety is identified by reference to place of origin. In Turkey, the variety producing the best quality comes from Giresun, and hazelnuts from all other locations are sometimes described as Levant. Within that broad category there are other relevant locations, in particular Ordu, Trabzon and Akcakoca. The present case is concerned mainly with the Giresun and Ordu varieties.
110. As for quality, the principal distinction derives from the use to which hazelnuts can be put, the higher quality being described as rocher, meaning that the nuts can be used in confectionary as whole kernels. The lower quality, described as paste, means that the nuts are suitable only for being ground into, and used as, paste. An intermediate quality is that of granulates, which calls for no detailed explanation. The quality boundary between rocher and paste is not impermeable. For example, a batch originally selected for use as rocher may, over time, deteriorate such that it is thereafter usable only as paste. Conversely, a batch sold as paste may be subjected to a process of re-selection, so as to yield a sub-batch of nuts suitable for use as rocher.
111. As for calibre, this is measured in millimetres, by reference to the diameter of the kernel. For present purposes, the main categories fall between 11 mm and 15 mm. The more exacting single calibre standard, for example 13-14 mm calls for hazelnuts with a diameter from 13.0 to less than 14.0 mm. The less stringent dual calibre permits batching of nuts for example between 11 to 13 mm and 13 to 15 mm.
112. Finally, classification by reference to specification enables hazelnuts to be bought and sold in batches conforming to precisely set standards in which maximum percentages are set for the presence of defective nuts, whether because of humidity levels, rot or mould, damaged kernels, cimiciate (a form of insect damage), insect infestation, twinned kernels, shrivelled nuts or the presence of fragments of shells imperfectly removed during the cracking process which I shall shortly describe.
113. During the relevant period, there was in force a Turkish standard known as TSE3075, issued by the Turkish Standards Institute in March 1978. A replacement (slightly more lenient) standard was introduced in March 2002, but it became common ground that this has no relevance to the issues which I have to decide, relating to the hazelnuts supplied to Ferrero during the 2000 and 2001 crop seasons.
114. Particular purchasers of hazelnuts could, and Ferrero routinely did, impose their own detailed specifications. In Ferrero's case these were usually more strict than TSE3075. In general therefore, Ferrero contracted to receive from its suppliers hazelnut consignments conforming to more rigorous specifications than those generally offered in the market, with the consequence that its suppliers and Baskan Gida in particular had to achieve those specifications either by sourcing higher than average specifications from their own suppliers, or by processing their supplies in such a way as to increase the quality of the specification delivered to Ferrero, by discarding sub-standard nuts.
115. The process whereby hazelnuts grown in a particular orchard in Turkey find their way into an export shipment to a customer such as Ferrero may be divided broadly into the following stages. The nuts first have to be dried, while still in shell. Nuts in shell may be stored for anything up to 3 years while retaining merchantable quality. Once shelled however, and unless frozen, they begin to deteriorate, such that, in kernel form, they have a shelf-life of not more than one year. The process of shell removal is known as cracking, and the general rule of thumb is that half the weight of the nuts in shell is produced by way of useable kernels. The shells themselves have a residual value, in particular for use in Turkey for domestic heating during the winter.
116. Once shelled, the kernels are calibrated, traditionally by being passed through large perforated drums. This is the first stage of what may loosely be called processing, the second stage of which consists of inspection, traditionally by a numerous female workforce picking out sub-standard nuts and other materials from the calibrated stream, on wooden tables in factory premises. During the period under review, more modern, less labour intensive techniques, including for example inspection by laser technology had begun to be introduced, more generally outside Turkey, and to a very limited extent within Turkey itself.
117. While undergoing the processes which I have briefly described, Turkish hazelnuts passed through the hands of a number of intermediaries, between the farmer and the foreign purchaser. These included manavs, crackers, brokers and exporters. In addition, a Turkish state-owned cooperative Fiskobirlik ("FKB") played an important role, essentially for the public policy objective of maintaining hazelnut prices for the benefit of farmers.
118. Manavs were local traders of hazelnuts in shell, buying direct from farmers. Crackers, as is implicit in their name, bought hazelnuts in shell from farmers and manavs and sold kernels to brokers or direct to exporters. There were about 180 crackers in Turkey at the material time, of widely varying sizes. Brokers, of which there were about 10 at the time, intermediated between manavs and crackers on the one hand and exporters on the other, in relation to the hazelnuts for export and, I assume, sold into the domestic market.
119. There were about 35 exporters, of which Baskan Gida was the largest. By exporter, I mean only that the company in question exported part of its hazelnut sales. Nor were exporters merely traders. Some of them, including Baskan Gida, had their own cracking and processing facilities, so that they could buy nuts in shell from manavs and brokers, as well as from FKB, which only bought and sold nuts in shell. It is possible, but in my judgment unlikely, that exporters also bought nuts direct from farmers. Since the expert evidence was to the effect that a typical farmer might produce no more than 1 or 2 mt of nuts per year, and the evidence showed that, at least in relation to Baskan Gida, nuts were purchased mainly by the 20 or 24 tonne truckload, it seems likely that purchases by exporters direct from farmers formed a negligible part of their hazelnut acquisitions.
120. FKB was a unique player in the Turkish hazelnut market. Its purpose, as I have described, was to support the market price by large scale purchases, usually early in the season, when an excess of supply from the harvest over demand threatened to reduce prices, and to sell later in the season, if possible, when demand began to outstrip supply. It sought to minimise the risk of deterioration of its substantial stocks by buying and selling in shell, and by storing the nuts in the meantime in temperature and humidity controlled warehouses. Unlike all other buyers and sellers in the market, FKB published its buying and selling prices. Not least because it stored all its stocks in shell, FKB had a reputation in the market for selling nuts of better than average quality. Nonetheless, anecdotal evidence was that, as might be expected of an organisation with a purpose other than the making of profits, FKB incurred very substantial losses in its market support activities, which had to be funded by the State, in the public interest, or at least in the interests of hazelnut farmers, who formed a substantial part of the electorate.
121. The Turkish hazelnut market was inherently volatile, due principally to the effect of a variable climate upon both the volume and quantity of each year's harvest, coupled with the difficulty for end-users in building up substantial buffer stocks caused by the limited shelf-life of the nuts, once cracked. Additional contributors to price volatility were, first, the susceptibility of the Black Sea producing areas to disruption by earthquake, and the substantial Turkish financial crisis, with consequential large devaluations of the Turkish lira and restricted availability of credit from Turkish banks at reasonable rates of interest, which seriously affected the market during 2001 and 2002. Since hazelnuts were, in general, exported for payment in foreign currencies, (usually dollars or euros), but acquired by exporters within Turkey in Turkish lira, wide fluctuations in the exchange rate could produce market distortions in which purchase price movements could appear quite different to an exporter than they did to a foreign buyer.
122. Price volatility occurred as much within as between crop years. In this respect, major contributors included currency fluctuations as I have described, the irregular interventions of FKB as buyer and seller, speculation (and in particular hoarding) by market participants at all levels between and including farmers and exporters, together with the differing requirements of end-users at different times of the season.
123. There was no bourse or exchange for hazelnuts within Turkey, and no published prices other than those of FKB. On any given day, prices for particular varieties, quantities and specifications had to be, and in practice could be, readily obtainable on the telephone, in particular from brokers. Both the hazelnut market experts who gave evidence to me had retained notes of daily prices offered by brokers in 2001 and 2002, from which it was possible to determine with some precision the domestic market prices prevailing at the relevant time.
124. The price payable to a broker, cracker, manav or to FKB for unprocessed hazelnuts (whether cracked or in shell) forms only a part of the costs to an exporter like Baskan Gida of making an export to a foreign purchaser like Ferrero. To its purchase cost had to be added an amount for processing the consignment to the calibre and specification required by the foreign purchaser, an amount for export costs (including tax) and an amount to cover the cost of financing the nuts between the date of payment to its supplier and the date of receipt of payment from its foreign buyer. If the exporter was to make a profit, it also had to factor in weight losses attributable to the rejection of substandard or sub-calibre nuts necessary during processing to achieve the foreign buyer's specification, less the anticipated proceeds of the sale of the rejected element to other buyers. Finally, the exporter had to build in a gross profit margin, sufficient to fund general overheads and leave a net profit for distribution to shareholders.
125. A major plank in the Banks' case is that Baskan Gida was, from the spring of 2001, frequently selling to Ferrero at prices bound to generate a loss to Baskan Gida, to Ferrero's knowledge, thereby revealing to Ferrero its grave and ever increasing financial difficulties, which Ferrero then dishonestly concealed when giving trade references for Baskan Gida to the Banks in July and December 2001. The question whether particular sales by Baskan Gida were loss-making to Baskan Gida was intensely debated between the hazelnut experts and in cross-examination of Ferrero's main witnesses. At the heart of the debate lay disagreements about processing, export, finance and overheads costs which Baskan Gida needed to add to its purchase costs in order to make a profit. I will address those issues at the points where they arise chronologically.

Baskan Gida

126. Baskan Gida ('gida' is Turkish for 'food') was the corporate vehicle for hazelnut trading and exporting by the Baskan family in Turkey which began in the 1940s, and which had become by 1991 both Turkey's and the world's single largest hazelnut exporter. By 2001 the family members mainly involved were two brothers, Ahmed and Cevat, and Cevat's sons Melih, Ismet and Resat, as well as their cousin Ahmed's son Fuat.
127. The Baskan's family hazelnut business appears to have originated as a manav, buying from growers and selling to crackers in the 1940s, and from there to cracking and acting as a broker on the domestic market, before moving into the export market in the 1980s. By the beginning of 2001 hazelnut exports had become overwhelmingly the largest part of Baskan Gida's business, and Ferrero its largest customer. Other well known customers included Storck, Haagen-Dazs, Nestlé, Kraft and Cadbury.
128. Although the Baskan hazelnut business originated in Giresun, and its headquarters and largest factory was located there throughout, it had by the beginning of 2001 also established smaller processing facilities at Hendek, Bulancak and Ordu, together with hazelnut storage facilities at Hendek, Bulancak and Cumayeri, in addition to its main warehouse facilities at Giresun. The issues in this case relate primarily to the facilities, and in particular the warehouses, at Giresun.
129. Baskan Gida's Giresun facility underwent a process of enlargement and modernisation which began in the late 1980s and continued through to the beginning of 2001. Its warehouses at Giresun were temperature and humidity controlled, and built in 1988 adjacent to its processing factory. Although with Ferrero's assistance and encouragement, Baskan Gida had embarked upon the acquisition of laser controlled calibration machinery from 1999, it still employed a large workforce, mainly consisting of women, in its processing factory, for example sifting out sub-standard nuts by visual inspection in the traditional manner. Baskan Gida's sales and administrative staff were also located at offices within its Giresun facility. It also operated a laboratory there for sampling and testing hazelnut batches.
130. An organogram prepared by PRI Financial Management ("PRI"), a firm of financial consultants acting for Baskan Gida in early 2001 shows Ahmet Baskan as chairman, Cevat Baskan as CEO, Fuat as chief financial officer, Resat Baskan as purchasing manager, and shows that the sales and marketing activities were headed by Melih Baskan with the assistance of a Mr Ishak Franko. Since neither Ahmet nor Cevat Baskan could speak English, the practical result was that in dealings between Baskan Gida and its non-Turkish customers and lenders, Melih and Fuat Baskan took the leading role. The only Baskan Gida employee who gave evidence before me was Ms Pay, who was described, somewhat grandly, in the 2001 management organogram as the Director of Logistics.
131. The evidence before me was unanimous to the effect that, until the beginning of 2001, the Baskan family were held in high esteem for both their business skills and their integrity, both in Turkey and among their foreign customers and business contacts.
132. There was circumstantial evidence that Baskan Gida had over the years built up supply chains, consisting of a small number of selected manavs and crackers, from which the Baskans purchased, on a regular basis, hazelnuts of types and quality better suited for sale to Ferrero than was generally available, thereby reducing their likely processing costs, for example by eliminating the need to re-calibrate some less demanding paste categories. This was Mr Casale's assumption, was echoed in the information about Baskan Gida prepared by its finance consultants PRI in 2001, and confirmed as his impression of Baskan Gida's market profile by Mr Haas-Rickertsen. The effect upon the economics of Baskan Gida's export business with Ferrero is virtually impossible to gauge, but this aspect of how Baskan Gida did business provides an important part of an understanding of Ferrero's motivation in cooperating with the Baskans to preserve their business in 2002.

Ferrero

133. Like Baskan Gida, the Ferrero group of companies was at the material time and had been since its inception a family owned business. It was founded by the Ferrero family in Alba in Northern Italy in 1946 as a confectionery manufacturer. Since then it has grown into an international group, operating 15 factories and employing over 19,000 staff worldwide, with a group turnover in the financial year 2006/2007 of €5.74 billion, to which its Italian operations contributed €2.16 billion. By the beginning of 2001 its main factories were in Italy, Germany and France, with smaller factories in, for example, the USA, Brazil and Australia. Its most famous and flagship product is Ferrero Rocher but its products Nutella, Kinder Surprise and Tic Tac are also well known internationally.
134. By the beginning of 2001 Ferrero's corporate structure was headed by a holding company based in Luxembourg, Ferrero International SA, with operating subsidiaries, including Ferrero Italy, Ferrero Germany and Ferrero France, in each of the countries where it did business. It had also established a research and development operation called Soremartec, headed by a Belgian subsidiary of Ferrero International, Soremartec SA, with an Italian subsidiary Soremartec Italia SRL. Ferrero had embarked upon the first stage of the centralisation of its worldwide purchasing activities, by the setting up of Ferrero Industrial, in 2000.
135. Ferrero's internal management structure was not, by the beginning of 2001, a precise reflection of its corporate structure. The group was internally divided, without particular regard to corporate or national boundaries, into three main units, namely Soremartec, the Business Unit and the Operations Unit. Soremartec was responsible for developing and setting the specifications and quality standards of Ferrero's products. For present purposes its importance lies in the fact that its task included laying down detailed rules about standards applicable to raw materials, including hazelnuts, which included setting dates in each crop season by which purchases had to be obtained from Turkey, and dates by when old season nuts had to be used in manufacture. The Business Unit was responsible for sales and marketing, while the Operations Unit was responsible for the manufacture of Ferrero's products. The purchasing (both of raw materials and packaging) came within the remit of the Operations Unit and, separately from Soremartec, Operations had its own responsibility for quality control.
136. In addition, Ferrero maintained an Administration and Management Services centre providing for administration of the group's activities worldwide. At the pinnacle of its management structure was Ferrero International SA's board of directors, together with an Executive Committee, which included the heads of each of the internal units which I have described, and also the chief executive officers of the group, namely Pietro and Giovanni Ferrero.
137. At the beginning of 2001 the General Director of the Operations Unit was a Mr Bologna, but he was replaced by Mr Do in June. At the same time he became chairman of Ferrero Industrial which, with its responsibility for coordinating all purchases worldwide, was part of the Operations Unit. By the beginning of 2001 Ferrero Industrial had two managing directors, Mr Rosa Brunet, responsible for raw materials purchases, and Mr Giovanni Poggio with responsibility for the purchase of packaging and technical equipment. Before his appointment to Ferrero Industrial in June 2000, Mr Rosa Brunet had previously been the Purchasing Director of Ferrero Italy, since 1986, with responsibility for all that company's purchasing requirements, including hazelnuts. Between 1985 and 1990, he had gained considerable experience of the hazelnut market in Turkey, in particular in connection with a joint venture company which Ferrero set up with 49% Turkish investors, called Balsu, for the purposes of hazelnut purchases for the group. The Balsu joint venture had been discontinued by 2000.
138. Mr Rosa Brunet was assisted by a number of Deputy Purchasing Managers, with responsibilities for particular raw materials. At the beginning of 2001, Mr Casale was the Deputy Purchasing Manager responsible for hazelnuts, cherries, almonds and alcohol. Like Mr Rosa Brunet he had, before the setting up of Ferrero Industrial, been involved in purchasing raw materials for Ferrero Italy, including hazelnuts, from about 1988. Neither Mr Rosa Brunet nor Mr Casale spoke Turkish, and they both appear to have used English for communication with Turkish hazelnut suppliers, including Baskan Gida, although Mr Casale also had a working knowledge of French, German and Spanish. Their use of English as a second language, both at the time, and in the witness box, has greatly simplified my assimilation of the documentary and oral evidence.

Ferrero Industrial's Hazelnut Purchasing Strategy

139. The thinking behind the formation of a central purchasing operation for the requirements of all Ferrero companies was the avoidance of internal competition between them in the same market places. The avoidance of that competition was achieved at the price of creating a single player in the market which was by the beginning of 2001 the largest single non-Turkish purchaser of hazelnuts in the world, the requirements of which were of themselves sufficient adversely to affect market prices. Ferrero's purchasing requirements were driven by the operational needs of its manufacturing plants throughout the world, which by no means, either individually or in the aggregate, matched the seasonal pattern of availability of hazelnuts, or in particular the periods when, early in the season, hazelnuts could be obtained at their freshest, before the passage of time while in storage reduced their quality.
140. Ferrero Industrial's natural response to the risk of forcing up the market by becoming an involuntary purchaser of large quantities of hazelnuts at any particular time, was to build up buffer stocks of hazelnuts, and for that purpose to use the group's own storage facilities, particularly within Europe, to ensure as far as possible that by careful temperature and humidity control, loss of quality while in store was kept to a minimum. By the beginning of 2001 Ferrero had also embarked upon the freezing of hazelnuts, but this was still on a small and experimental scale. Although some of its suppliers, including Baskan Gida, were developing their own temperature and humidity controlled storage facilities by 2001, Ferrero Industrial's routine preference was to obtain delivery of its buffer stocks into Europe as early as possible in each crop season, to the extent consistent with avoiding the creation of a seller's market, rather than relying upon late-season purchases of nuts which, perforce, would have been in storage in Turkey for a substantial period since being harvested. The risk of unacceptable loss of quality by late-season purchases was much greater in relation to rocher than it was in relation to paste, and it affected kernels much more than nuts in shell.
141. The importance of building up buffer stocks in excess of the operating companies' immediate manufacturing requirements was accentuated by two factors. The first was that the general success of Ferrero's business meant that its hazelnuts requirements were, and had for some time before the beginning of 2001 been, increasing year by year. The second was that, in order to avoid being at the mercy of suppliers, Ferrero planned to build up sufficient stocks of 'old season' stock to ensure that its manufacturing requirements could be met from those hazelnuts during the early part of the following season. In this respect there was something of a tug of war between, on the one hand, the purchasing department's desire to maintain buffer stocks to avoid being an involuntary buyer of new season crop and, on the other hand, the wish emanating from Soremartec and those responsible for quality control to minimise the use of old crop rather than new crop nuts.
142. Prior to the establishment of Ferrero Industrial, each Ferrero operating company had developed its own separate relationships with hazelnut suppliers. These relationships were, in effect, inherited by Ferrero Industrial, and thereafter developed for the purposes of its centralised purchasing strategy. The main policy elements affecting Ferrero's Industrial's relationship with its hazelnuts suppliers may be summarised as follows. First, excessive dependence upon one single supplier was, for obvious reasons, to be avoided. Thus during the period ending in 2000, Ferrero had three main Turkish suppliers, namely Baskan Gida, Oltan Gida and Yavuzlar, two Italian suppliers, Barbero and Stelliferi, and also obtained some supplies from smaller markets including the United States, Georgia and Azerbaijan.
143. Secondly, Ferrero Industrial continued the operating companies' earlier policy of seeking to establish long-term relationships with its main suppliers. Within those relationships, Ferrero sought to educate its suppliers as to its particular specification and quality requirements, to encourage them to adopt best modern practice in processing nuts and to avoid or minimise the risk of Ferrero being exploited by its suppliers' knowledge of its hazelnut requirements.
144. A major feature in Ferrero's relationships with its main Turkish hazelnut suppliers (principally Baskan Gida and Oltan Gida by the end of 2000) was the use of a cost-plus basis for agreeing prices, pursuant to which its suppliers would offer to sell to Ferrero at a price calculated by adding to the supplier's own cost of purchasing nuts, elements for processing, export and profit, at rates usually pre-agreed. I shall describe these arrangements in detail, as between Ferrero and Baskan Gida, in due course.

Ferrero's Trading Relationship with Baskan Gida Prior to 2001

145. Baskan Gida began supplying hazelnuts to Ferrero Germany in the mid-1980s, and to Ferrero Italy in 1991. By the mid-1990s it had become a major supplier to Ferrero Italy, and by the beginning of 2001 had become the largest supplier to the Ferrero Group, its aggregate supplies to Ferrero during the 1999 season having reached approximately 30,000mt, despite disruption caused to its processing factory at Cumayeri by an earthquake in August 1999 which killed one of the Baskan sons and a number of its factory workers. The same earthquake caused much greater devastation to Yavuzlar's factory, putting it largely out of operation as a significant supplier during that crop year.
146. By 2001, Baskan Gida's factories had been visited by numerous members of Ferrero's senior management, including Michele Ferrero, Mr Rosa Brunet and (much more frequently) Mr Casale. Furthermore, to assist in the coordination of deliveries, Ferrero had arranged for junior members of its purchasing staff to have a permanent presence at Baskan Gida's premises in Giresun. During the relevant period this function was carried out by Mr Altare and Mr Scavino, usually alternately. Furthermore, it had become habitual for members of the Baskan family to take Mr Casale on tours of selected hazelnut farms, while visiting Turkey, so as to enable him to observe the growing crop at first hand so as to gain a reliable impression of its likely size and quality, beyond that reported to him by the Baskans themselves. Nonetheless, language barriers prevented Mr Casale or any other members of Ferrero's senior management from having any significant discussions with persons active in Baskan Gida's own supply chain, or from forming meaningful business relationships with them. By the beginning of 2001 the day to day business relationship between Ferrero and Baskan Gida was conducted almost exclusively (in terms of anything other than routine administrative communications) between Mr Casale for Ferrero and Melih Baskan for Baskan Gida.
147. Both Mr Rosa Brunet and Mr Casale told me (and I accept) that by the beginning of 2000 Baskan Gida had established a solid reputation with Ferrero for quality, reliability and integrity as a hazelnut supplier. As to quality, its ability to satisfy Ferrero's exacting specification and quality standards had been sufficiently demonstrated to allow Ferrero without unacceptable risk to take delivery FOT at Baskan Gida's factories at Giresun and Cumayeri, without carrying out any pre-delivery quality inspection of its own, relying upon Baskan Gida's own laboratory testing of samples and effective self-certification of the specifications and quality of the nuts supplied.
148. As to reliability, Ferrero's management had been particularly impressed with Baskan Gida's ability to maintain deliveries and quality during and in the aftermath of the earthquake in 1999, without seeking to renegotiate delivery dates or prices. As to integrity, Ferrero had not experienced any attempt by Baskan Gida (by contrast for example with Yavuzlar) to exploit its knowledge of Ferrero's requirements, for example by building stocks of nuts when the market was low, with a view to supplying Ferrero at higher prices thereafter, a type of informed speculation to which Ferrero with its large requirements considered itself particularly vulnerable.
149. Particular Ferrero operating companies had prior to late 2000 felt sufficiently sure of Baskan Gida's reliability to make significant prepayments for hazelnut supplies without taking any security against the risk of default. The practice of making prepayments ceased however in 2000, after Ferrero Industrial's taking over all purchasing negotiations on a centralised basis, but there was no evidence that this reflected any reduction in Ferrero's perception of Baskan Gida's reliability. Rather, it appeared to have been a consequence of the Turkish financial crisis.
150. An essential plank in the Banks' case that Ferrero participated in Baskan Gida's fraud consisted of the allegation, pursued in enormous detail, that Ferrero's pattern of ordering hazelnuts from Baskan Gida changed during 2001 and 2002 from a previously well-settled commercial pattern in a manner that lacked any commercial justification, in a way which was therefore inherently suspicious. In order to address that central allegation, it is necessary for me to describe the pattern of dealings which had established itself between Ferrero Industrial and Baskan Gida by the end of 2000, in considerable detail.
151. At his office in Alba, Italy, Mr Casale had ready access to Ferrero's current budget for hazelnut procurement and pricing, and also to the anticipated requirements for hazelnuts of each of the Ferrero operating companies for which Ferrero Industrial made purchases, as well as their stock levels from time to time. He also had ready access on an informal basis to his boss Mr Rosa Brunet for guidance and assistance in deciding upon the amount to be purchased from each potential supplier. As the centralised purchasing department for hazelnuts, Ferrero Industrial was also charged with maintaining fairness as between the Ferrero operating companies, both in the allocation between them of relevant supplies and in the fixing of prices. Subject to those constraints, the amounts, qualities, specifications and prices of hazelnuts to be ordered on particular days from particular suppliers were within Mr Casale's discretion.
152. The process leading to the placing of a specific order with Baskan Gida began on the telephone, almost invariably between Mr Casale and Melih Baskan. In the high season (that is from August to December in any year) those conversations would take place at least weekly, frequently on a daily basis and often more than once a day. Mr Casale made a practice of avoiding disclosure even to a trusted supplier like Baskan Gida of Ferrero's annual requirements or even its immediate needs, not only to avoid Ferrero being taken advantage of by its immediate supplier, but also to guard against the risk of knowledge of Ferrero's requirements leaking into the general Turkish hazelnut market.
153. In addition to keeping Ferrero's annual requirements close to his chest, Mr Casale told me that it was not his practice simply to request that Baskan Gida (or any other supplier) supply a stated quantity of nuts. Again, he said that to disclose a specific Ferrero requirement, might turn the market against Ferrero. Rather, he said that his practice was to indicate on the telephone to Melih Baskan (and to competing suppliers like Oltan Gida) an interest in purchasing hazelnuts of a particular type at a particular price, frequently without specifying any particular quantity at that stage. Mr Casale's evidence suggests that discussion focused around a price in Turkish lira at which Baskan Gida would purchase the nuts, it being assumed that the price payable by Ferrero would be the aggregate of that base price, mark-ups at pre-agreed rates, and a conversion to US dollars or euros at then prevailing rates.
154. The next stage was for Baskan Gida to go into the Turkish market and seek to buy nuts at or around that price, until a sufficient quantity had been purchased to serve as the basis for a Ferrero order. Melih Baskan would send to Ferrero, for Mr Casale's attention, a series of Daily Purchase Reports ("DPRs") stating the amounts of hazelnuts purchased by Baskan Gida of the relevant type, and identifying the price payable and quoting average currency conversion rates for euros and US dollars for that day. The DPRs which I have studied appear to distinguish between Giresun and Ordu in terms of variety, and between nuts in shell and kernels. Mr Casale told me that he was usually informed on the telephone by Melih Baskan of the information recorded in the DPRs so that he did not need to study the DPRs as if they contained fresh information.
155. Once a sufficient tonnage of the types required by Ferrero had been purchased by Baskan Gida from its suppliers, (although not necessarily delivered), Mr Casale and Melih Baskan would then agree one or more specific orders, as between a particular Ferrero operating company and Baskan Gida. To assist in this process Baskan Gida routinely prepared and sent to Ferrero Industrial written Price Calculations ("PCs"). The first page of these usually summarised recent purchases by Baskan Gida by reference to a series of DPRs, in a manner designed to identify the average price paid for the aggregate tonnage in Turkish lira, together with the average conversion rates into euros and dollars during the period when the purchases had been made. To those average prices, in euros and/or dollars, Baskan Gida then added a series of mark ups at pre-agreed rates, differing for each type of variety and specification of nuts supplied, under the headings processing costs, export tax, export cost and profit, to produce a total price payable by Ferrero for each type, per metric ton. Occasionally, (and this applied only to a small minority of Ferrero's orders) a quotation for a CPT rather than FOT delivery would be made, including an amount for freight. The overwhelming bulk of Ferrero's orders were for delivery FOT.
156. The various mark-ups were pre-agreed in round dollar amounts and converted to euros in the PCs at then prevailing rates. By 'pre-agreed' I mean that mark-ups for particular varieties and specifications in successive PCs remained the same until, usually at the beginning of a season, they were re-negotiated. To take a random example, the mark-ups for Ordu nuts to Ferrero Italy's specification in October 2000 were US$120 for processing, US$100 for export tax, US$50 for export costs and US$140 for profit, with a discount of US$60 if the delivery was to be taken as paste, rather than the higher specification for rocher, since less processing costs would be incurred for paste. By way of comparison, the mark-ups in force on the same day for Giresun nuts at Ferrero Italy's specification were US$210 for processing, the same as Ordu for export tax, export costs and profit, with a discount of US$90 for paste as against rocher.
157. The ostensible purpose of negotiating and calculating prices between Ferrero and Baskan Gida in this way was that the price paid by Ferrero would always (unless specifically agreed to the contrary) bear a pre-ordained relationship with Baskan Gida's cost of acquisition of the same nuts in the Turkish market, purchased at substantially the same time, so as to exclude Baskan Gida from making a turn on the prices it negotiated with its supplier and the prices at which it sold to Ferrero based upon arm's length bargaining in each case. It was also ostensibly designed to prevent profits or losses being made or suffered by Baskan Gida due to market and exchange rate movements between the date of its purchase of nuts and the date of their on-sale to Ferrero.
158. Mr Casale and Mr Rosa Brunet both maintained in their evidence that while in general they trusted Baskan Gida honestly to state the prices at which it acquired hazelnuts before on-sale to Ferrero, there was nothing that Ferrero could do to check the accuracy of those prices or, therefore, to exclude the possibility that Baskan Gida was making additional profits beyond those reflected in the pre-agreed mark-ups, for example by purchasing when the market was lower than the prices stated, or to prevent Baskan Gida from making a profit by processing nuts at less than the stated processing mark-up. Ferrero's only means of satisfying itself that the prices quoted to it by Baskan Gida were realistic and fair was to compare them with the prices quoted by other suppliers, such as Oltan Gida, which at Ferrero's request used broadly similar methods of cost-plus calculation.
159. There was considerable inconsistency in the evidence as to whether the profit margin (US$140 per tonne in the examples quoted above) was intended to constitute a gross or net profit margin for Baskan Gida. The answer to that question turns on whether the amount for processing costs reflected an element for overheads, beyond these specific costs attributable to processing a metric tonne of that particular type and specification of hazelnut. In any event, the US$140 profit mark up had to accommodate Baskan Gida's cost of financing its acquisition of that parcel of nuts, between the date when it paid its suppliers and the date when Ferrero paid Baskan Gida, usually between one or two days after receipt of invoice, following delivery FOT.
160. Returning to the ordering process, following Mr Casale and Melih Baskan's agreement as to a specific delivery of nuts of a particular variety and specification to a particular Ferrero operating company, each party to the transaction then prepared contractual documents, in no set order of priority. Ferrero Industrial's practice was for Mr Casale to send, by email or fax, written instructions to the relevant Ferrero operating company to enable that company to place a purchase order with Baskan Gida. The instructions would contain all the necessary particulars to enable a purchase order to be completed and transmitted to Baskan Gida, including quantity, variety, specification, package size (usually 1 tonne big bags), method of delivery (FOT or CPT) and delivery period.
161. It was unnecessary for Mr Casale to send written instructions to Ferrero Italy, since its purchase orders were prepared in Mr Casale's office in Alba. By contrast, instructions for purchase orders to be made by Ferrero Germany were sent to Mr Bolowich at his office in Frankfurt. Similarly, instructions for purchase orders to be made by Ferrero France were sent to the appropriate employee at its factory in France.
162. Melih Baskan did not wait to receive a Ferrero purchase order before giving instructions for the preparation for Baskan Gida's own contractual documentation. No evidence was given of the precise process, but the inference that Melih Baskan did not wait for Ferrero's purchase order can safely be drawn from the fact that, in a minority of cases, but not infrequently, the Baskan form of contract reached Ferrero before the relevant operating company had sent, or in some cases, even been instructed to send, the relevant purchase order. Whereas all Ferrero purchase orders used the language of the place of incorporation of the relevant operating company, all Baskan Gida's contracts were in English. They contained substantially the same particulars in relation to each order as Ferrero's purchase orders, with one important exception, relating to delivery period. Whereas Ferrero's purchase orders usually identified a delivery period by reference to a month, (meaning, unless otherwise stated, delivery by the end of the stated month), Baskan Gida's contracts almost invariably simply stated, under the heading 'Shipment', the phrase "prompt from origin".
163. This discrepancy has given rise to a quite extraordinary amount of cross-examination and debate at trial, in particular in relation to the question whether deliveries made outside the period specified in Ferrero's purchase orders ought either as a matter of contract or commercial dealings between the parties to be regarded as late. The point matters because in the trade references which Ferrero Industrial gave the Banks for Baskan Gida in July and December 2001, Ferrero stated that it relied on Baskan Gida's ability to provide shipments "on a timely basis". As will appear, Baskan Gida was not by December 2001 routinely providing shipments within the delivery periods specified on Ferrero's purchase orders, such that, if those periods formed terms of the contracts for the supply of hazelnuts to which those purchase orders related, Baskan Gida was by December 2001 in breach of a large number of them.
164. The resolution of the question what the settled understanding or contract was between the Ferrero operating companies and Baskan Gida in relation to delivery times for hazelnut supplies, as at the end of 2000, is bedevilled by the very incomplete state of the evidence. The only person to give evidence involved in the making of the contracts (which in every case took place orally over the telephone) was Mr Casale. The only employee of Baskan Gida to give evidence, Mrs Pay, was not part of the sales staff, her duties being limited to the preparation of export documents and invoices. Baskan Gida has given no disclosure in the proceedings and it is clear that Ferrero's disclosure, and in particular the disclosure by Ferrero Industrial, has omitted a substantial part of the contemporaneous documentation, for reasons to which I will have to return. Subject to those difficulties, the following conclusions of fact relevant to this issue may be stated.
165. First, the discrepancy as to delivery dates or periods between Ferrero's and Baskan Gida's contractual documents relating to the same shipments was endemic by December 2000. Secondly, there is no evidence that this discrepancy gave rise to any difficulties at all between the parties by the end of 2000. Thirdly, most, but by no means all deliveries made during the 2000 season between August and December 2000 were made within the periods stated in the relevant Ferrero purchase orders. Nonetheless, over 2,000mt of hazelnuts were delivered outside those periods, without apparent complaint.
166. Fourthly, it does not appear that either party made much use of, or took much notice of, the other party's contractual documentation. Thus, Ferrero administered the transactions reflected in its purchase orders by recording the particulars contained in those orders on its computer systems, and using them for the purpose of monitoring deliveries and approving invoices for payment. Conversely, according to Mrs Pay, Baskan Gida recorded the details of its written contracts on its computer system, again for the purpose of preparing goods for delivery, obtaining customs clearance and sending out invoices. Baskan Gida's contracts were, when requested, countersigned and returned by Ferrero to Baskan Gida. If available to Ferrero Industrial at the time when it sent instructions to operating companies to prepare purchase orders, Baskan Gida's contracts were generally copied to the operating companies. Apart from that, they were simply filed away. It appears therefore that in the minds of those responsible for the management of the parties' respective obligations under the supply transactions, no one at Ferrero was particularly conscious that Baskan Gida purported uniformly to supply "prompt from origin" and no one at Baskan Gida appears to have been likely to have regarded delivery within the period stated in the Ferrero purchase orders as a contractually undertaken obligation, however desirable in order to satisfy an important customer.
167. Neither the documents nor the oral evidence provided any reliable guide to what Baskan Gida meant by "prompt from origin". None of the evidence (including the expert evidence) persuaded me that those English words had any settled meaning in the Turkish hazelnut market, or, perhaps less inherently improbably, in the international hazelnut market.
168. This is not a case in which the parties' silence about delivery dates leads to the implication that delivery was to be made within a reasonable time. In their contractual documents, they were both vociferous, but never ad idem. All three Ferrero Operating Defendants used standard terms and conditions which purported to make the delivery periods stated in their purchase orders terms of the ensuing contracts. Baskans Gida's standard form of contract was to the same effect in relation to the different delivery periods therein stated.
169. Mr Casale's evidence was that delivery periods or dates were never a matter about which he sought to obtain contractual certainty, provided that, as always happened, Baskan Gida maintained a track record of making deliveries in sufficient time for each of Ferrero's operating companies to avoid running out of stock. Nonetheless, whatever the position as a matter of strict contractual obligation (and I bear in mind that none of the hazelnut supply contracts were governed by English law) there was a mutual commercial expectation that, once ordered, hazelnuts would usually be delivered FOT in Turkey within about a month, save where the volume of orders was sufficient to overload Baskan Gida's processing capacity.
170. My reasons for this conclusion are as follows. First, the mutual understanding between the parties (whether or not true in fact) was that Baskan Gida had contracted to purchase sufficient hazelnuts of the appropriate type to satisfy a Ferrero order by the time it was made. This is the inevitable consequence of the process whereby orders were placed on the back of Baskan Gida's DPRs. There was no evidence that Baskans Gida purchased hazelnuts otherwise than for immediate delivery, but no evidence as to what, precisely, immediate delivery meant in Turkey. All that Baskan Gida needed to do therefore, as a matter of mutual understanding, was to take delivery of nuts from its suppliers, process and bag them and load them onto transport supplied by Ferrero.
171. Secondly, it was in the mutual interest of the parties that this process should take place as quickly as possible. First, Ferrero's interest was, as I have described, to arrange for hazelnuts to be stored in Europe in its own storage facilities, rather than to leave them in storage for long periods abroad. I have no reason to suppose that Baskan Gida was unaware of this. Its interest was to obtain payment for hazelnuts sold to Ferrero (usually within a day or two of invoicing, immediately following delivery FOT) as soon as possible, to minimise the period when it needed to finance the purchase of hazelnuts from its own suppliers. Ferrero had no commercial wish to delay delivery, for the purpose of postponing the date of payment, since throughout the relevant period, it appears that Ferrero's cash flow was amply sufficient to make payment for hazelnuts ordered, as soon as they became available.
172. There is a difficult issue as to Baskan Gida's processing capacity at the material time, which therefore impacts upon the question whether and to what extent the parties' expectations as to likely delivery periods were tempered by capacity restraints within Baskan Gida's factories. I have no doubt that Ferrero's status as Baskan Gida's largest and most important customer meant that available processing capacity was usually deployed to meeting Ferrero's orders as a first priority. Mr Abidali's evidence was that, both in terms of quality and reliability, Baskan Gida's performance of its obligations to less favoured customers was by no means as good. Mr Abidali's evidence was in that respect corroborated by surviving contemporaneous correspondence disclosed by Indo-Med's liquidator.
173. The parties have, with the assistance of their forensic experts, managed to agree comprehensive details of the daily and weekly volume of deliveries by Baskan Gida to Ferrero throughout the 2000 and 2001 crop seasons. It is tempting to conclude that Baskan Gida's capacity must have been as high as the highest rate disclosed by those figures. During the 2000 crop season, those figures suggest that Baskan Gida achieved weekly deliveries to Ferrero operating companies which exceeded 1,000mt on no less than six occasions, during September to November 2000. The highest was in excess of 2,500mt and there was one other delivery in excess of 2,300mt. During the 2001 crop season Baskan Gida achieved a weekly aggregate delivery to Ferrero companies exceeding 1,000mt on only one occasion, in September.
174. In my judgment these figures do not enable any reliable conclusions to be drawn as to Baskan Gida's processing capacity, for a number of reasons. First, the agreed delivery statistics are based upon incomplete records as to the precise dates of physical delivery, which are best evidenced by truck loading documents. Where those are absent, the agreed statistics adopt the invoice date. The trouble with invoice dates is that they may show misleading concentrations of deliveries, by reference to batches of invoices sent out by Baskan Gida's administration department, relating to actual deliveries spaced over a wider preceding period. Mrs Pay confirmed that this was so. She was asked in cross-examination to estimate Baskan Gida's daily loading capacity. She put it at about ten trucks (i.e. 240mt). This would support a weekly loading capacity of 1,440mt, assuming a six day week, but she was quick to point out that this was unlikely to be achieved day after day, and in any event depended upon the availability of nuts for loading, with their differing processing requirements. She also confirmed that invoices were not invariably dated or sent out on the day of delivery of the relevant goods, sometimes being completed days later. Although her evidence related to matters taking place years ago, it was about aspects of Baskan Gida's activities falling squarely within her field of responsibility.
175. The second reason (also reflected in Mrs Pay's evidence) is that the delivery statistics do not distinguish between different varieties, specifications and qualities of nuts, all of which needed differing levels of processing in order to convert incoming supplies at the Turkish standard to the different specifications required by the various Ferrero operating companies. There is therefore no set processing time per metric tonne or lorry load of hazelnuts capable of being identified, and therefore no unitary measure of processing capacity capable of being applied to any set of Ferrero purchase orders, for the purpose of ascertaining whether it overloaded Baskan Gida's processing capacity.
176. Thirdly, no attempted estimate of Baskan Gida's processing capacity would accommodate occasional interruptions, such as might be caused by the discovery that a particular batch of incoming hazelnuts fell short of expected quality standards, necessitating replacement or longer processing. Another potential source of interruption was Ramadan, during which the obligation to fast during daylight hours is said adversely to affect the productivity of the Turkish workforce.
177. A further reason for inability to define processing capacity arises from the fact that some deliveries were accumulated for the purpose of being put all on a single ship. This is likely to have led to a group of deliveries being concentrated within a particular short period, giving the misleading impression that the necessary processing all took place immediately prior to delivery.
178. All that can be said with any confidence is that Ferrero's purchase orders were more likely to stretch Baskan Gida's processing capacity early in the season, that is between August and December in any year, than at any time thereafter, both because this was the period of peak availability of hazelnuts following harvest and because it coincided with Ferrero's wish both to buy when the supply was high, so as to avoid the risk of increased prices, and to buy early for storage in Europe rather than late, after storage in Turkey. Departure from a commercial expectation of delivery within approximately one month was therefore inherently more likely as the result of capacity issues during that period, than at any other time in the year. This is specifically borne out by the agreed statistics for delivery periods actually achieved during the 2000 crop year, when there were a series of late deliveries outside the period specified in Ferrero purchase orders between July and October, and virtually none between November and March.
179. It is necessary briefly to describe the process within Ferrero by which delivery of hazelnuts by Baskan Gida led to payment of its invoices. The procedures differed slightly as between Ferrero Italy on the one hand and Ferrero Germany and France on the other, largely due to the fact that Ferrero Industrial and Ferrero Italy shared common offices. I only heard detailed evidence relating to the deliveries to Ferrero Italy and Ferrero Germany. It is a fair inference that the processes relating to Ferrero France were broadly the same as those described in relation to Ferrero Germany.
180. Taking Ferrero Italy first, the reliable evidence about this came mainly from Mrs Di Matteo, albeit that she only dealt with the last stage in the process. Baskan Gida would send invoices relating to Ferrero Italy, together with evidence of delivery, to Ferrero Industrial, where they would be checked and approved for payment by Mr Casale's team. I accept Mr Casale's evidence that he did not routinely carry out this process personally, although he and Mr Rosa Brunet both initialled numerous invoices to signify that they had been approved for payment by their staff. Mr Casale's evidence was that he or Mr Rosa Brunet sometimes initialled invoices after they had been passed to Ferrero Italy for payment, for example when they were both away from the office at the material time. It is sufficient for present purposes for me to say that, whenever initialled, invoices needed Mr Casale's department's approval before being passed to Ferrero Italy for payment. Mrs Di Matteo confirmed that the practice of having incoming invoices initialled by Ferrero Industrial (which she called the Purchasing Department) was used only for supplies where payment was due before receipt into Ferrero Italy's own warehouse or factory. It therefore applied to all Baskan Gida's deliveries.
181. The next stage was for invoices to be sent to Ferrero Italy's import department, charged with ensuring compliance with customs formalities. For this purpose they used the invoice and accompanying evidence of delivery, and noted relevant details on a computer database before passing invoices to Ferrero Italy's accounts department. With access to the same database, the accounts department could assume without further checking that delivery had been made (although of course the hazelnuts were still weeks away from arrival at Ferrero Italy's warehouse). The job of the accounts department was to ensure that payment for the delivered hazelnuts was due to Baskan Gida, and duly recorded as having been made. For this purpose the accounts department relied primarily on Ferrero Italy's purchase orders. Any particular order might generate a whole stream of invoices, as hazelnuts were delivered, lorry load by lorry load, to satisfy an order which might for a quantity in excess of 1,000mt. Checking invoices against purchase orders enabled the accounts department to satisfy itself first, that the nuts had been ordered, secondly that Baskan Gida was charging the correct price in its invoice, and thirdly that the nuts had not yet been paid for. Once approved for payment, the invoice was sent to the treasury department, whose only task was to ensure that payment was actually made, in accordance with any agreed timeframe, to the appropriate bank nominated by Baskan Gida. It was no part of the treasury department's function to decide whether an invoice was payable, or to check invoices against purchase orders, still less against Baskan Gida's form of contracts. Since payments were routinely made by bank transfer, it was the responsibility of the treasury department to ensure that both Ferrero Italy's bankers had facilities for direct payment to the bank nominated by Baskan Gida, and to ensure that instructions were given for payment to the correct bank.
182. For this purpose the treasury department maintained a computerised list of Ferrero Italy's suppliers, identifying their currently nominated bank. Where suppliers such as Baskan Gida frequently requested payment to a number of different banks, it was necessary for the treasury department to ensure that the necessary bank to bank links existed between Ferrero Italy's bank and each of those used by the supplier. Baskan Gida (uniquely in Mrs Di Matteo's recollection) supplied a list of banks for that purpose to the Ferrero Italy treasury department, at the beginning of the season. Thus it was necessary in relation to each incoming Baskan Gida invoice for Mrs Di Matteo's staff to check first whether it was the current bank identified for payment on the department's database, if not, secondly whether it was on Baskan Gida's list, and if not, thirdly to check with Ferrero's bank whether there existed a satisfactory payment arrangement between it and the new bank identified on the relevant invoice.
183. Fundamental to the treasury department's approach was, in the absence of any special reasons to do otherwise, a rule that payment had to be made to any supplier by paying the bank (if any) nominated by the supplier on the relevant invoice.
184. Turning to Ferrero Germany, Baskan Gida sent duplicate invoices both to Ferrero Industrial (at its offices in Alba) and to Ferrero Germany direct. It was Mr Bolowich's responsibility to approve payment of invoices to Ferrero Germany. He would ascertain that delivery had been made, either directly with Ferrero Industrial's employees in Giresun, or with Ferrero's transport contractors based in Istanbul. Having checked that payment was due, by comparison with Ferrero Germany's purchase order, Mr Bolowich would authorise payment by signing the invoice and passing it to Ferrero Germany's treasury department to arrange for timely payment to the appropriate bank. As I have said, in the absence of evidence to the contrary, I infer that the procedure relating to deliveries to Ferrero France was much the same as that adopted by Ferrero Germany.

Indo-Med and its Relationship with Baskan Gida Prior to 2001

185. Indo-Med was a much more modest operation than either Ferrero or Baskan Gida. It was set up in 1982 by Mr Abidali and his brother Saifuddin. Its precise beneficial ownership is an issue yet to be resolved, but it is clear that at the material time Indo-med was controlled by Mr Abidali, by virtue of an ultimate beneficial ownership of himself or his family.
186. Throughout the material period, Indo-Med carried on business as an international commodities trader, buying and selling hazelnuts and other materials on its own account, never taking physical delivery or holding stocks, but rather buying against orders from its customers for delivery direct by its suppliers to its customers.
187. Baskan Gida was Indo-Med's main supplier of hazelnuts, and the two companies enjoyed a trading relationship which went back to 1992, the essence of which was that Indo-Med purchased hazelnuts from Baskan Gida for delivery direct to its customers around the world, settling its contractual liabilities to Baskan Gida as purchaser on a running account between the two companies. Mr Abidali described the essence of the terms of trade between Indo-Med and Baskan Gida as being that Baskan Gida would supply Indo-Med with hazelnuts at cost, Indo-Med would on-sell them at a profit, and that Indo-Med and Baskan Gida would then share the profit between them equally. This relationship demanded a high degree of mutual trust, since Indo-Med had to rely upon Baskan Gida's honesty as to the cost of supplying hazelnuts to Indo-Med, and Baskan Gida had to rely upon Indo-Med's integrity in making a full disclosure of the relevant profit to be shared. So far as I can ascertain, neither party sought to carry out any due diligence checks upon the other's compliance with those obligations, which were conducted with a high degree of informality, profit shares being identified during the course of infrequent after-the-event reconciliation meetings between Melih Baskan and Mr Abidali.
188. Indo-Med got into serious financial difficulties due to defaults by its Russian customers arising out of the Russian financial crisis in 1998, following which Baskan Gida (no doubt because of Indo-Med's importance to it as an outlet for hazelnuts) provided substantial support to Indo-Med, by supporting Indo-Med's working capital needs by allowing Indo-Med's inter-company account to become heavily indebted. It is a reasonable inference that, thereafter, Mr Abidali felt personally indebted to the Baskan family, and to Melih Baskan in particular, for their assistance to him during Indo-Med's financial troubles. In short, I infer that Mr Abidali recognised at least a moral obligation to help the Baskan family, should they in turn incur financial difficulties in circumstances where he or his company had the ability to provide practical help.
189. An important feature of the case is the closeness of the relationship between Mr Abidali and Melih Baskan throughout the relevant period, a closeness which went beyond mere business cordiality, and persisted well beyond the date when Mr Abidali must have become aware of the Baskans' fraud upon the Banks, to the extent that, as late as 2003, Mr Abidali invited Melih Baskan to his son's wedding. This relationship was well established by the end of 2000.

The 2000 Hazelnut Crop

190. By the end of 2000, the parties were well embarked upon processing, selling, buying and trading in the Turkish hazelnut crop which was harvested during the previous summer. It was by all accounts a poor crop, yielding (according to official figures) approximately 500,000mt, as opposed to 640,000 in the previous year and 750,000 in the following year. The market was also adversely affected first, by the disruption to exporters' activities (and in particular their processing facilities) caused by the earthquake which I have described, and secondly by the serious Turkish financial crisis, which began in the second half of 2000, and which came to a head in early 2001. The principal adverse consequence of this economic crisis for exporters such as Baskan Gida was a large rise in the interest rates at which Turkish banks were prepared to lend to their Turkish customers. The effect of this on the cost to Baskan Gida of maintaining its working capital was exacerbated by the fact that in about November 2000 Ferrero decided to cease making pre-payments for hazelnuts deliveries by Turkish suppliers, a decision which Mr Casale attributed to a concern at the credit risk inherent in making pre-payments to companies in a country undergoing an economic crisis. He said, and I accept, that the decision to end pre-payments was not based upon any specific concerns about Baskan Gida itself, being applied even-handedly to all Ferrero's Turkish suppliers.
191. Notwithstanding these gathering difficulties, Baskan Gida managed by the end of December 2000 to deliver to the Ferrero operating companies slightly over 16,000mt of hazelnut kernels of all types from that season's new crop. Although a little over 2,000mt were delivered outside the delivery period specified in the Ferrero Purchase Orders, it does not appear that financial difficulties had detracted from the reliability of Baskan Gida's performance as Ferrero's largest hazelnut supplier by the end of that year, and the contrary has not been suggested. Business therefore continued to the beginning of 2001 without there being, or being any basis for, any apprehension on the part of Ferrero or Indo-Med that Baskan Gida's long and satisfactory track record as an honest and reliable supplier of high quality hazelnuts was being adversely affected, either by the poor harvest, or by the growing Turkish economic crisis.

2001

January to July

192. This period includes the first of the events about which there are serious issues of fact, and culminates with the first of the actions of Ferrero alleged to give rise to liability to the Banks, namely Ferrero Industrial's trade reference for Baskan Gida given, after meetings with representatives of the Banks in early July, in writing by Mr Casale's email on 31^st July ("the July Letter"). The unifying theme during this period consisted of sustained efforts by Baskan Gida to find alternative (or additional) sources of finance for its trading activities, and the response of those invited to assist, including Ferrero, the Banks and Mr Abidali. In short, by the end of July, Ferrero had declined to assist from its own resources, although it made limited efforts to find bank finance for Baskan Gida in Italy. Mr Abidali had found Mr Dahod, and brokered a financing agreement which had begun to bear fruit, while the Banks had gone no further than the early stages of a due diligence examination of a finance proposal.
193. The factual issues arising during this period may broadly be summarised as follows:

i) How grave had Baskan Gida's financial predicament become by the end of the 2000 crop season (i.e. July 2001)? Was it balance sheet insolvent, commercially insolvent (i.e. unable to pay its debts as they fell due) or even hopelessly insolvent (i.e. bound to fail in due course)?

ii) What was Ferrero's understanding of Baskan Gida's financial position by the time it provided its trade reference to the Banks? To what extent (if at all) did Ferrero's understanding of Baskan Gida's financial predicament undermine the truth, carefulness or honesty of that reference?

iii) What, precisely, took place at the meeting or meetings in early July between Mr Casale and representatives of the Banks? In particular, were any oral representations made, giving added meaning to the contents of the July Letter?

194. Before becoming immersed in the detail, it is important to bear in mind that the Banks do not allege that, as early as July, Baskan Gida or, a fortiori, Mr Abidali or Ferrero had embarked upon a conspiracy to defraud the Banks, in the sense of doing anything with that common objective, even though it is alleged that Ferrero's trade reference was dishonest.

The Turkish Economic Crisis and its Effect on Baskan Gida

195. Statistics helpfully prepared by the Banks' forensic expert and his team (and not in dispute) showed that an abrupt devaluation in February 2001 led to a fall, as against the US dollar, of 50% in the value of the Turkish lira by October of the same year, a rate of devaluation greatly in excess of Turkey's serious and escalating rate of inflation, measured by its consumer price index. For a company like Baskan Gida which exported 95% of its turnover for payment mainly in dollars or euros, but which paid for its labour and materials mainly in Turkish lira, this short term disparity between the rate of inflation and the rate of devaluation ought to have been advantageous since, whether looked at in terms of US dollars or Turkish lira, the value of its sales was likely to outstrip its cost of sales, until wage and price inflation cancelled or reversed that trend (as, in due course, they later did).
196. It appears however that two related factors more than cancelled out this apparent short term advantage. The first was that, (as the experts broadly agreed) Baskan Gida appears to have borrowed from its Turkish bankers the finance necessary to support both working capital, the cost of acquisition and improvement of fixed assets and some unexplained historic debts, in dollars rather than in lira. Secondly, at the heart of the Turkish economic crisis lay a lack of bank liquidity, which led to Baskan Gida's Turkish banks charging it ever increasing levels of interest, reaching (according to some reports), 20% or more. Faced therefore with the double whammy of borrowing in an appreciating foreign currency at exorbitant rates of interest from local banks, it is no surprise that Baskan Gida's response was to seek foreign sources of finance, to remedy a potentially disastrous predicament.
197. After an unsuccessful attempt at the beginning of the year to persuade Ferrero to resume pre-payments, the Baskans made a more formal attempt to obtain substantial long-term finance from Ferrero in March. By a letter to Mr Rosa Brunet and Mr Casale dated 26^th March 2001, Baskan Gida sought US$30 million of finance from Ferrero, of which it proposed to use $14.5 million for capital expenditure to improve its processing capacity. The invitation was supported by a detailed explanation why Baskan Gida's business with Ferrero was said to have become unprofitable, and was wrapped up within an offer to become Ferrero's exclusive supplier of Turkish hazelnuts, and to achieve repayment over 10 years by taking a reduced profit share. Alternatively, it was suggested that finance could be obtained from a bank of Ferrero's choice, with Ferrero's promise to buy 30,000 to 40,000 mt of hazelnuts per annum from Baskan Gida constituting that bank's main security for repayment.
198. Mr Casale and Mr Rosa Brunet's evidence was that they interpreted Baskan Gida's letter as an attempt to secure sole supplier status with Ferrero and that, since this conflicted with Ferrero's policy, they rejected it out of hand, after a brief discussion with Mr Nasi, Ferrero Industrial's then chairman. No surviving documents record Ferrero's reaction to the letter, but I accept that evidence.
199. Mr Wardell put it to both Mr Casale and Mr Rosa Brunet that, since Baskan Gida's letter allocated only US$14.5 million of the proposed US$30 million investment as being for capital expenditure, the rest must have been to repair a US$15 million hole in Baskan Gida's balance sheet, thus revealing that Baskan Gida was in grave financial difficulty. They both denied gaining any such impression. In particular they made it clear that they took Baskan Gida's assertion that its business with Ferrero was loss-making with a pinch of salt, pointing out that Ferrero was buying from Baskan Gida at broadly the same prices, with broadly the same built-in gross profit margin, as it was buying from Baskan Gida's competitors in Turkey, and that all Turkish exporters were anxious to do business with Ferrero on those terms. Further, Mr Casale pointed to the fact that, in his view, there had been no interruptions in the regularity and quality of Baskan Gida's deliveries, such as might be attributable to financial difficulties within Baskan Gida. As to this, the agreed statistics for deliveries in early 2001 showed only a single failure by Baskan Gida to deliver within the periods specified in Ferrero purchase orders for January to March inclusive (namely 6 out of 20 batches ordered by Ferrero Germany on 14^th February). Actual weekly deliveries showed a dip in February, and an unusual zero for the week beginning 5^th March, but a return to a healthy 400 mt plus per week for the 4 weeks beginning on 19^th March, which was a satisfactory level, having regard to the low volume of the 2000/01 crop.
200. In my judgment, while Baskan Gida's letter did disclose to Ferrero its desire to obtain substantial trade finance at lower interest rates than it was incurring in Turkey (then at the height of its financial crisis), Ferrero were both entitled to and in fact did treat Baskan Gida's tale of woe in its letter of 26^th March as special pleading designed to obtain exclusive supplier status, rather than as evidence of grave, still less terminal, financial difficulty.
201. Ferrero did nonetheless, during the next few months, make limited attempts to assist Baskan Gida in obtaining finance outside Turkey at lower rates of interest, in particular by pursuing inquiries of potential Italian lenders, and from Rabobank, but in the event none came to fruition, for reasons which I need not describe. In the meantime, Baskan Gida directed its attention to obtaining finance from other sources.

The FKB Contracts

202. The second event which the Banks allege put Ferrero on notice of Baskan Gida's grave financial difficulties consisted of the negotiation of a series of contracts for hazelnut supply by Baskan Gida to Ferrero, sourced from FKB. The Banks' case is that Ferrero must have appreciated that nothing short of the gravest financial difficulties could have persuaded Baskan Gida to contract on such gravely disadvantageous terms. Ferrero's case is that Baskan Gida gave an apparently plausible commercial reason for offering Ferrero substantial discounts off the ordinary cost plus pricing basis in relation to the FKB contracts, and that neither the making nor the performance by Baskan Gida of those contracts gave any indication that Baskan Gida was in financial difficulty.
203. The immediate background to the making of the FKB contracts was that, during a difficult crop season giving rise to a generally tight market, Baskan Gida had by mid-April 2001 delivered a little over 21,000mt of hazelnut kernels to Ferrero, but that Ferrero anticipated a substantial continuing requirement for 2000 crop season nuts, both for its manufacturing needs during the rest of that season, and to provide a buffer to avoid it being held to ransom at the beginning of the 2001 crop season. For its part, FKB had amassed substantial stocks for price maintenance purposes early in the season and, by April 2001, was likely to be the holder of the best quality stock, by reason of the combination of selecting good quality for storage, holding nuts in shell, and the use of humidity and temperature controlled warehouses. FKB announced that it would sell hazelnuts at US$1.70 per kilo (i.e. US$1,700 per mt) ex-warehouse. Converting to kernels at a 50% yield meant that FKB was offering at a kernels price of US$3,400, but of course the buyer would have to incur the cost of cracking.
204. On 20^th April 2001 Melih Baskan sent Mr Casale a PC designed to illustrate the US dollar and Euro prices at which, alternatively, Baskan Gida could supply Ferrero with kernels for paste sourced from FKB or from the market, which was then offering at a lower price.
205. His cost plus calculation for kernels sourced from FKB was (in US$ per mt) as follows:

Product Cost	3,400.00
Processing Costs	210.00
Export Tax	100.00
Export Costs	50.00
Profit	140.00 3,900.00

His cost plus calculation for kernels sourced from the market was as follows:

Product Cost	3,130.08
Processing Costs	120.00
Export Tax	100.00
Export Costs	50.00
Profit	140.00 3,540.08

206. To each calculation Melih Baskan then applied the following discounts. For the FKB sourced kernels he offered a US$120 deduction for paste (implying that the price calculation was primarily for rocher), a full US$140 deduction for profit and a full $50 deduction for export costs, producing a net price of US$3,590.00. For nuts sourced from the market he offered merely a US$90 deduction for paste, producing a net price of US$3,450.08.
207. Further PCs followed on the same day, all based on an assumed FKB source at US$3,400, producing a price for Ordu kernels for Ferrero Italy's specification FOT Giresun at US$3,590 and C&F Alba at US$3,695, and for FOT Giresun at Ferrero Germany and France's paste specification at US$3,540, reflecting a US$50 reduction for the lower specifications of Germany and France, as against Italy. The last of those PCs began with the following preface:

"Please find below the calculation for Ordu Quality for Italy, Germany and France. As we discussed, I would like to confirm that we agreed with FKB that if FKB reduces their price for 2000 crop before end of current crop we will get the difference and as I confirmed on phone we will get this difference back to you."

208. Thereafter, orders flowed from Ferrero Italy, Germany and France, at broadly (but not quite uniformly) those quoted prices, in April, May and June 2001 for an aggregate of slightly over 6,300mt. They constituted the entirety of Ferrero's remaining orders to Baskan Gida for the 2000 crop season. Subject to an issue whether the first of those orders (by Ferrero Italy for 120mt on 20^th April at US$3,490 per tonne) was intended to be sourced from FKB, it was common ground that Ferrero's expectation was that Baskan Gida's source of nuts for satisfying all those orders was to be FKB, although this was nowhere specified in any of the purchase orders or in Baskan Gida's contracts.
209. Leaving on one side for the moment Mr Casale's and Mr Rosa Brunet's evidence about these events, the natural inference from the first of the PCs which I have described, followed by PCs and orders based on FKB as the source for the nuts, is that Baskan Gida was offering Ferrero special discounts off the ordinary cost plus calculations then in force as an incentive to persuade Ferrero to take supplies sourced from FKB rather than the market, despite the fact that the market was on 20^th April 2001 offering nuts at a significantly lower price, and that Ferrero agreed to do so, even though this meant accepting a slightly higher price for FKB nuts than those sourced on the market, the difference being (in relation to Ferrero Italy) approximately US$140 per tonne. On the face of the documents, the discounts appeared to involve Baskan Gida in abandoning the whole of its US$140 profit margin, swallowing its own US$50 export costs and increasing by US$30 its rocher to paste reduction.
210. That bald analysis raises the obvious commercial questions: first, why Ferrero should have been prepared to buy FKB sourced nuts when presented with a cheaper quote by Baskan Gida for the same specification sourced from the market, and secondly, why Baskan Gida should have offered Ferrero to supply FKB sourced nuts at discounts which, on their face, not merely wiped out the whole of its gross profit margin, but incurred an apparent loss of US$80 per tonne.
211. Mr Casale's evidence (which Mr Rosa Brunet corroborated, as having been for the most part communicated to him by Mr Casale at the time) was as follows. With Mr Rosa Brunet's approval, he had communicated to Melih Baskan an indication (without being precise at to quantity) that Ferrero wished to acquire a few thousand mt of kernels during the residue of the 2000 season. Melih Baskan's response was that Baskan Gida was reluctant to make further substantial purchases in the market for fear of pushing up prices at a late stage in the season, and causing itself losses on unrelated orders from other customers against which (by implication) Baskan Gida had not covered itself, and in relation to which Mr Casale assumed that Baskan Gida anticipated making profits at current market prices. As an alternative, Melih Baskan suggested that Ferrero's requirements could be fulfilled by using supplies from FKB. Because those would cost more to Baskan Gida than supplies from the market, Melih Baskan offered to reduce the added cost to Ferrero of buying nuts sourced from FKB by the discounts which I have described. Mr Casale regarded it as implicit in Melih Baskan's proposal that, if Ferrero accepted it, it would not itself purchase hazelnuts in the market, and thereby raise prices in much the same way as if Baskan Gida had purchased them on Ferrero's behalf.
212. Mr Casale said that it did not occur to him at the time, whatever the implications from the discounts offered by Melih Baskan, that Baskan Gida would necessarily make a loss on sales sourced from FKB, on the terms proposed. He said that the negotiations took place on the telephone, and that when he later received the PCs confirming them, he did not study them with that degree of attention which would, he accepted, have disclosed a probable loss for Baskan Gida, on the face of the calculations. His main point was that it was Baskan Gida rather than Ferrero which had proposed those terms, and it never occurred to him that Baskan Gida would deliberately propose a series of large volume loss-making contracts. He said that, to the extent that the discounts suggested involved the abandonment of a previously agreed profit margin, this was expressly proposed by Melih Baskan on the basis that, by purchasing from FKB, Baskan Gida would preserve the profitability of contracts already made with other customers, by avoiding pushing up prices.
213. As for the rationale for Ferrero contracting to acquire nuts sourced from FKB at a higher price than the same specification sourced from the market, Mr Casale said that he and Mr Rosa Brunet in consultation with Mr Nasi, concluded that the small extra price was worth paying, because it avoided Ferrero imposing its substantial continuing requirements for 2000 crop nuts on a tight market, thereby encouraging speculators to buy and hoard nuts, with a view to holding Ferrero to ransom at much higher prices at the end of the season. In that regard Mr Casale prayed in aid a draft letter found on his laptop intended to be sent by Ferrero's Turkish lawyers to Yavuzlar, specifically describing Ferrero's disapprobation of exporters who speculated against Ferrero by buying and hoarding, as an illustration of that policy rationale.
214. The Banks attacked this account with the utmost vigour. In the forefront of their case was the allegation that, in fact, the FKB contracts were much more seriously loss-making than might be derived from a cursory review of the PCs which I have described. They supported this by Mr Solak's evidence which, after the rectification of some double counting, set out by reference to detailed calculations how, in his opinion, the FKB contracts must have lost Baskan Gida an aggregate of US$2.099 million. Every part of Mr Casale's account of the negotiation of those contracts was challenged in cross-examination. In particular it was put to him that Ferrero would never have continued to deal with Baskan Gida if Melih Baskan had, as Mr Casale described, revealed to him that Baskan Gida had taken speculative short positions in dealing with other customers. It was even suggested that Mr Casale's draft letter to Yavuzlar, which was undoubtedly disclosed late, was a recent fabrication, conjured up to support Ferrero's case.
215. For his part, Ferrero's expert Mr Haas-Rickertsen initially sought to justify an opinion that the terms of the FKB contracts left Baskan Gida room for profit, but it was based upon taking Melih Baskan's stated processing costs as being true. In a supplemental report, substituting his own opinion as to production costs, Mr Haas-Rickertsen reached the view, which he described in cross-examination as preferable to his earlier view, that the FKB contracts were, on his opinion as to associated costs, likely to lead to a loss for Baskan Gida in the sum of US$1.29 million, although he continued to maintain that it was possible (albeit unprovable) that Baskan Gida could have exploited opportunities significantly to reduce that loss. Both he and Mr Casale pointed to a range of possibilities available to Baskan Gida, such as purchasing from the market rather than FKB (and therefore at lower cost), purchasing at a discount from FKB, reduced processing costs attributable to the generally high quality of the FKB product, achieving a higher than 50% yield when cracking nuts in shell, and gains made on the receipt of whole warehouse deliveries where the actual volumes exceeded the registered quantities. By the end of his cross-examination, Mr Haas-Rickertsen had been constrained to concede that most of these opportunities, if exploited by Baskan Gida, would only have made marginal differences.
216. Forensic examination of the issues as to the commerciality or otherwise of the FKB contracts occupied a quite extraordinary amount of the parties' efforts and time during the trial. No stone was left unturned, and every aspect of the cost analysis of the acquisition and processing of FKB shells into Ferrero's specification kernels was wrangled over in the minutest detail. It would overburden this necessarily very long judgment for me to engage with, or determine, every aspect of that major battle. I bear in mind in particular that the relevance of the FKB contracts to the Banks' claims is not their commerciality per se but the impression as to Baskan Gida's financial position which the negotiation of them conveyed to Ferrero. Nonetheless, a reliable judgment as to that impression cannot be wholly divorced from the underlying reality, and a proper appreciation of Baskan Gida's financial position, and the motives underlying its business decisions, plays a part in a number of other issues which I must also decide.
217. I am in no doubt, that by April 2001, Baskan Gida was approaching, if not already struggling with, serious financial difficulties. Its own quarterly management accounts showed only wafer thin profits and an excessive exposure to short-term borrowing, together with a tiny net positive equity. Although its accounts for 30^th June 2001 would in due course continue to show a small surplus of shareholders' funds, its auditors later advised that, when a mistreatment of financial expenses was corrected, it had a net negative equity as at that date of the Turkish lira equivalent of US$7.5 million. Assuming that the auditors' approach presented a true and fair view, it was probably balance sheet insolvent by mid-April 2001. The evidence however does not show whether it was commercially insolvent, nor did financial difficulties prevent it from substantially discharging its delivery obligations to Ferrero within the period specified in its purchase orders.
218. I consider it likely that by 20^th April 2001, Baskan Gida was in a significant speculative short position vis-à-vis the Turkish hazelnut market. There is only slight documentary evidence of this, in the form of a document retained by Indo-Med and a note by Mr Solak recording Baskan Gida having offered to sell forward over this period. Melih Baskan's assertion to that effect to Mr Casale was corroborated by the concurrent views of both Mr Solak and Mr Haas-Rickertsen, to the effect that the Turkish hazelnut market opinion was that Baskan Gida was a notorious short speculator, generally taking a more bullish view of available supply than the rest of the market, and therefore shorting the market in the expectation of a late season fall in prices. The experts were also agreed that in April 2001 the market was thin, and that although prices were still below the FKB offered level, the impact of any substantial demand (such as the 6,000mt contemplated by Ferrero) would be bound to drive up prices, to the severe embarrassment of any short speculator. Whether that was, in April 2001, likely to strip Baskan Gida of an anticipated profit (as Mr Casale said he thought) or create or aggravate an impending loss, is impossible to ascertain without access to Baskan Gida's working records, and probably does not matter.
219. The expert evidence suggests a strong probability that the FKB contracts were in fact loss-making to Baskan Gida, and Melih Baskan must have understood this at the time. On their face, the PCs which I have described projected an US$80 loss per tonne. Applied to the aggregate tonnage under these contracts, this projects a loss a little in excess of US$500,000. I have already referred to the higher estimates of US$1.25 million (Haas-Rickertsen) and US$2.099 million (Solak). For present purposes it does not matter which of those three alternatives is correct. What does matter is the commercial rationale leading Baskan Gida to propose loss-making terms to Ferrero for a large amount of business.
220. In my judgment the motivation was as follows. First, at the heart of all Baskan Gida's attempts in 2001 to raise substantial further finance lay its ability to assert a high level of ongoing business with a blue chip customer, namely Ferrero. That this represented a key attraction to any potential lender is apparent from Baskan Gida's letter to Mr Casale and Mr Rosa Brunet in March, from the central structure of what became the Dahod loan agreement, and from its presentation shortly thereafter to lenders in the City of London, including the Banks. From the perspective of obtaining trade finance, its business with Ferrero was, quite simply, the jewel in Baskan Gida's crown.
221. Secondly, since Baskan Gida was short the market, if it could not find an off-market source to meet Ferrero's demand, then Ferrero would be likely to satisfy it (whether through Baskan Gida or other exporters) from the ordinary Turkish domestic market, thereby forcing up prices to Baskan Gida's grave disadvantage. In this context, the expert evidence powerfully suggests that what Melih Baskan told Mr Casale about Baskan Gida's short position was true.
222. Thirdly, and this emerged only at a very late stage in the evidence, it appears that, contrary to Mr Solak's opinion, FKB may have been prepared to sell nuts on deferred payment terms to reliable customers, charging an interest related premium for the credit period thereby provided. That this was a possibility emerged only when Mr Haas-Rickertsen acknowledged it under cross-examination. Until then it appeared to be common ground that FKB would require payment on delivery. For that reason, this possibility was never put to Mr Casale in cross-examination. If deferred payment terms were available to Baskan Gida from FKB, this would have been a powerful reason for persuading Ferrero to buy nuts sourced from FKB rather than from the market and, since the credit period afforded to Baskan Gida would provide it with a much needed boost to its cash flow, it may well have been worth incurring a loss on the transaction in order to obtain that advantage, as a stop gap, in advance of the obtaining substantial further trade finance.
223. Those being Baskan Gida's probable motives for negotiating the loss-making terms of the FKB contracts, the question remains whether the Banks have succeeded in showing that, contrary to Mr Casale's and Mr Rosa Brunet's evidence, Ferrero must thereby have appreciated that Baskan Gida was a company in terminal financial trouble, such as to falsify Mr Casale's July trade reference. In my judgment, their evidence was slightly undermined, but not nearly enough for the Banks' purposes. My conclusions on this question are as follows. First, Baskan Gida had not prior to 20^th April conducted itself in such a way as to disclose severe financial difficulties to Ferrero. It had performed satisfactorily as a supplier during a difficult crop season, and Ferrero could look back at a long satisfactory relationship with, apparently, the largest and most successful Turkish hazelnut exporter in the business. In particular, I find that Mr Casale had no reason to assume that Baskan Gida's short position in the hazelnut market was loss-making, rather than profitable by April 2001, or that Mr Casale appreciated that nuts might be obtainable by Baskan Gida from FKB on deferred payment terms, such a belief not being suggested to him in cross-examination.
224. I specifically reject the Banks' case that Ferrero had a general abhorrence of speculators, short and long. Mr Casale's draft letter to Yavuzlar clearly corroborates his evidence that Ferrero's particular aversion was to hoarders (i.e. long speculators), so that the revelation by Melih Baskan to him that Baskan Gida was short the market in April 2001 would not of itself have inclined Ferrero to seek alternative suppliers. I reject the suggestion that the draft letter was a self-serving fabrication.
225. The one respect in which, in my judgment, Mr Casale's evidence was less than wholly persuasive on these issues lay in his attempts to justify, after the event, a conclusion that Baskan Gida had room to make profits out of the terms of the FKB contracts. In fairness, this part of Mr Casale's evidence appears largely to have consisted of speculation after the event, his real recollection being that he did not give the profitability or otherwise of the FKB contracts to Baskan Gida much active thought at the time. In my judgment both he and Mr Rosa Brunet probably realised that Baskan Gida was prepared to take a loss on the FKB contracts to protect its short position from becoming an embarrassment due to the effect on the market of Ferrero's requirements, and they probably also believed, by parity of reasoning with the thrust of Baskan Gida's letter in March, that its policy was to retain as much as possible of Ferrero's business, even if that meant, from time to time, eliminating profit or even taking a modest loss. Nonetheless, I have no reason to suppose that Mr Casale or Mr Rosa Brunet had the slightest idea that the terms of the FKB contracts might expose Baskan Gida to losses of the magnitude suggested either by Mr Solak or (upon later reflection) by Mr Haas-Rickertsen. I can well envisage why a bulk purchaser dealing with suppliers on cost plus terms would find it difficult to believe that its suppliers deliberately understated their anticipated costs, as the expert evidence suggests that Melih Baskan did in the April PCs.
226. Little more needs to be said about Ferrero's perceptions of Baskan Gida's financial position prior to the giving of the July trade references. Baskan Gida satisfied almost all Ferrero's continuing requirements for 2000 crop hazelnuts, apart from two orders from Oltan Gida, which although at first sight apparently contrary to the understanding that Ferrero would not go into the market while buying FKB sourced nuts from Baskan Gida, may have been orders which Oltan itself sourced from FKB. Although Baskan Gida's compliance with the delivery periods in Ferrero's purchase orders during this period was not quite as good as it had been during the period from January to April, there were no significant delivery failures and the delays beyond those delivery periods do not appear to have been substantial, virtually all deliveries having been made within 44 days of the date of the relevant contracts.

The Dahod Agreement

227. Mr Abidali assisted the Baskans in finding finance for Baskan Gida's working capital in three distinct ways. First, he arranged for Indo-Med's bankers Habib Bank Zurich ("HBZ") to make available in May 2001 a discount finance facility to Baskan Gida of up to US$1 million, which he guaranteed. Under its terms, Indo-Med advanced money to Baskan Gida against sight of invoices from Baskan Gida to Indo-Med, together with documents confirming shipments of hazelnuts by Baskan Gida to Indo-Med's customers. Unlike Ferrero, which paid generally within two days of delivery FOT Giresun, Indo-Med's customers paid only after delivery to their premises in Europe and elsewhere, so that the HBZ facility provided a useful bridge, if limited in amount, between the production by Baskan Gida of the hazelnuts necessary to supply Indo-Med's customers, and payment by those customers to Indo-Med, and by Indo-Med to Baskan Gida.
228. Secondly, also in May, Mr Abidali assisted Melih Baskan at the early stages of Baskan's inquiries into the availability of international finance from London based banks, to replace borrowing at high interest from Turkish banks. This led to Mr Abidali collecting Melih Baskan and Mr Franko from Heathrow airport at the beginning of June, and driving them and a Mr Victor Politis of PRI to initial meetings at the Banks' London offices at the outset of the negotiation of the Facility. Mr Abidali described this assistance as nothing more than acting as "a chauffeur and gofer", and involving him in no analysis or understanding of the type of facility being negotiated. He was introduced to the Banks at meetings on 1^st June 2001 as Baskan Gida's London agent. In a very loose sense, since Indo-Med obtained customers for Baskan Gida's hazelnuts, and was based in London, this was true, but Mr Abidali was at pains to distance himself from any agency from Baskan Gida in connection in raising bank finance. There being no evidence to the contrary, save for his being described as Baskan Gida's agent, I am inclined to believe him.
229. Thirdly and most significantly, Mr Abidali was asked by Melih Baskan, shortly after the 1^st June meetings, whether Indo-Med or some other private (i.e. non-bank) source might be interested in providing finance for Baskan Gida, and in particular for that part of Baskan Gida's business consisting of sales to Ferrero.
230. Mr Abidali was and remains a member of the international Dawoodi Bohra religious community, a Muslim Shiite sect numbering about one million, consisting mainly of businessmen and professionals, based mainly in the Indian sub-continent, but with members in the Middle East, East Africa, Europe, North America and the Far East. Mr Abidali made known the opportunity to invest in Baskan Gida within his community, and was introduced in early June 2001 to Mr Ashraf Dahod (otherwise known and described in many documents as Sheikh Murtaza) as a potential investor. Mr Dahod had a highly successful career in the field of internet and other communication systems and software. He and his wife Shamim were resident in Massachusetts, USA and had also established a charitable foundation known as the Nooril-Iman Charitable Foundation Inc. ("NICF").
231. By the end of June that introduction had led to an agreement whereby Mr and Mrs Dahod and the NICF were to lend between US$10 to 15 million during the whole of the 2001 crop season for the sole purpose of financing Baskan Gida's sales of hazelnuts to Ferrero.
232. Although numerous aspects of the making, recording, performance and unwinding of this agreement are so extraordinary that they stretch the court's powers of suspension of disbelief to the limit, the making and essential terms of the Dahod Agreement have not been challenged by the Banks as an invention, and Ferrero's case in relation to them is broadly neutral. Nonetheless, since aspects of the performance and in particular the unwinding of the Dahod Agreement lie at the heart of the matters principally in issue between all the parties, it is necessary to describe the making and terms of it in some detail.
233. The first extraordinary aspect of this story is that during the three week period between Mr Abidali's first introduction to Mr Dahod, and the making of the Dahod Agreement (5^th to 25^th June 2001) there appear to have been virtually no inquiries or other forms of due diligence by Mr Dahod, before his commitment to the advance of very substantial sums by way of investment in a business with which he had no prior experience, run by people he had never met, in a country going through a financial crisis. Apart from the agreement itself, the only relevant documents which have survived are a draft proposal written by Mr Abidali dated 8^th June and circulated by fax and email to Mr Baskan and Mr Dahod, and an exchange of short emails showing that Mr Abidali met Mr Dahod in Boston shortly beforehand, during which he explained the proposal in more detail. In short, it appears that the Dahods simply took the merits of the investment on trust from Mr Abidali. Apart from the prospect of a very substantial return on capital, it appears that Mr Dahod simply placed implicit trust in the say-so of a fellow member of the community, and that he was prepared to accept Mr Abidali's reference for the Baskans, borne out of their many years' satisfactory mutual dealings.
234. The proposed terms circulated by Mr Abidali on 8^th June were structured as a joint venture project, the subject matter being the procurement on behalf of Ferrero and the supply to them of hazelnut kernels from Turkey, with a target of not less than 30,000mt during the 2001 crop year. The parties were described as (1) Baskan Gida (2) Sheikh Murtaza (or his nominee) and (3) Indo-Med. Their respective responsibilities were defined as (1) the purchase processing and shipping to Ferrero on a cost plus pricing structure of the hazelnuts "with full contractual responsibility" (2) the provision of cash resources to finance the purchase of the raw hazelnuts until shipping and invoicing to Ferrero, estimated at between US$10 to 15 million, and (3) the provision of "the intermediary infrastructure and banking channels to manage the financing and invoicing in a clear and open method".
235. Under the heading "formula of remuneration/risk" Baskan Gida's remuneration was described as coming from the built-in profitability of the processing costs charged to Ferrero under the existing agreed cost plus formula. Mr Dahod and Indo-Med's remuneration was to consist of sharing, in the proportion 60/40, the whole of the US$140 per mt profit margin agreed between Baskan Gida and Ferrero under the existing cost plus structure. The duration of the agreement was to be for the whole of the 2001 crop season, specifically from 11^th July 2001 to 10^th July 2002, with scope for agreed extension to one or more subsequent seasons. It was a condition of the proposed agreement that the funds should be used by Baskan Gida "solely for the Ferrero business and no other". The agreement was to be conducted pursuant to Islamic tenets of faith as defined by Sharia law, as recognised by and practised by the Dawoodi Bohra community, in accordance with the instructions of its leader the Dai-al-Mutlaq, who was to determine any relevant disputes, and was to remain subject to contract until agreed by all parties and made the subject of "permission and the blessings of the Dai-al-Mutlaq".
236. The Dahod Agreement itself consists of a handwritten mainly Arabic text, signed by Mr Abidali, Mr Dahod and Melih Baskan and purportedly dated 25^th June 2001. Mr Dahod's and Mr Abidali's evidence was that it was prepared by Mr Abidali, signed by the two of them at a dinner in Toronto on 25^th June, approved and blessed by the spiritual leader on the following day and signed at some unspecified later date, but before 19^th July 2001, by Melih Baskan.
237. A reading of the document in translation suggests that, far from constituting a definitive statement of the terms of the joint venture, it was in reality a form of application for approval and blessing for submission to the Dai-al-Mutlaq, setting out a general description of the joint venture, and containing specificity only in relation to the split of profits between Mr Dahod and Indo-Med, revised at the last moment from 60/40 to 70/30. Subject to that provision, Mr Abidali said that the terms of the joint venture were as set out in his proposal dated 8^th June 2001, with the following additional provisions. First, Indo-Med was to act in all respects as the Dahod family's agent in the performance of the agreement, responsible for receiving investment funds from the Dahods, and their transmission to Baskan Gida, for monitoring Baskan Gida's performance of its obligations and, most importantly, for receiving the proceeds of sale of hazelnuts, by payment direct from Ferrero, for deducting the agreed $140 per tonne profit share, and for remitting the balance by way of revolving finance back to Baskan Gida to fund the next tranche of hazelnut procurement and processing. In Mr Abidali's words in cross-examination, the Dahods were to be silent investors, trusting him and Indo-Med entirely for the management and protection of their investment.
238. The Dahod Agreement is, in its context, quite extraordinary in its commercial terms. It was described by Mr Abidali as funding designed to replace Baskan Gida's Turkish bank funding of its working capital, due to those banks' unacceptably high interest rates. Yet the return on capital contemplated by the Dahod Agreement represented a substantially heavier burden to Baskan Gida in terms of the cost of working capital than the interest rate then being charged by its Turkish banks. On a contemplated 30,000 tonne turnover of business with Ferrero, the Dahod Agreement promised, at a rate of US$140 per tonne, a combined profit share for the Dahods and Indo-Med of US$4.2 million, i.e. a 42% return on the minimum US$10 million capital promised, and a still very large 28% return if the full US$15 million was required. The manuscript Dahod Agreement itself contemplated that Mr Dahod and Indo-Med would share an annual profit in the region of US$5 million. In cross-examination, Mr Abidali could offer no explanation why, otherwise than because of extreme financial embarrassment, Baskan Gida should have offered him and Mr Dahod such a large return, but he unconvincingly denied any awareness that Baskan Gida was in grave financial difficulty by that time.
239. Secondly, although the first advances under the Dahod Agreement were (as is demonstrated by surviving bank statements) made in July 2001, Mr Abidali was plainly aware that Baskan Gida was pursuing its attempts to obtain trade finance from the Banks thereafter. Mr Abidali explained that his understanding was that the finance sought from the Banks was to run in parallel with that provided with under the Dahod Agreement, and to be used for the non-Ferrero business of Baskan Gida, in due course replacing the HBZ discount facility. Some corroboration of this perception is to be found in a letter from Mr Abidali to HBZ on 21^st June 2001, which refers to the Baskans being "in the process of finalising a major Trade Finance Package in the City of London to become operative during August this year, in time for the new season of Hazelnut exports. As such the use of this (i.e. the HBZ) facility is no longer appropriate." By that letter, Mr Abidali sought and obtained a continuation of the existing HBZ Facility only until the end of August 2001.
240. Much more difficult to understand, if the Dahod Agreement was made in the manner alleged, is Baskan Gida's complete failure to make any mention of it of any kind in its detailed negotiations with the Banks, culminating in the Facility agreed in December 2001. When it is borne in mind that the profitable business with Ferrero was, from start to finish in its negotiations with the Banks, described by Baskan Gida as the jewel in its crown, the failure to mention, at any time after 25^th June 2001, that this business was already subject to a funded joint venture under which the US$140 per metric tonne profit margin had been passed on in full to Mr Dahod and Indo-Med, is inexplicable unless either, the Dahod Agreement is an invention, or the Baskans were determined, from June onwards, to deceive the Banks, or perhaps the Dahods, or both.
241. Although the first stage of performance the Dahod Agreement, consisting of initial payments by the Dahods to Indo-Med, for onward transmission to Baskan Gida, occurred in July, it is convenient to deal with them in the next section of this judgment, covering the period August to December 2001.
242. I have thus far described the negotiation and making of the Dahod Agreement as presented by Mr Abidali and Mr Dahod in their oral evidence, and as corroborated by the two relevant surviving documents. On balance, and notwithstanding the extraordinary aspects both of its terms and of the way in which it was negotiated, I find that the Dahods did provide funding for Baskan Gida's Ferrero business on the basis of a US$140 per mt profit share, without significant due diligence and in the highly informal and trusting manner which I have described. As will appear, neither Baskan Gida nor Indo-Med performed their respective obligations under it. On the contrary, Mr Dahod's trust in each of them was thoroughly abused.

Baskan Gida Approaches the Banks

243. In addition to asking Mr Abidali and Ferrero for help in finding financing, Baskan Gida escalated the search in May 2001, by approaching a New York based firm of financial consultants, PRI, probably in early May. In turn PRI sought to identify potential London based lenders by using the London consultancy firm Quantum Trade Finance Limited ("Quantum").
244. PRI spent the latter part of May briefing themselves about Baskan Gida's business and preparing a "Frequently Asked Questions" sheet ("FAQ") and either PRI or Quantum prepared an indicative term sheet for the purpose of describing the requested facility to interested banks. PRI and Quantum also prepared a brief two page descriptive summary of Baskan Gida.
245. Initial contact with both KBC and BTM was made by Quantum during the last ten days in May, leading to meetings at KBC on 29^th May and 1^st June, and at BTM on 1^st June, attended by Melih Baskan, and (save for the first meeting with KBC) Mr Abidali and representatives from Quantum and PRI.
246. The thrust of the presentation to the Banks was that Baskan Gida was an old established business, had for 10 years been the world's largest hazelnut exporter, had annual sales worth approximately US$200 million, and spent US$130 to 140 million purchasing hazelnuts in Turkey for cash for processing and onward sale, to a glittering array of blue chip customers headed, as to 65 to 70% of its sales, by Ferrero, which bought on average 30,000mt of hazelnuts per annum from Baskan Gida, on a cost plus basis paying Baskan Gida a fixed US$120 premium over cost. Baskan Gida was described as having an enviable reputation for quality and reliability, as exporting 90% to 95% of its sales, and as avoiding speculation save for 10% of its turnover "where it takes positions in order to stay in the market and informed". Critically, Baskan Gida was described as having a turn-round period between purchase of raw hazelnuts and on-sale after processing of a mere five days. Its longstanding relationship with Ferrero, and Ferrero's commitment to purchase large quantities on a cost plus basis was prominently trumpeted, both in the FAQ and in the two page summary. At that early stage, the proposed facility was presented as designed to finance Ferrero's purchase of hazelnuts in the harvest season (August to November), and availability was sought from mid-August 2001 in a maximum revolving amount of US$50 million, with a 180 day facility period. It was not at that stage presented as Ferrero specific, being designed to finance supplies to all Baskan Gida's export customers. The proposal was presented as being capable of being undertaken by a single bank, or by two or more banks acting together. The principal security offered was to consist of a legal assignment of the receivables due from Baskan Gida's customers, together with a legal assignment of title to the hazelnuts while in Baskan Gida's warehouses.
247. Both BTM and KBC, at this stage entirely separately from each other, responded positively to these approaches. Both set up two man teams to handle the negotiation and due diligence required, consisting of a relationship manager and a credit analyst. For BTM the team consisted of Mr Byles and Mr Philbin (in the latter case only until October 2001). For KBC the team consisted of Mr van Broekhoven and Mr Lewis. The conduct of the Banks in connection with the due diligence, the negotiation, the drafting and the operation of the Facility has been criticised as negligent by Ferrero in terms which occupy 40 pages of the re-re-re-amended defence, and 112 pages (excluding exhibits) of their banking expert's first report. The determination of the contributory negligence issues requires each Bank's conduct to be separately viewed, in particular during the due diligence period before they began acting together. A full description of each of the Banks' conduct of the due diligence, negotiation, drafting and operation of the Facility would occupy a long judgment of its own. Nonetheless, because there is little dispute of fact about any of these matters, and because to a significant extent the parties' respective experts reached a large measure of agreement, it is unnecessary for me to set out the entire history of those matters in painstaking detail. For present purposes, I shall confine myself to a summary of the Banks' conduct of those matters, and descend to greater factual detail, to the extent necessary, only at the stage of analysis of the specific allegations of contributory negligence.
248. Taking KBC first, Mr van Broekhoven and Mr Lewis decided at an early stage to look at the proposal in the round, so as to obtain in principle support from their senior managers in the Global Trade Finance Group in London before conducting detailed due diligence. For this purpose they sent detailed questions to Quantum, investigated the available credit information about Baskan Gida's main customers, and prepared and sent to Quantum KBC's own draft term sheet, to serve as the basis for further discussion. They also obtained a bank reference on Baskan Gida from Yapi Kredi, one of Baskan Gida's Turkish bankers, before reporting to the London based Business Committee of the GTFG, ahead of its meeting on 29^th June, at which approval in principle was given, together with authority for the conduct of a full due diligence exercise.
249. Events followed a similar pattern at BTM, where Mr Byles and Mr Philbin pursued inquiries of Quantum, made contact with BTM's local representative officer in Istanbul, and obtained bank references from Yapi Kredi and Akbank. By the end of June Mr Byles had begun drafting a preliminary application to BTM's European Credit Division, and in the meantime circulated to his immediate managers in the European Business Division a note setting out his support for the proposal.

July 2001 – the Banks' Meeting with Mr Casale, and the Ferrero July Trade Reference

The July meetings

250. Preparations for visits by both the Banks to Baskan Gida's headquarters in Giresun were on foot by the end of June, the pressure coming mainly from PRI on behalf of Baskan Gida, which wished to have a facility in place by August. At this stage, each of the Banks was pursuing its inquiries separately, although by then aware of the other Bank's interest. The teams from both Banks were also aware, prior to travelling to Turkey, that they could expect an opportunity to meet a representative from Ferrero. In the event, Mr Casale was visiting Giresun on the 11^th to 12^th July, in connection with his review of the growing 2001 crop, and attended brief meetings with both Banks' teams, members of the Baskan family and representatives of PRI, probably on the afternoon of the 11^th July.
251. The Banks do not base their case in deceit (or for that matter in conspiracy) upon anything said or omitted by Mr Casale when he met their representatives in Giresun on that day. Rather, the Banks rely upon what was said at their meetings with Mr Casale as setting the context for the trade reference contained in Mr Casale's 31^st July Letter. The Banks also allege, and Ferrero denies, that the broad nature of the way in which, if the Facility went ahead, the Banks would be likely to rely on Ferrero's conduct, was explained to Mr Casale.
252. In cross-examination Mr Casale stood firm in his recollection that he received only about five minutes' advance notice that he was to be asked to meet representatives of the Banks, on a day otherwise devoted to routine business with the Baskans relating to the new crop season. Apart from his evidence, I would have regarded the probabilities as being against him being given such short notice, but on this point I am inclined to believe him. It follows that he did not go to the meeting armed either with detailed information with which to assist the Banks, or with an appreciation of the particular reliance which might be placed by the Banks on anything he said.
253. The best surviving evidence of what did happen at the meetings between Mr Casale and the Banks comes from notes prepared respectively by Mr Philbin for BTM and Mr Lewis for KBC. Mr Philbin's original manuscript notes made at the time are exactly contemporaneous, but scrappy and rather uninformative. By contrast, Mr Lewis's visit report was only made on 23^rd July, but was based on contemporaneous manuscript notes which he then threw away. A separate section of Mr Lewis's report is given over to his meeting with Mr Casale, and I accept his evidence in cross-examination that it is unlikely that he confused information coming from Ferrero with information coming from the Baskans.
254. Mr Casale could only remember one meeting with representatives of the Banks, and had a lasting recollection only of Mr Byles, triggered when he spoke to him again the following year. On balance, I find that there were two meetings, separately with each banking team, probably close together, and both short. Unsurprisingly, Mr Casale made no contemporaneous notes at the meeting, and did not pretend to have any very specific recollection of precisely what was said.
255. Having listened to a detailed cross-examination of Mr Casale, on the basis of Mr Lewis's note, I am satisfied that the note constitutes a reasonably accurate summary of what Mr Casale said, albeit that he may not have used precisely the language (such as "cost plus" basis) or been as precise with figures (such as the percentage of sales to Ferrero by Baskan denominated in Euros) as Mr Lewis's note suggests. I am also inclined to believe Mr Casale's denial that he confirmed that Baskan Gida "only" purchased hazelnuts on the back of Ferrero orders. That was certainly the normative basis upon which Mr Casale dealt with Melih Baskan, but I accept that Mr Casale would have been unlikely to assure the Banks categorically about there being no departure from that basis by Baskan Gida, the timing of its purchases being a matter outwith Mr Casale's knowledge.
256. I can therefore summarise what Mr Casale said at his meeting with KBC's representatives simply by transcribing the relevant passage of Mr Lewis's visit report, subject to the reservations mentioned above:

( Baskan has been Ferrero's biggest hazelnut supplier for the last 8 years.

( Ferrero operates a centralised purchasing system; Casale is the sole buyer in Turkey for Ferrero world-wide.

( As a manufacturer it is Ferrero's policy not to enter into financing arrangements with its suppliers, although Ferrero does pay on receipt of faxed documents from Baskan in order to assist its cash flow.

( Ferrero would have no problem in paying direct to KBC if instructed to do so by Baskan.

( Although Ferrero does not provide formal guarantees for its subsidiaries, it is the policy of the company to ensure all subsidiaries meet their obligations.

( Quantity and timing of purchases depend on price, crop, quality and market conditions, so it is impossible to predict these at this time.

( All purchase orders are verbal, followed by a faxed deal confirmation. Casale confirmed that there had never been any default under this mechanism, and that if a verbal contract were not fulfilled, the persons/company involved would not have a future in the industry.

( 90% of last year's sales to Ferrero were denominated in Euros.

( Casale confirmed that Baskan only purchases hazelnuts on the back of Ferrero orders.

( Pricing is determined on a "cost plus" basis, with a profit margin to Baskan of USD120- USD 140 per MT. Although it is obviously in Ferrero's interests to get the price down, Casale confirmed that Ferrero worked with Baskan to identify costs savings, rather than reduce Baskan's profit margin. It is in the interests of Ferrero to ensure Baskan remains a viable entity given the long and successful relationship that has evolved, and given that Ferrero relies on a consistent quality and reliable source of supply in order to ensure its final product is acceptable.

257. Mr Casale was also cross-examined on two passages in Mr Philbin's notes. The first was:

"Not pressuring suppliers to buy forward."

Mr Casale said that this might have been a slight mis-recording of his saying that Ferrero did not like dealing with suppliers who speculated by buying in advance of Ferrero orders. The second passage was:

"Ferrero orders honoured immediately. Occasionally, Baskan will slightly over-buy depending on fluctuating market conditions."

As to this, Mr Casale said that he kept an insufficiently close watch on the precise time-lag between orders and deliveries to be able to have made such a statement. He would have said that Ferrero had the capability to deliver quickly after orders were placed. As for the second part of the extracted quote, Mr Casale said that since contracts followed purchases by Baskan Gida recorded in the PCs, he thought it unlikely that he would have said precisely what Mr Philbin recorded.

258. As to these points, I consider it probable that there was some difference of emphasis between what Mr Casale was trying to say and what Mr Philbin recorded. They are not of great consequence. For reasons which will appear later, Mr Casale was throughout his evidence at pains to play down the extent to which he monitored the timing of Baskan Gida's deliveries, and I have concluded that he was probably better informed in that regard than he has been prepared to admit.
259. I turn to what the Banks allege they told Mr Casale at their meetings with him. It is common ground (at least after cross-examination) that Mr Casale understood that, if they decided to lend to Baskan Gida, the Banks would be placing considerable reliance upon the strength of the Baskan Gida/Ferrero relationship. Beyond that, I am not persuaded that the Banks were any more specific. It is alleged that Mr Byles told Mr Casale at his meeting with him that he would be asking for the gist of what Mr Casale had said to be put in writing. Mr Casale denied any recollection of being asked to do so at the meeting, but it is common ground that his 31^st July Letter was the result of a request, originating from BTM and passed on to him by Baskan Gida.
260. It was specifically put to Mr Casale, and denied by him, that the Banks told him that they intended to rely specifically upon what he said about the profit margin built in to the cost-plus basis upon which Ferrero and Baskan Gida normally traded. Of course, by July 2001, this had for some time been honoured in the breach, rather in the observance, since all contracts after 20^th April had excluded any such profit margin, because of the special circumstances of the FKB arrangement. Nonetheless no case has been pleaded that this was a relevant dishonest misrepresentation by Mr Casale made orally at his meetings with the Banks. Nor was it in fact untrue, as a description of the normal basis on which Baskan Gida and Ferrero did business, and in my judgment Mr Casale had no reason to suppose that the normal basis would not be resumed at the beginning of the 2001 crop season, as indeed it was, at least initially.
261. For completeness I should add that during their visit to Turkey, both banking teams were given a guided tour of the Giresun factory, and at meetings with the Baskans asked for and obtained substantial information about the manner in which Baskan Gida carried on business. In addition, KBC's team met the manager of the Giresun branch of Yapi Kredi (a Mr Kici) and, among other things, obtained confirmation that local bank interest rates for foreign currency lending to Turkish businesses had exceeded 20%. No criticism is made of the manner in which either of the banking teams conducted their Turkish due diligence, so that it need not be set out in any greater detail.

The July Letter

262. Due diligence by both banking teams continued during July, leading to a presentation by Mr Lewis and Mr van Broekhoven to KBC's Business Committee in London on 24^th July, at which the proposal in its then form was rejected due to lack of adequate security over the hazelnuts pending delivery. On 23^rd July Mr Byles and Mr Philbin submitted a full credit application to KBC's European Credit Division.
263. On 31^st July in the morning there was an exchange of emails between Melih Baskan and Mr Casale following an earlier telephone call between them, during which Mr Casale said that Melih Baskan had asked him to put into writing what he had told the Banks at his meetings with their teams earlier that month. Mr Casale said that he had suggested to Melih Baskan that he prepare an appropriate draft, having himself heard what Mr Casale had said. Such a draft duly followed, attached to Melih Baskan's email to Mr Casale, which identified BTM as the bank requiring a trade reference. The draft consisted of a letter to Mr Byles of BTM from Mr Casale, and by return email Mr Casale duly sent that letter over his typed signature, without amending the draft. It is the most important single document in the case and I set out its terms below in full:

"Dear Mr Byles

I am writing to confirm our discussion of FERRERO's experience with Baskan as a supplier. Our company has been working with Baskan Gida since 1991, and on a large-scale basis after initial testing since 1993. Hazelnuts are FERRERO's single most important source of raw materials, and Baskan has been for several years our largest single supplier of hazelnuts. We have a very close working relationship to coordinate the market strategy for this important commodity, maintaining communication on a daily basis during the harvest season.

As the Deputy Purchasing Manager for FERRERO, I am responsible for coordinating this relationship and can report no significant difficulties since we bagan [sic] cooperating with Baskan. Since the centralization of FERRERO's purchasing department in 1999, we have purchased over 30,000 tons of natural hazelnuts from Baskan annually. We expect these trends to continue.

In terms of quality, we rely on Baskan's ability to provide consistent shipments of the highest standard on a timely basis. In fact, out of 200,000 tons supplied since 1991, I am aware of only 4 or 5 truckloads amounting to a total of approximately 100 tons, that FERRERO did not accept under the standards presented by Baskan. These shipments were simply regraded and accepted as a less expensive category. We expect these demanding quality standards to continue on both sides of the relationship

Please let me know if I can provide any further information.

Regards,

Alessio Casale"

264. The 31^st July Letter was duly forwarded by Melih Baskan, via PRI, to Mr Philbin at BTM. Mr Lewis said that he subsequently read and relied on it, but could not recall how he had received it. Certainly no mention is made in the Credit Application which, shortly thereafter, Mr Lewis prepared in conjunction with Mr van Broekhoven, for KBC's Credit Committee in Brussels.
265. The Banks allege that the 31^st July Letter contained four dishonest misrepresentations by Mr Casale as follows:

i) that the Ferrero operating companies had experienced no significant difficulties with Baskan Gida;

ii) that Baskan Gida was able to (and did) provide consistent shipments to the Ferrero operating companies on a timely basis;

iii) that the Ferrero operating companies expected to continue purchasing up to 30,000mt of hazelnuts annually; and

iv) that contracts entered into between Ferrero and Baskan Gida for the supply of hazelnuts were on a proper commercial basis and, in particular, the prices paid by the Ferrero operating companies were fixed in accordance with the prices actually paid by Baskan Gida and included an agreed sum for Baskan Gida's profits.

266. Alternatively, the Banks claim that the representations made in the 31^st July Letter were continuing representations, and became untrue before the completion of the Facility on 19^th December 2001. For present purposes, I shall deal only with two issues of fact, namely whether the 31^st July Letter contained representations which were then (i.e. on 31^st July) untrue and, if so, whether they were known by Mr Casale to be untrue when he made them. It is important at the outset to bear in mind that in these three short paragraphs, Mr Casale was summarising his description of a trading relationship going back 10 years, and 8 years on a large scale.
267. Taking each alleged misrepresentation in turn, I have identified no basis upon which the first was untrue. Reliance was placed in the RAPOC upon Ferrero having pre-paid Baskan Gida in relation to certain orders in 2000, but nothing was done at trial to show that pre-payments were or flowed out of difficulties in the trading relationship, whether by cross-examination or otherwise. Reliance was placed upon the relatively few occasions upon which Baskan Gida's deliveries took place outside the delivery period specified in Ferrero's Purchase Orders, during the 2000 season. When it is borne in mind that this was a difficult season, in which Baskan Gida substantially succeeded in supplying Ferrero's requirements out of a small crop in a tight market, I consider that Mr Casale was entitled to say that Baskan Gida's performance during that year had been satisfactory, and unaccompanied by significant difficulties. I do not regard the making by Baskan Gida and rejection by Ferrero of Baskan Gida's funding proposal in March 2001 as amounting to a significant difficulty in the trading relationship, still less the negotiation of the FKB contracts, by which Ferrero obtained substantial supplies, late in a difficult season, at what Mr Casale was entitled to regard as satisfactory prices.
268. As for the second alleged misrepresentation (consistent shipments on a timely basis), again, and largely for the same reasons, I do not consider that this was otherwise than true, in relation to the only periods prior to the end of July 2001 or subjected at trial to any detailed scrutiny. An attempt was made to rely upon the absence of any deliveries during one week in March 2001, but that occurred at the very height of the Turkish financial crisis, there appear to have been no orders in respect of which deliveries were outstanding beyond the periods specified in the purchase orders, and deliveries at a satisfactory rate resumed immediately thereafter.
269. The greatest reliance was placed by the Banks on the third alleged misrepresentation (that Ferrero expected purchasing to continue at the rate of 30,000mt per annum). Mr Wardell submitted that this cannot in fact have been Ferrero's expectation by the end of July 2001, because of what he described as the abundant evidence of Baskan Gida's growing financial difficulties. To be precise, the pleaded allegation is that "Ferrero did not expect to continue purchasing up to 30,000mt of hazelnuts from Baskan Gida annually without substantial new funds being made available": (my underlining).
270. Again, I find that Ferrero did expect orders and deliveries of hazelnuts by Baskan Gida to continue at an approximate rate of 30,000mt per annum. That rate had by July 2001 been achieved, Ferrero's requirements were on an upward trend and, although it had determined to resist Baskan Gida's wish to become its sole supplier of Turkish hazelnuts, nothing in the evidence suggests that, by bringing on an alternative competing supplier (as Ferrero did with Oltan Gida in the 2001 crop year) Ferrero intended to cut Baskan Gida's level of supplies down. Furthermore, Mr Casale did by July 2001 expect, and with good reason, that the 2001 crop season would produce substantially greater volumes than had the 2000 season.
271. I have already rejected the Banks' primary case in this respect, namely that Ferrero and Mr Casale in particular were by July 2001 aware that Baskan Gida's financial difficulties had reached a level that made its continuation in business for another full year unlikely. Nothing happened in or about the performance of the FKB contracts that did or should have suggested to Mr Casale that financial difficulties would shortly impede Baskan Gida's ability to go on meeting Ferrero's demands. Furthermore, Mr Casale's expectation that a 30,000mt per annum trend would continue was not expressed to have been based upon an assumption that substantial further funding would not be made available to Baskan Gida. To Mr Casale's knowledge, Baskan Gida was actively pursuing the obtaining of such further funding with the Banks, and, so far as I can ascertain, there was no basis upon which Mr Casale either did, still less should have, assumed that any funding necessary for Baskan Gida to maintain its level of supplies would not in fact be obtained. In those circumstances, it seems to me that the pleaded premise for the untruthfulness of this alleged misrepresentation falls away.
272. As to the final alleged misrepresentation (as to the basis upon which contracts were entered into), Mr Strauss accepted that there was an implied representation that contracts with Baskan Gida were entered into on a proper commercial basis. Apart from that, the 31^st July Letter was silent as to the terms upon which Ferrero contracted with Baskan Gida. It seems to me impossible to read by implication into the 31^st July Letter what Mr Casale had told KBC at the meeting earlier that month about the cost plus basis upon which the parties contracted. It was at the meeting with KBC that Mr Casale had given this description, whereas the July letter was sought from him by Baskan Gida in response to a request from BTM, and it was written to BTM. More generally, even if Mr Casale had given a similar explanation at his meeting with BTM, where an intending lender seeks written confirmation of information originally supplied informally by a principal customer of the borrower, and obtains in writing something less than it believed had been said orally, then the natural inference is that the customer intends to commit itself only to that which is confirmed in writing, and not to any additional oral material that is not confirmed. Accordingly, although the FKB contracts constituted a significant recent departure from the basis of contracting described orally by Mr Casale at the July meetings, neither he nor Ferrero committed themselves to that description by way of trade reference, when asked to do so at the end of the month.

August to December 2001

273. This section covers the period leading up to and including the making of the Facility Agreement, the notification to Ferrero of the debt assignments which formed part of it, the making of the investment promised under the Dahod Agreement, and the first stage of the export of hazelnuts by Baskan Gida to Ferrero from the 2001 crop. It also includes the decision by the Baskan family to perpetrate a fraud on the Banks and the period when the Banks allege (but the Abidali defendants deny) that Mr Abidali and Indo-Med agreed to participate in that fraud.
274. As against Ferrero, the Banks allege that the Ferrero defendants also commenced participation in an unlawful means conspiracy against them during this period, by making the Framework Contract and Acknowledgments of Assignment falsely suggesting that it was business as usual with Baskan Gida, by repeating in December the trade reference in the July Letter knowing it to be untrue, by secretly changing the basis of their trading arrangements with Baskan Gida, by entering into a series of uncommercial contracts designed to assist Baskan Gida in obtaining by deception advances from the Banks under the Facility and by countersigning Baskan Gida contracts which falsely stated their dates and relevant delivery periods, knowing that the Baskans would use them to deceive the Banks.
275. It is also the period during which the Ferrero defendants say that the Banks committed most (but not all) of the acts and omissions alleged to constitute contributory negligence on their part but, for reasons already given, I shall defer until a later discrete section of this judgment any minute examination of the acts and omissions upon which those allegations are based, save where relevant to other issues, or necessary for the purposes of providing a coherent chronological account of relevant events.

The Dahod Investment

276. By the beginning of August 2001, the Dahod Investors had already transferred US$3.15 million to Indo-Med for onward transmission to Baskan Gida. US$100,000 (less bank charges) had been sent on 11^th July and US$3.05 million (again less bank charges) on 18^th July. Acknowledging receipt on that day, Mr Abidali told Mr Dahod by email that he was remitting the first tranches of the Dahod Investment to cover a contract relating to the "current crop supply" notwithstanding that the Dahod Agreement had been for finance in relation to the new (i.e 2001) crop.
277. In fact, Mr Abidali did not remit the whole of the US$3.15 million by transfer to Baskan Gida for the purposes of financing new business with Ferrero. He used US$100,000 to reduce Baskan Gida's overdraft with HBZ under the discount facility which Mr Abidali had arranged, and which Indo-Med had guaranteed. As I have described, it had nothing at all to do with Baskan Gida's business with Ferrero. The use of any part of the Dahod Investment in connection with Baskan Gida's non-Ferrero business was contrary to the Dahod Agreement, yet Mr Dahod was not told about the use actually made of the US$100,000 and, in an email which Mr Abidali sent to him on 26^th July, it was clearly implied that the whole of the US$3.15 million had been used in accordance with the Dahod Agreement. When taxed with this in cross-examination, Mr Abidali unconvincingly responded that, to him, the business funded by the HBZ discount agreement was Ferrero business. It clearly was not, as Mr Abidali well knew.
278. At Mr Abidali's request, the Dahod Investors provided a further US$2 million on 8^th August, which, like the earlier payments, Mr Abidali caused to be transferred onto Turkish bank accounts identified by Baskan Gida. Although responsible to the Dahod Investors for making sure that their money was properly deployed for the purposes of the joint venture (i.e. exclusively on new Ferrero business), Mr Dahod did nothing to ensure that the Dahod monies which he did transmit to Baskan Gida were either kept separate from Baskan Gida's general funds, or used for new Ferrero business. Mr Abidali lamely told me that he had received Melih Baskan's assurance to that effect, and that, naively, he then trusted him.
279. The next payment by Indo-Med of Dahod Investors' money to Baskan Gida occurred on 29^th August. The circumstances deserve close attention. At 3.15 p.m. on 28^th August Indo-Med received a faxed demand from HBZ stating that as a result of the discount facility the account was overdrawn to the tune of US$437,128.28 and requesting prompt payment by the end of the month. Mr Abidali forwarded it to Melih Baskan with the manuscript comment "ouch!".
280. Just before 6 p.m. Mr Abidali sent an urgent email to Mr Dahod, saying:

"The biz has got off to a great start, we are badly handicapped with funds at the moment with all the monies blocked. We need additional resources to avoid stumbling badly and the additional 5 million you expected last week will be just about enough to avoid this happening immediately. Please can you tell me when this will be coming and rush same."

Shortly thereafter, Mr Dahod replied that the money would arrive imminently, from NILP, and apologised for the delay.

281. US$5 million duly arrived in Indo-Med's HBZ account early on 29^th August and, after seeking Melih Baskan's instructions by email, Mr Abidali arranged for US$4.5 million to be converted into Euros, and for the Euros to be transmitted in three unequal amounts to different Baskan Gida accounts in Turkey. The remaining US$500,000 was used to satisfy HBZ's demand to clear the overdraft incurred on the guaranteed discount facility, thereby avoiding a call on Indo-Med's guarantee. By 9 o'clock on 29^th August, in a further urgent email, Mr Abidali had told Mr Dahod that the funds had arrived, and that he was "in the process of dispersing them to Turkey" saying nothing about the retention of US$500,000 in London for purposes wholly unconnected with the Ferrero business, and beneficial to Indo-Med. Adding insult to injury, he continued "thank you also for your trust in me". When faced with this further serious breach of the Dahod Agreement in cross-examination, and with an apparent blatant lie to Mr Dahod, he could only say that since the HBZ account was a Baskan Gida account, and Baskan Gida was in Turkey, he neither lied nor broke the Dahod Agreement. In reality, the effect of those transactions was that, by the end of August, Mr Abidali had, to Melih Baskan's knowledge, but unknown to Mr Dahod, extracted US$600,000 from payments made to him as a fiduciary, and used them for his own benefit, by reducing Baskan Gida's then indebtedness to Indo-Med, incurred in non-Ferrero business.
282. By the end of August, the application by Baskan Gida of the Dahod Investment should have led to the funding of significant old and new crop deliveries to Ferrero, and to payment by Ferrero which, in accordance with the Dahod Agreement, should have been made direct to Indo-Med, so that its and the Dahod Investors' profit shares could be deducted before the residue was recirculated. In fact, nothing of that kind occurred. In cross-examination Mr Abidali said that he had relied on Baskan Gida to set up that payment arrangement with Ferrero and that, apart from one payment, it never happened, because Ferrero declined to pay to Indo-Med a debt due on a contract made with Baskan Gida. There is some slight evidence of a request by Mr Abidali to Melih Baskan to put such an arrangement in place, but only in late September, and a request to that effect by Baskan Gida to Ferrero, but Mr Abidali acknowledged in cross-examination that the fact that this important part of the agreed payment structure was never put in place represented a miserable failure on his part. In an email to Mr Dahod on 19^th September, Mr Abidali merely said:

"Everything runs smoothly. There are some changes on how the funds are being rotated for ease of function which we can discuss when you are here."

283. Having regard to the large size of the Dahod Investment, the manner in which the business supposedly funded by it was reported back to the Dahod investors was extraordinarily inadequate. At the end of September, Mr Abidali provided a first report by email of the tonnage shipped to Ferrero under the Dahod Agreement, in the aggregate sum of 4820mt, based upon a purely oral report to him by Melih Baskan. In October, Baskan Gida started to prepare single page statements of tonnage supplied to Ferrero, broken down between the various Ferrero operating companies, which Mr Abidali passed on to Mr Dahod. The reports are silent as to the prices paid by Ferrero, or as to the specific use made of the Dahod Investment, but of course the US$140mt profit share could be simply calculated. By 22^nd October Baskan Gida had, via Indo-Med, reported to the Dahod Investors that a total of 8,592mt of processed kernels had been delivered to Ferrero, from which Mr Dahod could calculate that a profit share was due to him and Indo-Med in the sum of US$1,202,880. Not one cent of it had by then been paid, nor had the profit share by then been referred to in any surviving written communication between the parties, following the making of the Dahod Agreement in June. Mr Abidali's explanation for its non-payment by the end of October was that he assumed that it was being recirculated to finance ongoing Ferrero business and that, since Mr Dahod was not chasing for it, there was no need for him to do anything about it.
284. The Dahod Investors provided a further US$98,500 to Baskan Gida via Indo-Med on 27^th September and, despite the unsatisfactory lack of profit share payments or proper reports, a further US$3 million on 1^st November, bringing their total investment in Indo-Med's business to the aggregate of US$8,748,500 and €4,923,413 (crediting as invested the US$600,000 applied by Mr Abidali in reduction of Baskan Gida's debts to Indo-Med). Those very large sums had been invested on the basis of Mr Dahod's apparent simple trust in Mr Abidali, before he had met any member of the Baskan family, visited its premises or carried out any due diligence in relation to its affairs.

The Banks' Continuing Due Diligence

285. In total contrast to the Dahod Investors, the Banks continued during August, September and October with a leisurely but very detailed due diligence in relation to the proposed facility. From the end of the first week in August the two Banks began to work together as likely co-lenders. KBC's Credit Committee approved the transaction subject to conditions on 20^th August, and BTM's Credit Committee approval was given two days later. In both cases, approval was given subject to conditions and, inevitably, they were not identical. Getting from two different Credit Committee approvals to a fully contracted and operational joint lending facility still involved a large amount of work, in terms of further due diligence, the negotiation of detailed terms satisfying the outstanding conditions, and the reduction of the transaction into a complex series of written agreements.
286. At the risk of oversimplification, the essence of the structure contemplated by both Banks at the stage of Credit Committee approval was that loans made to Baskan Gida to finance the purchase of hazelnuts would be repaid by Baskan Gida's European based purchasers ("offtakers") by the assignment by Baskan Gida to the Banks of its contractual entitlement to payment by the offtakers. The Banks analysed their risk exposure between loan and repayment as falling into two distinct periods. The first lay between the making of each tranche of the loan and the delivery of the hazelnuts purchased therewith, after processing, to the offtakers. The second lay between the delivery date and the receipt of payment by the Banks from the offtakers.
287. As to the first period, the Banks' principal exposure arose from the risk that Baskan Gida might fail to perform its contractual obligations to process and deliver hazelnuts to the offtakers, generally described as "performance risk". As to the latter, the Banks' exposure lay in the risk that the Banks might not receive payment from the offtakers for hazelnuts delivered. That risk turned mainly on the creditworthiness of the offtakers, but also involved transactional issues, such as the enforceability and terms of Baskan Gida's contractual rights to payment, and their proper assignment.
288. The appraisal of the Baskan Gida performance risk necessitated an analysis of its track record as a quality supplier (as to which the Ferrero trade reference played a central part), and an analysis of the extent of the risk that Baskan Gida might cease to be able to carry on its business during the period of the Facility, which necessitated at least some investigation of its commercial solvency. Against the performance risk, the Banks proposed to take security in the form of a pledge over the nuts purchased with the advances, pending processing and on-sale to the offtakers. The Banks' principal security for due payment by the offtakers after delivery lay in obtaining a watertight assignment of Baskan Gida's contractual rights to payment. At the time of Credit Committee approval, it was contemplated that the Facility would finance sales by Baskan Gida not merely to Ferrero, but to other substantial European companies. In the event, for reasons which do not matter, the Facility was agreed in terms which in fact only financed Baskan Gida's business with Ferrero.
289. The putting into place of the structure which I have briefly summarised involved engagement with Ferrero in four respects. First, the Banks had to be (and in due course easily were) satisfied as to Ferrero's creditworthiness. Secondly the Banks needed to obtain Ferrero's description of its trading relationship with Baskan Gida, and a satisfactory trade reference. This had largely been dealt with by the end of July, as I have described, but the description of the detail of the trading relationship was at that stage only oral. Thirdly, the Banks needed to satisfy themselves as to the structure and terms of trade between Baskan Gida and Ferrero, so as to give substance to the contractual rights to payment which were to be assigned. This they sought to do by requesting Baskan Gida to invite Ferrero to sign a framework contract setting out those terms in writing. Fourthly, the Banks needed to be satisfied that Baskan Gida's assignment of its rights to payment for hazelnuts sold to Ferrero would lead, as a matter of contract and in practice, to Ferrero making payment to the Banks. This necessitated pre-agreement with Ferrero that this would occur, and a documentary trail demonstrating both assignment of the relevant rights and notification of the assignment to Ferrero as debtor.
290. Although the Banks and their solicitors took the lead in specifying and drafting the written terms upon which those aspects of Ferrero's engagement in the Facility structure would be documented, the Banks left the obtaining of Ferrero's engagement in those matters almost entirely to Baskan Gida, no doubt for fear of undermining the apparently satisfactory trading relationship upon which the viability of the Facility depended, by officious intermeddling.
291. The Banks' case is that in each of the respects in which Ferrero's engagement was sought in connection with the Facility, save only for the first, Ferrero acted dishonestly, in furtherance of Baskan Gida's plan to defraud the Banks. The essential backdrop to the appraisal of those allegations consists of two parts. The first consists of Baskan Gida's continuing day to day business relationship with Ferrero. The second consists of an attempt to trace, on very inadequate evidence, the development of the Baskans' attitude towards the preservation of the family's hazelnut business, from one which began with a search for finance to save Baskan Gida, and ended with a fraud on the Banks designed to enable the business to be extracted from Baskan Gida, and carried on through a phoenix company. Since the first of those two parts of the backdrop is relatively fully documented, and the underlying statistics agreed, I shall take it first, by describing the negotiation and sale of new crop hazelnuts by Baskan Gida to Ferrero from August until 15^th October 2001, the date when the Banks allege that, by signing the Framework Contract, Ferrero committed its next significant part in the alleged fraud.

The 2001 Crop – August to Mid-October 2001

292. Reliable statistics show, and the hazelnut market experts agreed, that the 2001 crop was a very large one, amounting to between 725,000 and 750,000mt of in-shell hazelnuts, by comparison with the 2000 crop of 500,000mt, and the two previous years' crops of about, or slightly less than, 650,000mt.
293. It proved however also to be a crop with significant quality problems, which may be summarised as a high proportion of the less marketable 9-11mm kernels, high levels of mould and aflatoxin in the Black Sea varieties and high levels of cimiciate in the Akcakoca variety.
294. The absence of serious frosts earlier in the year had led the market to anticipate a large crop, but the experts agreed (and surviving documents recording crop estimates showed) that its actual size was generally underestimated by the market until well into 2002. Similarly, the quality problems to which I have referred also took some considerable time to emerge, as a matter of general market awareness.
295. The experts agreed that the market also generally misread the likely policy of FKB during the season. Pressure from the IMF on the Turkish government to cease funding market intervention for price maintenance purposes by FKB led to an expectation that its activity in the 2001 season would be substantially reduced, by comparison with previous years. In fact, although its purchases as a percentage of the market fell slightly, due to the increased size of the crop, the volume of its purchases actually increased, from about 94,000mt to 125,000mt of in-shell nuts.
296. The price consequences arising from these features, coupled with the market's initial misunderstanding of some of them, may be summarised as follows. I do so by reference to the Turkish lira price for 100 kilos of 11-13mm kernels, produced to TS3075, available for purchase by exporters such as Baskan Gida in Turkey. Prices started at 2.7 million, falling slightly to 2.65 million for most of September and passing the 3 million mark on 10^th October. They then rose during the rest of October to 3.6 million, encouraged by a published FKB bid price, and then remained within a range of 3.3 to 3.6 million right through until early February, following which they traded within a tight range of 3.2 to 3.3 million until early April, followed by a very slight reduction through to mid-May, falling back to 3 million on 21^st May. Thereafter, the growing appreciation of an over-supplied market led to a sustained reduction to a range of 2.5 to 2.7 million through to July, followed by a very slight peak during the last two weeks of the season. Prices for 13-15 mm nuts followed a similar pattern, at slightly higher levels. Exchange rate fluctuations meant that US dollar and Euro prices for exported hazelnuts followed a significantly different pattern. It was also a period during which a temporary lag in the rate of Turkish consumer price inflation against the depreciation of the lira, caused by the sharp devaluation in February 2001, was eventually cancelled out and indeed slightly reversed by April 2002, due to the combination of further price inflation and a correction in the lira/US dollar exchange rate between October 2001 and February 2002.
297. Due to the benign growing conditions, deliveries of the new season crop began earlier than in the previous year. Mr Casale and Mr Rosa Brunet visited Baskan Gida's offices at Giresun on 20^th August, as part of a visit to Turkey which lasted from 19^th to 23^rd August. By this time, they told me in evidence that Ferrero Industrial had decided upon its purchasing policy for the new crop season. Their evidence on this was vigorously challenged, and I must therefore deal with this issue in some detail.
298. At the heart of the Banks' case against Ferrero in conspiracy, is the allegation that a substantial part of the hazelnut orders which Ferrero placed with Baskan Gida during the 2001 crop season were for hazelnuts which Ferrero neither needed nor wished to purchase, being in excess of its production requirements at the time. The allegation is that Ferrero sought dishonestly to assist Baskan Gida in its fraud on the Banks by placing orders for unwanted nuts which Baskan Gida then used to justify draw-down requests under the Facility. Ferrero's case by contrast is that the orders placed on Baskan Gida during the 2001 season were the product of a pre-determined policy based entirely on its own commercial interests, adjusted from time to time during the season as changing market circumstances and other unforeseen events required.
299. Mr Rosa Brunet and Mr Casale gave broadly consistent evidence as to the formulation and content of Ferrero's purchasing strategy for the 2001 season. It was confirmed in detail by the evidence of Mr Do, who took over as chairman of Ferrero Industrial and head of Ferrero's Operations Unit in June 2001. Its main elements may be summarised as follows. First, anticipating a plentiful crop and therefore reasonable prices, they planned to build up Ferrero's stocks of stored hazelnuts well beyond the budgeted production requirements of the operating companies. Secondly, they sought to achieve such an increase in stock levels in excess of production requirements by maintaining deliveries from Baskan Gida at broadly current levels, and by increasing deliveries from competing Turkish suppliers, in particular Oltan Gida. Thirdly, they intended, so far as market conditions and prices permitted it, to continue with their established policy of purchasing early rather than late in the season, for reasons for which I have already set out.
300. The commercial rationale for that policy, and in particular for the increase of stocks in excess of production requirements, was explained by Mr Do, Mr Rosa Brunet and Mr Casale as follows. First, they anticipated that, by contrast with the two previous years, the 2001 crop would be plentiful, with correspondingly reasonable prices. Secondly, they had in mind that at times during the previous year, tightness of supply had led to stocks being reduced below satisfactory levels, albeit that no interruption in confectionary production had actually occurred. Thirdly, they anticipated that Ferrero's general track record as an expanding business might well lead to hazelnut requirements in excess of the budgeted production figures. Fourthly, Ferrero had both the cash flow with which to finance the holding of increased stocks and recourse to good quality warehouse facilities in Europe for storage, as well as its developing programme for freezing hazelnuts for long term storage. Fifthly, they (and Mr Casale in particular) doubted the capacity of the hazelnut trees in Turkey to produce two bumper crops in succession, and thereby anticipated that the 2002 crop would probably be less plentiful than their expectations for the 2001 crop. Finally, they (and Mr Do in particular) wanted to ensure, by diversifying their sources of supply, that Ferrero would not become excessively beholden to a single Turkish supplier.
301. Mr Wardell for the Banks challenged this evidence as, in substance, a dishonest attempt to rationalise, after the event, orders from Baskan Gida which exceeded any commercially defensible requirements. Emphasis was placed in particular upon the absence of any description of such a policy in any surviving document, and I was asked to infer that the absence from disclosure of any significant quantity of documentary records of management's deliberations and decisions about hazelnut purchasing policy was probably the result of a deliberate suppression or destruction of documents by Ferrero in connection with this litigation. That inference was, of course, squarely put to Ferrero's witnesses, and equally vigorously denied, reliance being placed upon a management style of doing things orally at meetings without taking extensive attendance notes or minutes, arising from the originally family based nature of the Ferrero business.
302. Apart from the paucity of documents, the Banks had little ammunition with which to challenge Ferrero's evidence about its hazelnut purchasing policy for the 2001 season. On the face of it, both the policy itself and the reasons given for it by the Ferrero witnesses were perfectly rational, when assessed against market circumstances and perceptions which are not in dispute, and against the essential vulnerability of Ferrero as, increasingly, the world's largest scale involuntary buyer in a volatile market.
303. Quite separately, the likelihood that Ferrero had the policy which their witnesses described is reinforced by the undoubted fact that the acquisition of increased stocks in excess of production requirements and a diversification of their sources of supply was just what Ferrero did during the season which followed. As will appear, even when Baskan Gida went out of business in February 2002, and thereby offered Ferrero a golden opportunity to resile from any unperformed orders (on the Banks' case) in excess of its real requirements, Ferrero confirmed all those orders with Aksu Gida, and later with Baskan Yuksel, and took delivery of the majority of them, such shortfalls in delivery against orders by the end of the season being reasonably attributable to other factors.
304. Against those considerations, all of which militate in favour of the acceptance of Ferrero's evidence on this issue, the paucity of documents recording management decision-making about purchasing policy does appear strange to an English trained eye. Ferrero was, by 2001, a large, increasingly centralised, sophisticated and multi-national business, whatever its family roots. Nonetheless I consider it would be wrong to place too much reliance upon expectations of the degree of retained documentation generated by a comparable English business, in deciding whether a surprising absence of documents is sufficient, virtually on its own, to undermine otherwise consistent and credible testimony. In this instance, I am not persuaded on that account to disbelieve the Ferrero witnesses' description of the purchasing policy which they adopted for the 2001 crop.
305. Mr Casale and Mr Rosa Brunet took the opportunity of their 20^th August visit to Giresun to negotiate a reduction in the standard price uplifts for processing and export costs used during the previous season. There is no contemporaneous document recording this renegotiation, and Mr Casale could not clearly remember having done so, although he regarded it as probable. He drew attention in his witness statement to a surviving study made within Ferrero in June 2001 which suggested that Baskan Gida had been charging slightly higher mark-ups than Oltan Gida for equivalent products. A comparison between PCs for purchases of the 2000 and 2001 crops shows that such reductions must have been negotiated at about this time. Their precise amounts do not matter. Since mark-ups for processing and export costs were always identified in US dollars in Baskan Gida's PCs, and Turkish lira price inflation was still well behind its exchange depreciation with the US dollar, it is not surprising to find that Ferrero felt entitled to ask for, and Baskan Gida felt compelled to concede, reductions in its costs mark-ups. Mr Casale's evidence (which I accept) was that Oltan Gida conceded similar reductions in its mark-ups at the same time, and that other suppliers may also have done so.
306. It appears to have gone without saying at the 20^th August meeting that the basis of negotiation of supply contracts between Baskan Gida and Ferrero would continue to be that recorded in the DPRs and PCs which I have described as typical of the previous year's trading (prior to the FKB Contracts). Although Ferrero through Mr Casale was aware that Baskan Gida had not secured the bank facilities outside Turkey which it had been seeking (since inquiries as to Ferrero's creditworthiness were still being received at that time), he said that he had been informed in general terms that Baskan Gida had obtained funding from Mr Dahod, albeit that he was not told about the name of the lender, the terms upon which it was being provided, and in particular not told about Baskan Gida's profit sharing arrangement with Indo-Med and the Dahod Investors.
307. Mr Casale took the opportunity presented by the 20^th August visit to place Ferrero's first orders on Baskan Gida for the new crop. He placed six orders, split evenly between Ferrero Italy and Ferrero Germany, for an aggregate of 1,128mt, and mainly for rocher. The requested delivery periods in the Purchase Orders were, as to three by the end of August and as to the other three by the end of September. As usual, all Baskan Gida's corresponding contracts stated "prompt from origin". All were delivered by 15^th October, and all within the purchase order period, save for three lorry loads of a 240mt order by Ferrero Germany.
308. By the end of August Ferrero had placed seven further orders on Baskan Gida, again mainly for rocher, bringing its aggregate orders by the end of the month up to 4,248mt. By far the largest single order was a rocher order by Ferrero Italy of 1,704mt, to which Baskan Gida applied its contract number 5052, placed on 27^th August.
309. By the end of that month Ferrero had placed even larger orders on Oltan Gida, amounting in the aggregate to 7,444mt. When this was pointed out to Mr Casale in cross-examination, he replied that Ferrero Italy had deliberately concentrated its rocher orders on Baskan Gida, which had a better track record in the supply of that high quality product, and which required a greater level of processing, and Ferrero had therefore focused more of its higher volume paste requirements in orders on Oltan Gida. Ferrero Germany ordered rocher from Oltan Gida and Baskan Gida, but was less exacting in its requirements than Ferrero Italy. In fact, by the end of August, Ferrero Italy's orders for rocher from Baskan Gida were ahead of orders from Oltan Gida (2,520mt against 1,704mt) and this disparity continued thereafter, so that by the beginning of October, the aggregate rocher orders from Oltan Gida by Ferrero Italy were 6,000mt and 4,368mt respectively. During the same period, the overall hazelnut orders placed by Ferrero on Baskan Gida almost caught up those placed on Oltan Gida. By 16^th October, the aggregate orders placed by Ferrero on Baskan Gida were 16,008mt, several thousand mt ahead of the equivalent level by the same time in the previous year.
310. By comparison with the increased rate of Ferrero orders, Baskan Gida's rate of delivery to Ferrero at the beginning of the 2001 season fell short of the rate achieved the previous year, although the actual volume delivered by 15^th October, during a two week longer period to date, was greater by 1,400mt. Of greater significance, the amount ordered but not yet delivered as at 15^th October was 8,424mt. Of course, by no means all that amount were overdue, in the sense that the period specified in Ferrero's relevant purchase orders had passed by then. Nonetheless, some 3,240mt did fall into that category.
311. The Banks' case was that a failure on that scale by Baskan Gida to deliver tonnage within the period specified in Ferrero's Purchase Orders must have set alarm bells ringing at Ferrero, in particular about Baskan Gida's continuing financial stability, so as to make the suggested 'business as usual' representation in the Framework Contract which Mr Casale signed on 15^th October untrue to his knowledge. I shall in due course come to deal with the question whether any such representation is to be found in the Framework Contract, but I must first deal with Mr Casale's trenchant denial that, as at 15^th October, he was aware of, or suspected, any such thing.
312. Mr Casale's first point (in the order of logical analysis) was that the periods specified in Ferrero's Purchase Orders were not negotiated as contractual terms with which Ferrero expected strict compliance, but rather were inserted, almost on a conventional basis, by reference to the next month end, or, in relation to orders placed late in the month, the end of the following month. It will be apparent from my analysis of the way in which Ferrero and Baskan Gida traded before the 2001 season that there is much force in this point. The expression of delivery dates in Ferrero's purchase orders and Baskan Gida's contracts were hardly ever ad idem, and yet the constant disparity appeared to have caused no difficulties between the parties before the beginning of the relevant period. This is borne out by reference to the agreed analysis of deliveries against delivery periods for the 2000 crop, in relation to which the volume of late deliveries against orders placed up to and including 5^th October amounted in aggregate to more than 2,000mt.
313. Mr Casale's second point, fully corroborated by Mr Bolowich, was that, since Ferrero Industrial's policy was to build stocks for relatively long term storage rather than to provide raw materials for production on a 'just in time' basis, the precise dates of delivery by Baskan Gida, or by any other supplier, were not of great concern to him, as long as the operating companies' stocks continued to increase, as he satisfied himself that they did. He said that, following the placing of orders, he paid little detailed attention to the flow of deliveries, leaving the fine-tuning of deliveries, together with the making of the necessary transport arrangements, and notification of expected arrival to the operating companies, to his subordinates. Since Ferrero had by then ceased making pre-payments, and only paid upon delivery, and since cash flow management was not Mr Casale's responsibility or, apparently, a significant concern generally in relation to hazelnut procurement, this evidence is not inherently incredible, and Mr Bolowich's corroboration of the general thrust of it adds considerable weight. I have little doubt that the information systems within Ferrero Industrial were sufficient to enable Mr Casale to inform himself with little trouble about Baskan Gida and his other suppliers' respective performances against specified delivery periods. In fact, there is evidence that Oltan Gida took the trouble to trumpet its own performance in regular statements to Ferrero, although there is no evidence that Baskan Gida, or any other supplier, did the same.
314. In my judgment, Mr Casale did nonetheless rather over-egg the pudding in this respect. It seems to me, as a matter of common-sense, that in deciding from time to time with which of various competing suppliers to place Ferrero's next series of orders Mr Casale must, as an evidently thorough and competent businessman, have taken some account of the amounts ordered but yet to be delivered from each of them. While other considerations, such as price and quality in particular may well have played a predominant role in his choice, it seems to me unreal to suppose that he remained in blissful ignorance of the rate at which existing orders were being fulfilled by Baskan Gida.
315. There are two pieces of evidence which suggest, with uncertain force, that Ferrero was in fact dissatisfied with Baskan Gida's rate of deliveries by 15^th October. The first consists of an email from Mr Abidali to Mr Dahod on 10^th October, seeking further instalments of his promised finance, on the basis that:

"Melih is under pressure to increase the pace…(and)...needs additional funding"

Mr Abidali said that although he could not recall his conversation with Melih Baskan about this, the pressure could only have come from Ferrero. The second was Mrs Pay's evidence that Ferrero's on-site representatives (Messrs Altare and Scavino) were complaining about delivery delays.

316. Mr Abidali is hardly a reliable source, and Melih Baskan may well have felt pressurised by the sheer volume of Ferrero's orders, without any expression of dissatisfaction. Mrs Pay was a generally reliable witness, but her evidence is not specific as to date. It may have related to the period after 15^th October, and it was in any event undermined by her evidence about there being many days in October and November when no deliveries, or only one delivery were made, when the surviving records suggest otherwise.
317. Any comparison by Mr Casale of the absolute rate of deliveries as between Baskan Gida and Oltan Gida would certainly have revealed Baskan Gida in a poor light, suggesting that, albeit complicated by the different specifications being demanded of each of them, Baskan Gida's delivery capacity was falling well behind that of its main rival.
318. Nonetheless, assuming as I do that Mr Casale was probably better aware of the level of Baskan Gida's delivery performance by 15^th October, and generally, than he has been prepared to admit, I am still satisfied that he was telling the truth when asserting that, when he signed the Framework Contract, nothing in Baskan Gida's delivery achievements had by then caused him to believe, or even suspect, that Baskan Gida had reached a stage of terminal financial difficulty.
319. As to the underlying reality, there is evidence that by mid-October 2001, Baskan Gida's senior management had reached a stage where, short of desperate measures, including fraud, they did not anticipate being able to pull the company through its financial predicament. As will appear below, when I review the question when the Baskans did decide to embark upon a fraud on the Banks, it is probable that they were on the point of doing so by 15^th October. Nonetheless there is no evidence that they had at this early stage resolved to take anyone at Ferrero into their confidence in this respect. On the footing that sophisticated fraudsters generally communicate their dishonest plans only to those who need to know, I consider that the probabilities are very strongly against the Baskans having given any advance intimation of their intended fraud to Ferrero at this stage, and I conclude that they did not.
320. The Banks' only other basis for alleging that, as early as mid-October, Ferrero knew that the Baskans were up to no good, derives from their allegation that, by this time, Baskan Gida was defrauding its Turkish bankers, in the manner which must have come to the attention of Ferrero. By the time of closing speeches, the Banks had abandoned the allegation that Ferrero knew in 2001 that Baskan Gida was defrauding its Turkish bankers. I shall return to the underlying question whether Baskan Gida was in fact doing so.

The Framework Contract

321. Mr Casale's evidence (both in his witness statement and confirmed in cross-examination) was that the first he heard of what became the Framework Contract was when Melih Baskan told him on the telephone on 20^th September 2001 that he was going to send a document designed to describe the business relationship between Baskan Gida and Ferrero to third parties from whom Baskan Gida wished to obtain finance. In cross-examination, with some support from surviving documents, it was put to him that he had been involved in the process of the production of a framework contract much earlier. It was certainly an early requirement of the Banks, having been identified as a likely requirement at the outset of the negotiations. Furthermore, an email from Mr Stambler of PRI to Mr Byles on 28^th August described an early draft of it as having already been reviewed by Mr Casale. In a letter to Melih Baskan on 21^st September, Mr Politis reminded him that the draft agreement had been sent for the Baskans' review on 26^th June.
322. I am unable to treat PRI, and Mr Stambler in particular, as a reliable source of information (in particular information obtained from the Baskans), not least because in Mr Politis's letter of 21^st September he noted that, in negotiations with BTM, Mr Stambler had been "forced to lie to the banks" in relation to an important matter under negotiation. On any view, it is clear that significant pressure from Baskan Gida upon Ferrero in relation to what became the Framework Contract began in late September, leading to the conclusion of the Framework Contract on 15^th October.
323. The Banks' case is that in signing the Framework Contract, Mr Casale and Ferrero acted dishonestly, because the description of the trading relationship between Baskan Gida and Ferrero therein contained was, to his knowledge by that time, materially untrue. The Framework Contract is also of importance to the assignment issues, because it is the earliest English language document known to have been received by Mr Casale in which the word "assignment" is used, and Mr Casale's evidence was that, from start to finish, he never really understood what the word meant.
324. On 20^th September Melih Baskan sent an Italian language version of the then draft Framework Contract to Mr Casale, telling him:

"As mentioned we really need this frame work in order to close the file and to start to draw money. If you see any part or even for the whole frame work I can sign separately any letter which shows that there is not any obligations for Ferrero Group…"

Later that day Ismet Baskan sent Mr Casale an English language version of the then current draft. It was a relatively long-winded English-style ten page document (not including its appendices) providing for English law and London arbitration, and contemplated a commitment by Ferrero Italy for itself and all its subsidiaries, on the misconceived basis that Ferrero Italy was the parent of all the Ferrero operating companies. It was, unsurprisingly, rejected out of hand by Mr Casale, who on the telephone with Melih Baskan told him that he would be prepared for Ferrero Industrial to sign a brief statement setting out the basis of Ferrero's then current contractual arrangements with Baskan Gida, providing that it involved Ferrero in entering into no new obligations, or in altering the way it then did business with Baskan Gida. Mr Casale discussed the making of a framework contract with a Mr Giannotta, an accountant employed by Ferrero Italy (in particular about the choice of law clause) and more generally with Mr Rosa Brunet, but did not take legal advice about it.

325. Further negotiations on the telephone led to the submission by Melih Baskan of a much simpler four page draft on 9^th October, drawing attention in the accompanying email to clauses 8 and 9, which had been omitted from an earlier draft sent on 5^th October, and which were concerned with applicable law and assignment.
326. Mr Casale and Mr Rosa Brunet signed this version and returned it to Baskan Gida on 15^th October. No side letter of the type offered by Melih Baskan in his 20^th September email was asked for or given. The Framework Contract was expressed to be made between Baskan Gida and Ferrero Industrial, which was expressed to contract "for itself and on behalf of each of Ferrero Companies ("Ferrero")." Clauses 2 and 3 deserve quoting in full:

"2. QUANTITY

The quantity of the hazelnuts supplied by Baskan is the quality of the hazelnuts generally being supplied at the time and place of acceptance, according to the specifications in the contracts, which the hazelnuts are required to meet.

3. METHOD OF SUPPLY

3.1. The Daily Purchasing Report is sent to Ferrero on a daily basis. The report shows the amount and price of different quality hazelnuts that are purchased by Baskan. The average TL/$, TL/Euro, Euro/$ and $/Euro rates are determined in the report. A copy of the Daily Purchasing Report is attached.

3.2 On the average in every 2-3 days time, Ferrero informs Baskan about the amount and specification of hazelnuts each Ferrero Company demands. At the same time the price calculations are made and sent to Ferrero. Based on the price calculations the invoices are prepared. A copy of the Price calculation Sheet is attached.

3.3 Subject to terms of the contracts, Baskan delivers the goods to Ferrero on FOT and/or CPT terms at Baskan's facilities."

327. Clause 6 began as follows:

"6. PAYMENT

6.1.(i) Payment for each order are made by Ferrero to Baskan on the date after the delivery to Ferrero takes place against presentation of the following documents:

a) Inspection Certificate prepared by Baskan;

b) Delivery Certificate specifying:

i. Quality grade of merchandise

ii. Total quantity of delivery

iii. Truck license plate number

c) Euro 1 customs clearance form;

d) Invoice;

e) Confirmation from the transportation company for the receipt of the cargo.

(ii) Invoice is issued on FOT basis and/or CPT Europe."

Clause 6.2 was drafted so as to identify the Baskan Gida accounts into which the Ferrero operating companies were to make payment, but the details were not completed until February 2002. Clause 6.3 stated that payment would be made on the same working day as that upon which the documents identified in clause 6.1(i) reached Ferrero, or on the following working day if documents arrived at weekends or out of hours.

328. By clause 8 the governing law was identified as that of the country in which the Ferrero company signing any particular supply contract was located. Clause 9 provided as follows:

"9. ASSINGMENT [sic]

9.1. Either party can assign its rights under by way of security or otherwise, without having to obtain the approval of the other party. However, a party, which has assigned its rights, will still be responsible for performing its obligations under this Framework of General Terms and conditions.

9.2. Any such assignment will be effected by notice in writing from the assignor countersigned by the assignee to signify its acceptance of the obligations under the framework of General Terms and Conditions. Upon the making of any such assignment, the assignor will remain bound to perform or secure performance of the said obligations (as so accepted) by the assignee."

329. The Banks maintain that, as a description of the then current trading relationship between Ferrero and Baskan Gida, the Framework Contract contained untruths, to which Mr Casale and Mr Rosa Brunet by signing the document subscribed dishonestly. The first, in the last sentence of clause 2, was the statement that Ferrero expected to continue to purchase hazelnuts from Baskan Gida at an annual rate in excess of 30,000mt. Reliance was placed upon the fact that, at that stage, orders to and deliveries from Oltan Gida had outstripped those to and from Baskan Gida, and that Baskan Gida had not kept up with the delivery periods specified in Ferrero's purchase orders. It was therefore suggested in cross-examination that, both by reason of what it observed of Baskan Gida's performance, and also by reason of its own substantially increased level of orders to Oltan Gida, Ferrero no longer genuinely expected to do 30,000mt of business with Baskan Gida during the 2001 season.
330. Secondly, the Banks maintain that the average rate of orders described in clause 3.2 as being "every 2/3 days" was not, to Mr Casale's knowledge, being achieved in relation to the 2001 crop, as it had been during the previous year. This allegation formed a minor aspect of the more general proposition that it was no longer business as usual between Ferrero and Baskan Gida.
331. Thirdly, the Banks say that Mr Casale must have known by 15^th October, from Baskan Gida's arrears of deliveries, that it could not have purchased the hazelnuts necessary to make delivery against all Ferrero's orders, as suggested by clause 3.1 of the Framework Contract. Other allegations, such as that there had been loss making (and therefore uncommercial) contracts by 15^th October, were not pursued.
332. In my judgment, the signing of the Framework Contract involved no dishonest misrepresentation by either Mr Rosa Brunet or by Mr Casale, or therefore by Ferrero. In fact, the average rate of orders by October 2001 had moved out from 2 to 3 days to 4 to 5 days, but this small misrepresentation was in my judgment immaterial. Furthermore, I accept Mr Casale's evidence that the reduction in frequency of orders was attributable to the availability of a larger crop, and the ability therefore to increase order size and place them less frequently. More generally, for the reasons given in detail in my description of the ordering and delivery of 2001 crop hazelnuts up until 15^th October, I have not been persuaded that anything had occurred sufficient to displace either Ferrero's policy of broadly maintaining the level of its orders from Baskan Gida, or its expectation that the same level of deliveries would in fact eventually be made.
333. Although Mr Casale understated in his evidence his level of awareness of Baskan Gida's arrears of deliveries, I am not persuaded that he had given any particular thought to the reasons for it. For all he knew, they could have been caused by processing capacity problems in the Baskan factories, or by delayed deliveries to Baskan Gida of nuts which it had ordered from its suppliers. I think it highly unlikely that the Baskans would have told him the truth, which was indeed that financial constraints were hampering the rate at which they could fulfil Ferrero's orders, by preventing them from ordering supplies described as purchased in the DPRs.
334. In reaching that conclusion I have not ignored the fact that 15^th October 2001 was the very last day upon which Baskan Gida delivered up to date DPRs and PCs to Ferrero as part of the ordering process, or that the last order placed by Ferrero on Baskan Gida which had been negotiated in a conventional manner by reference to that process was placed on 16^th October. There is simply no evidence to suggest that, without the benefit of hindsight, anyone at Ferrero knew on 15^th October that this process was about to come to an end.
335. Subject therefore to the minor inaccuracy as to periodicity of orders, and to the failure to include copies of DPRs or PCs referred to in clause 3.1 and clause 3.2 or to complete the banking details in clause 6.2, the Framework Contract provided a broadly accurate description of the basis upon which, on the day when it was signed by Ferrero, it carried on the business of purchasing hazelnuts with Baskan Gida.
336. The banks claim that the Framework Contract contained two further representations, namely:

i) That Ferrero expected the trading relationship with Baskan Gida to continue on the terms described in the future;

ii) The representations contained in the July Letter remained true.

337. In my judgment the Framework Contract made neither of those representations. Mr Casale had rejected an earlier draft which contained terms about how the relationship might be altered or terminated. In that context, the Framework Contract was silent as to how it might be performed in the future. Even if it contained such a representation, the Banks were soon thereafter, and before entering into the Facility, disabused of any such assumption, by two events which I shall describe. The Framework Contract said nothing about the July Letter, expressly or by implication. Even if it had done in the manner alleged, that would have been substantially true.
338. Clause 9 of the Framework Contract (relating to assignment) was in substantially the same form as the assignment clause in the original draft sent, in English and Italian, on 20^th September. It was accidentally omitted, together with clause 8 about choice of law, from the 5^th October draft, which Mr Casale said he had studied with care, and both clauses were then added back on 9^th October. Mr Casale attempted unconvincingly to deny studying clause 9 when it reappeared alongside clause 8, although since he had originally objected to the earlier choice of English law I infer that he must have studied the new clause 8 carefully. He was driven (by an agenda to deny any understanding of the English word 'assignment' in later documents) to say he had not given any consideration to the meaning of clause 9, and that he signed the Framework Contract (in fact on every page) without understanding that part of it. In my judgment, having received a non-technical Italian translation of the word assignment on 20^th September, he knew perfectly well that it meant transfer or cede to another, and approved the Framework Contract on that basis.

The Baskans Embark Upon their Fraud on the Banks

339. It is difficult to be precise as to the moment in time at which the continuing deterioration in their financial affairs led the Baskans to decide to attempt, as it were to steal, rather merely to borrow, money from the Banks. In my judgment, it must have roughly coincided with the equally elusive moment at which the Baskans appreciated that Baskan Gida was hopelessly insolvent (in the sense that they could no longer see light at the end of the tunnel). I have already concluded that Baskan Gida was balance sheet insolvent by the end of June. The fact that it could only pay its outstanding debt to HBZ (guaranteed by Indo-Med) at the end of August by misapplying part of the Dahod Investment suggests that it was commercially insolvent by then, but that is not of itself a sufficient reason for concluding that the Baskans did not regard the negotiation of a large facility from the Banks as offering a long term solution to their company's difficulties.
340. Mrs Pay gave a vivid account of Baskan Gida's deteriorating financial position during 2001 in her witness statement. It included the following paragraph:

"69. I can remember things were so serious that in August or September 2001, I suggested to Fuat Baskan that the company should sell its recently acquired assets such as the Giz Bizkuvi factory as a matter of urgency. I was surprised by Fuat Baskan's response. He said that it was too late to save anything. It was the first time that I had heard a member of the family or the company management suggest that the company would not survive."

In cross-examination on this paragraph she said that Fuat Baskan might have intended to refer not to the whole of Baskan Gida's business, but rather to the Giz Bizkuvi factory itself. Nonetheless, her evidence offers a small pointer towards identifying the relevant time.

341. More concrete, albeit heavily contested, evidence that the Baskans no longer deserved their long established reputation for commercial propriety is to be found in their treatment of their Turkish banks during the last quarter of 2001. The Banks' case is that the Baskans were by then well embarked upon a systematic fraud against Baskan Gida's Turkish bankers, by raising finance against supposed, but in fact non-existent, deliveries to Ferrero, to Indo-Med and to Cap Industries.
342. The question whether the Baskans were defrauding Baskan Gida's Turkish banks during this period is not of direct relevance to the Banks' claim and is only one among a number of factors going to the question when the Baskans embarked upon the fraud on the Banks which all the active parties in this litigation accept took place. Nonetheless an important part of the Banks' original case against both Indo-Med and Ferrero was that they each knew, by November and December 2001 respectively, that the Baskans were defrauding their Turkish bankers. In closing the Banks abandoned the allegation that Ferrero knew this, but persisted in it as against Mr Abidali. No analysis of that allegation can sensibly be conducted without first deciding whether, and if so how and when, the Baskans did defraud Baskan Gida's Turkish banks.
343. In the foreground of any consideration of this question by an English court lies the fact that Ahmet and Cevat Baskan together with Latif Aksu were all prosecuted and acquitted of just such an offence by the Giresun High Criminal Court in 2002/03, and that the acquittal followed and concurred with an opinion of an expert investigating committee appointed by the court, to the effect that the allegation of fraud was not merely unproven, but positively incorrect. The gist of the charge against the defendants in the Turkish criminal court was that they had raised money from Baskan Gida's Turkish bankers by dishonestly pretending that hazelnuts had been exported, by the use of misleading customs and other documents, including invoices, against which the Turkish banks lent. The conclusion of the expert committee and of the court was that the relevant loans had constituted pre-export finance, that is loans made against draft customs and other documents and pro-forma invoices, and that the customs and freighting documents presented to the Turkish banks demonstrated on their face that the relevant hazelnuts had yet to be exported, so that the Turkish banks could not have been misled in the manner alleged.
344. Nothing daunted, the Banks persisted in their allegation that the Baskans had indeed been defrauding their Turkish bankers, by reference not only to the materials before the Turkish criminal court, but also by reference to additional evidence obtained from the Banks' own forensic inquiries, together with oral evidence from Ms Pay, and an important admission extracted from Mr Abidali in cross-examination, based upon documents disclosed by Indo-Med's liquidator.
345. On the case put to the Turkish criminal court, it mattered not whether the Turkish bank borrowing related to proposed exports of hazelnuts which were the subject of orders or contracts between Baskan Gida and its relevant customers at the time when the borrowings were made. The alleged fraud was simply pretending that the nuts had actually been exported when they had not. Accordingly that question does not appear to have been subjected to any detailed forensic analysis in the Turkish criminal court, and I infer that both the expert committee and the court acquitted the defendants on the assumption, not significantly challenged, that such orders or contracts existed. Furthermore, the Turkish court's focus was on the alleged falsity of customs documents rather than invoices.
346. The Banks' attempted resurrection of the fraud case dismissed by the Turkish criminal court was based upon two main foundations. The first was that, in relation to a substantial number of the purported intended exports to Indo-Med on the back of which Baskan Gida obtained pre-export finance, there was simply no pre-existing underlying transaction between Baskan Gida and Indo-Med, such that Baskan Gida obtained from its Turkish banks finance based upon the false premise that hazelnuts had even been ordered. By contrast, the Banks accepted that there were genuine orders from Ferrero in respect of all the pre-export finance which Baskan Gida obtained from its Turkish bankers, based upon intended deliveries to Ferrero.
347. The second foundation of the Banks' case was that, throughout, Baskan Gida used false invoices when raising pre-export finance in relation to fictitious orders from Indo-Med and genuine orders from Ferrero. For present purposes it is sufficient for me to concentrate on the first of those two planks in the Banks' case.
348. On 20^th September 2001 HBZ (Indo-Med's bankers) sent Indo-Med a set of twenty invoices with accompanying CMR documents which it had received from Emlak Bankasi, one of Baskan Gida's Turkish banks, requesting Indo-Med's authority to debit its account with payment of invoices with an aggregate face value of US$949,600, in relation to some 400mt of hazelnuts. None of the copy invoices identified a Baskan Gida/Indo-Med contract number, as they should have done. By 15^th October 2001 at the latest Mr Brian Venables, a bookkeeper employed by Indo-Med, had prepared a list of these invoices, and in cross-examination Mr Abidali accepted that within about two days of receipt of those copy invoices from HBZ, Mr Venables would have appreciated that there were no underlying contracts for the supply of hazelnuts to Indo-Med, to which the invoices presented to HBZ and copied to Indo-Med related.
349. I shall return in due course to address the question what these events must have revealed to Mr Abidali about the Baskans' practices. For present purposes, Mr Abidali's admission, based upon reviewing the documents in question (and in particular his own bookkeeper's schedule) that Baskan Gida's Turkish banks were seeking payment in respect of supposed consignments of hazelnuts which had not even been ordered by Indo-Med, demonstrates the truth of the Banks' case that, in a respect not investigated by the Turkish criminal court, Baskan Gida was defrauding at least one of its Turkish banks by borrowing, pre-export, against pretended orders by Indo-Med, by the end of September 2001.
350. Evidence that Baskan Gida had descended to fraud as against one of its bankers under the financial pressures facing it in late 2001 does not of itself prove that the Baskans had formed a similar intention with regard to the Banks by that time. Prior to the end of October 2001, there is only one clear example of dishonesty in the Baskans' dealing with the Banks. From the outset, Baskan Gida's presentation through PRI of a lending opportunity to the Banks had centred upon its apparently profitable business with Ferrero as the jewel in its crown, even though at that stage it was contemplated that any facility would also finance business with other European customers. At no time during those negotiations did Baskan Gida reveal the existence or performance of the Dahod Agreement, which not only constituted, as between the parties, a self-sufficient basis for financing Baskan Gida's business with Ferrero, but also transferred to Indo-Med and the Dahod Investors the whole of Baskan Gida's US$140 per mt agreed profit margin on that business, under an agreement with the Dahod Investors designed to persist throughout the 2001 season. The Banks were allowed to assume that Baskan Gida's business with Ferrero was then financed by Baskan Gida's Turkish bankers, and that the Turkish borrowings attributable to that business would, if the Facility went ahead, be paid down out of the first advances made by the Banks.
351. It may be that the Baskans hoped to deal with this piece of double financing by treating the Dahod Investment, contrary to the terms of the Dahod Agreement, as in essence a stop-gap pending the establishment of a larger and cheaper facility with the Banks. The very high profit margin agreed lends force to that view. If so, there is no evidence that they mentioned this either to Mr Abidali or to the Dahod Investors. It is nonetheless difficult in those circumstances to treat the concealment of the Dahod Agreement and Investment from the Banks as compelling evidence of an already formed intention by the Baskans to defraud the Banks.

The 45 Day Lie

352. The existence of an intention by the Baskans to defraud the Banks becomes unmistakable from the beginning of November 2001. Until then, the presentation by Baskan Gida of its business relationship with Ferrero suggested that, on average, Baskan Gida received payment from Ferrero in respect of any particular delivery of hazelnuts, about 2 to 3 weeks after buying the necessary nuts from its suppliers. Thus for example, a document headed "Baskan Payment Cycle" sent to the Banks in June 2001 identified the period needed for purchasing and processing as 14 to 21 days (for large orders sent by ship) and 7 to 14 days (for smaller orders sent by truck), the period for payment as being in each case 2 days after shipment, with consequential periods of Turkish exposure for the Banks as 16 to 23 days for large orders and 10 to 16 days for small orders.
353. On 19^th November 2001 Mr Stambler of PRI wrote to Mr Byles and Mr Lewis, acknowledging that a 14 day average turn around between purchase by Baskan Gida of nuts from its suppliers and payment for an export of the same nuts by Ferrero had historically been typical of the Baskan Gida/Ferrero business relationship, and continued as follows:

"When the peak purchasing and shipment period (August-November) ended at the beginning of November, Ferrero modified the process in order to take advantage of price fluctuations and high quality crop. As a result, whenever high quality crop enters the market or the price of hazelnuts drops to a preferable level, Ferrero asks Baskan to purchase nuts on or about the day it places the order and specifies a future date (maximum 6 weeks) for shipment.

Baskan purchases the merchandise and stores the merchandise in its warehouse until the shipment date as shown on the purchase order. The corresponding storage and financing costs are factored into the sales price. Under the new terms, Ferrero continues to make payments immediately after Baskan ships the goods.

The process can be summarised as follows:

Day 0: Ferrero sends purchase order to Baskan and specifies a ship date (6 weeks maximum)

Day 0-5: Baskan purchases the nuts and stores them in its warehouse

Day 6-42: Baskan ships the goods to Ferrero

Day 42-45: Baskan receives payment from Ferrero."

354. The Banks took this statement at face value, re-structured the Facility to allow for a longer period between advance and repayment, and did not verify any part of it with Ferrero, from whom the statement had been concealed. It represented a significant alteration in the trading relationship from that described in the Framework Contract, which was silent about Baskan Gida storing nuts for Ferrero, or adding storage and financing costs to the invoice price.
355. In fact, almost every part of the passage which I have quoted was untrue. In particular:

i) The peak purchasing and shipment period had not passed, although, for quite different reasons, Ferrero had placed no further orders with Baskan Gida since 16^th October.

ii) Ferrero had done nothing to modify the process, whether for the reason stated or at all.

iii) Specifically, Ferrero had not asked Baskan Gida to purchase nuts with a view to keeping them for up to 6 weeks before shipment, nor had Baskan Gida done so.

iv) There had been no adjustment to the sales price by reference to increased storage and financing costs.

The only true part of the statement was that, as usual, Ferrero continued to make prompt payment for nuts on delivery.

356. The reason why Baskan Gida caused PRI to make this statement (to which I shall refer, like the parties, as "the 45 Day Lie") may readily be inferred. Baskan Gida wished to obtain a longer period of use of the Banks' advances than would be justified by reference to its business with Ferrero which was in the process of becoming the only part of its business to be financed by the Facility. Accordingly, Baskan Gida presented a wholly untrue and misleading description of the financing period necessary in relation to nuts to be supplied to Ferrero.
357. The extension of the financing period from approximately 14 to 45 days in due course played a vital part in enabling Baskan Gida to misapply the Banks' advances, between the period when the money started to be advanced and the date when non-payment put the Banks on notice that something was amiss. While no documents or other evidence enable it to be said with certainty that Baskan Gida planned prior to January 2002 to misuse the advances by the purchase of hazelnuts and the disposal of them to a phoenix company within the extended finance period, it is again probable that from the moment when the 45 Day Lie was prepared, the Baskans had such a scheme in mind.
358. The Banks' case is that both the 45 Day Lie and the phoenix scheme which it was intended to facilitate were planned at meetings between Melih Baskan and Mr Abidali on 1^st November, in London. Mr Abidali's case is that Melih Baskan came to London with Mr Franko to take part in a trade fair, and that he took the opportunity of their visit to arrange for a first meeting between Mr Dahod and Melih Baskan, so that Mr Dahod could, as it were, clap eyes on an important member of the Baskan family responsible for managing his investment, receive a presentation about the business in which he had invested, and ask any relevant questions about it. Mr Abidali denies knowing anything specific about the then state of Baskan Gida's negotiations with the Banks, any involvement in the 45 Day Lie, and any knowledge of or participation in the putting together of the phoenix scheme until January 2002, when he says that it was put to him for the first time during a visit by him to Giresun for an unconnected purpose to do with litigation in the USA. My findings on this issue now follow.
359. I have found no reason to doubt that the original purpose for Melih Baskan's and Mr Franko's visit to London was to participate in a trade fair. Equally, it is clear from correspondence subsequently disclosed by Indo-Med's liquidator that Mr Abidali did take the initiative of using their visit as a means of effecting a first introduction between Mr Dahod and a member of the Baskan family.
360. With the assistance of a Barclaycard statement and Mr Abidali's admissions in cross-examination, it was established that he picked up Melih Baskan and Mr Franko at Heathrow airport on 1^st November, took them back to Indo-Med's offices, then to their hotel, and that, in the evening, all three of them went out for dinner at a Chinese restaurant.
361. The meeting with Mr Dahod took place the following day. Documents made available to the Banks by PRI, together with metadata from one of its computers in New York, show that an early version of what became the 45 Day Lie was the subject matter of emails passing between the Baskans and PRI on 1^st November. After further work, the 45 Day Lie made its first appearance in the course of a telephone conference between Mr Politis of PRI, Mr Lewis and Mr Byles on 16^th November, before being spelt out in terms in the letter from Mr Stambler of PRI, to which I have referred.
362. On that rather slender basis, Mr Wardell submitted (and put vigorously in cross-examination) that Mr Abidali had taken the opportunity of their Monday meetings to discuss with Melih Baskan a scheme whereby the Dahod Investment would be repaid, including the transfer of Baskan Gida's assets to a phoenix company after they had been swelled by means of the working capital provided by the Banks' lending. Mr Wardell did not suggest that this had taken place or continued on 2^nd November at the meeting with Mr Dahod, who was not cross-examined on that basis.
363. For their part, both Mr Abidali and Mr Dahod denied being aware of any financial predicament of Baskan Gida sufficient to warrant any discussion, let alone planning, of an early repayment of the Dahod Investment before the end of the 2001 season (i.e. July 2002). On their evidence, they regarded the investment as prospering, by reference to the tonnage which Melih Baskan had told them had by then been shipped to Ferrero, and were not unduly concerned that their profit share had been re-circulated in the financing of that business. Their case was that on 2^nd November they did ask Melih Baskan to provide better reports and accounting than he had to date about the progress of the joint venture, and asked Melih Baskan to prepare an agreement recording the monies by then invested. Much later, Mr Abidali used this evidence as an excuse for backdating important parts of the contractual structure leading to the asset transfer to Aksu Gida, to make it appear that a Protocol Agreement for the repayment of the Dahod Investment, together with associated promissory notes, had been put in place in November 2001 in response to that request.
364. The Banks tried to bolster their case that Mr Abidali was complicit in the Baskans' intended fraud on the Banks by this time by the allegation that by then he both knew about and was assisting the Baskans' fraud on their Turkish banks. I have explained the basis of the allegation that Mr Abidali knew of it. The allegation that he participated in it arises from the fact that he did not protest Emlak Bankasi's demand on HBZ until 2002. Mr Abidali's evidence was that early demands for payment were commonplace, and they were routinely put on one side with no thought being given to them.
365. In my judgment, the Banks' attempt to implicate Mr Abidali in a conspiracy with Baskan Gida to defraud them as early as 1^st November fails for lack of evidence. Conversely, I do not accept Mr Abidali's account, weakly supported by Mr Dahod, that they asked Melih Baskan for a written agreement containing a repayment programme on 2^nd November. I will take each issue in turn.
366. As for the Banks' conspiracy allegation, I consider that the probability that the Baskans had certain brief communications with PRI relating to what became the 45 Day Lie, on 1^st November, while Melih Baskan was in London, is too slender a foundation for implicating Mr Abidali at that early stage in the Baskans' by then dishonest plans against the Banks. It is not clear that he was personally involved in those communications. Mr Wardell sought to reinforce the Banks' case on this point by cross-examining Mr Abidali along the lines that he must by then have appreciated how severely Baskan Gida's performance had fallen off due to its financial predicament. When it is borne in mind that the final instalment of the Dahod Investment had only just been made, in the form of a further US$3 million, it seems to me unreal to suppose that Mr Abidali went from optimism about the investment to a state of such pessimism that he was ready to participate in a fraud in order to obtain repayment for Mr Dahod, in the space of one week. In that context I have not on balance been persuaded that by the beginning of November 2001 Mr Abidali knew that the Baskans were defrauding their Turkish bankers, still less that he agreed to assist them by remaining quiet about Emlak Bankasi's demands. I think it probable both that the Emlak Bankasi demands had been reported to him, and that he suspected that something was amiss, but not that a fraud was in progress.
367. More importantly, I consider that the probabilities weigh strongly against the Baskans having involved Mr Abidali in their plans at that early stage. To take him into their confidence would have involved revealing in full Baskan Gida's hopeless financial predicament, but at a time when they could by no means be sure that the Banks' Facility would definitely be put in place. In other words, they needed to maintain the confidence of the Dahod Investors, and of Mr Abidali as their adviser and representative, for as long as the obtaining of the Banks' finance remained uncertain.
368. Mr Dahod's evidence about the 2^nd November meeting was that the presentation by Melih Baskan was one of optimism, in which he attributed his failure to provide sufficient written detail about progress to pressure of business rather than financial embarrassment. That part of his evidence accords with my view as to the probabilities, at that delicate stage for the Baskans.
369. As for Mr Abidali's case that he asked for a written agreement as to repayment on 2^nd November, that is in my judgment equally improbable. While Melih Baskan may have promised to provide more regular reports of tonnages shipped to Ferrero, I reject Mr Abidali's evidence that Melih Baskan said that:

"Baskan Gida would also be happy to provide written confirmation of its indebtedness under the Dahod Agreement, by way of a loan agreement, promissory notes or other suitable documentation. He undertook to look into this following his return to Turkey."

370. No such documents were provided by Baskan Gida to the Dahod Investors in 2001, and the surviving records of further communications involving the Dahod Investors, Mr Abidali and Baskan Gida for the rest of 2001 disclose no reference to such a promise, still less any complaint that, had it been made, Melih Baskan had not honoured it.
371. I do however accept Mr Abidali's case, and his evidence that Melih Baskan told them nothing about the progress or content of Baskan Gida's ongoing negotiations with the Banks. Again, the probability is that, faced with having financed the new season's Ferrero business from the Dahod Investment, while seeking finance for the same business from the Banks, Melih Baskan chose to say as little about that growing embarrassment as he could get away with, at least until the Banks' Facility was safely in place.

Ferrero/Baskan Gida trading – Mid-October to 13^th December 2001

372. Having secured the Framework Contract, the remaining commitments from Ferrero sought by the Banks before the Facility could be put in place were, first, a re-statement of the trade reference in the July Letter and secondly, an acknowledgement of the assignment of debts by Baskan Gida to the Banks in relation to sales of hazelnuts to Ferrero. Both of these were provided by Ferrero Industrial in early December. Since both the repeated trade reference and the Ferrero Industrial acknowledgement of assignment are alleged to have dishonestly misstated the then state of the Baskan Gida business relationship, it is necessary to describe its development between mid-October and early December in some detail.
373. As I have already mentioned, DPRs and PCs ceased altogether after 15^th October, and after 16^th October Ferrero placed no further new orders with Baskan Gida until 6^th December. Meanwhile, deliveries by Baskan Gida continued at a similar weekly rate to that achieved from mid-September onwards (between 872mt and 448mt) before declining to just over 300mt and 200mt in the first two weeks of December. Aggregate deliveries rose from 6,480mt by 15^th October to 11,448mt by 12^th December. This was significantly worse than the delivery rate achieved during the previous year, and worse than the rates being achieved by Oltan Gida during the same period.
374. There is unchallenged evidence from Mrs Pay that Ferrero representatives at Giresun were complaining in general terms about delays in deliveries during October and November, but the reliability of her recollection is (as I have said) a little undermined by her next comment, that there were many days in October and November when no deliveries, or only one out of five or more planned deliveries, were made. This assertion is contradicted by the detailed agreed delivery schedules. Ferrero did not call either of its local representatives, Messrs Altare and Scavino, to give evidence. It is a fair inference that there was some expression of dissatisfaction by them about Baskan Gida's delivery performance.
375. Particular emphasis was placed by Mr Wardell on Baskan Gida's underperformance against orders from Ferrero France. Mr Casale put this down in cross-examination to storage capacity problems in France, saying that delivery dates were therefore deliberately put back at Ferrero France's request. Despite a spirited attempt by Mr Wardell to suggest that this was an opportunistic on-the-spot lie, the month-end agreed stock figures for Ferrero France largely bore out Mr Casale's recollection on this point, at least until the end of October. Thereafter surviving documents suggest that Ferrero France's storage capacity increased substantially.
376. The pattern of orders changed more fundamentally than the pattern of deliveries during that period, and calls for correspondingly closer examination. The pause in Ferrero orders between 15^th October and 6^th December coincided with a rise in prices, during October, from TL2.6 million (100kg 11-13mm TS3075 kernels) at the end September to TL3.6 million by the end of October, exacerbated by FKB's intervention. Ferrero's case (supported by Mr Casale's evidence) was that, faced with an (unexpected) tightening of the market, Ferrero simply backed off, at least so far as Baskan Gida was concerned. That was a sufficient reason for the flow of DPRs and PCs to cease for the time being since, as I have described, they were generated by Ferrero indicating an interest to Baskan Gida in making purchases, invariably by Mr Casale so informing Melih Baskan over the telephone.
377. The Banks' case is that Ferrero did not withdraw from the market at all, but merely discontinued ordering from Baskan Gida, on account of its deteriorating delivery performance, and the backlog which had by then built up. They categorise this period as one during which Ferrero's relationship with Baskan Gida broke down. This is therefore an important issue calling for a close look at the orders which Ferrero did place with other suppliers during that period.
378. The Banks rely upon five orders, the relevant details of which are as follows:

i) 312mt of rocher ordered by Ferrero Germany on 19^th October from Oltan Gida at €2,580 per mt.

ii) 1,000mt of Ordu ordered by Ferrero Italy from Barbero Dario on 8^th November.

iii) 600mt of Levant paste ordered by Ferrero France from Oltan Gida on 8^th November at €2,788 per mt.

iv) 312mt of Ordu rocher ordered by Ferrero Germany on 16^th November from Oltan Gida at €2,890 per mt.

v) 96mt of rocher ordered by Ferrero Germany from Oltan Gida on the same day at the same price.

379. By comparison, the last order placed with Baskan Gida, by Ferrero France on 16^th October was for 600mt of Levant paste at €2,324 per mt.
380. The first two of the five orders to which I have referred can be put on one side. The first was made just as the price rise was happening, was not at a substantially higher price than Ferrero had been paying for rocher kernels, and fits in to the series of orders which Mr Casale described as having been made before Ferrero temporarily withdrew from the market. The second was not an order for original nuts, but a reselection order, whereby Ferrero Italy arranged for Barbero to reselect among nuts already obtained from Turkey, so as to yield a consignment suitable for freezing. That leaves one order by Ferrero France for 600mt on 8^th November, and two orders for an aggregate of 408mt by Ferrero Germany, all from Oltan Gida. Those were all at prices which broadly reflected the increase in the market price, and appear prima facie inconsistent with a case that Ferrero withdrew totally from the market during that period.
381. Mr Casale's evidence that he did regard the market as having (from Ferrero's perspective) suffered a hardening during late October and November, from which he expected it in due course to recover, is supported by a note found on his personal computer and made on 12^th November 2001, containing a brief commentary of his on the then market for almonds and hazelnuts. The hazelnuts section reads as follows:

"For hazelnuts, like almonds, 2001 has seen the biggest ever harvest, bringing prices crashing down from 2000 levels to about $2 per kg.

Prices bounced back in October 2001 as a result of high demand caused by intervention by the European Commission (threatening protectionism) and by the reluctance of European producers to sell.

At present the market price is about US$2.50 per kg and huge quantities of in-shell goods are still unsold."

382. Mr Casale explained that the reference to intervention by the European Commission related to a threatened import tariff on hazelnuts, which in the event never materialised, but the threat of which caused European producers to withhold stocks hoping for a market rise in Europe, and European buyers to increase their orders from Turkey ahead of the feared intervention, thereby causing the market in Turkey to harden.
383. The Ferrero Germany and France orders from Oltan Gida during November demonstrate that Ferrero did not entirely withdraw from the market. Nonetheless, even taking those orders into account, its level of buying was very much less in November than in either October or December 2001. Ferrero Italy, the operating company with which Mr Casale was most closely associated, made no fresh purchases at all. The evidence thus largely but not entirely bears out Mr Casale's explanation for the dearth of orders from Baskan Gida during that period.
384. The question nonetheless remains why Ferrero France and Germany made the modest orders which they did in November from Oltan Gida rather than Baskan Gida. Mr Casale suggested that, by contrast with Ferrero Italy, Ferrero Germany's principal supplier for rocher nuts at that time was Oltan Gida rather than Baskan Gida. In my judgment the precise reason must remain, after this lapse of time, a matter for speculation. Nonetheless I am satisfied that the cessation of orders from Baskan Gida from mid-October through November 2001 was not the result of a loss of confidence by Ferrero in its ability to supply, mainly because of the substantial orders which Ferrero placed in early December, to which I will return in detail in due course.
385. On 19^th November, the very same day on which the 45 Day Lie was put in writing by PRI to the Banks, Melih Baskan wrote to Mr Casale setting out, after an earlier discussion, Baskan Gida's expectations about the rest of the 2001 season. The gist of the letter ("the 19^th November Letter") was to suggest that the increased prices which by then had persisted for nearly a month were the result of market misinformation supported by FKB about the size of the crop, to the effect that it was smaller than Baskan Gida believed it to be, coupled with widespread retention of hazelnuts from the market at all levels, from growers to exporters. As he put it: "everybody from manav to exporters are long up to head". Melih Baskan expressed his opinion that there was probably still about 400,000mt of nuts in shell yet to reach the market from a crop which he estimated at about 750,000 to 800,000mt, basing himself on estimates of the amounts still in the hands of growers, crackers, traders manavs and exporters. He continued as follows:

"Analysing all above and considering quantity coming from growers will be less during Ramadan, we think that the market will remain at the same levels of today's with slight increases and decreases until new year and starting January starting market will be easier and later in March.01 to May.01 will be lower even below FKB's purchasing price and that even without having any money FKB may receive goods from growers.

Also the situation of 2002 crop for this moment looks to be quite good on basis of mail (sp male) flowers even there was some talks in the market that mail flowers were less. Of course new crop developments to be followed which will be a point which can also be a point for the market.

Dear Casale, as we discussed we would like to confirm that we can do 17,000 – 18,000 metric tons until end of February.2002 both for Rocher and Pasta. As mentioned we would request if can have some like 2,000 – 3,000 metric tons Standart to be shipped during April.2002 and if possible May.2002 for which we can discussed to finalise the contract from these days if you would prefere."

386. Fax transmission details show that this letter was undoubtedly sent on 19^th November by Baskan Gida to Ferrero, but it was roundly challenged as a sham, concocted by Melih Baskan and Mr Casale together to provide an after the event market orientated justification for the subsequent placing by Ferrero of very large, long-dated orders on Baskan Gida at prices which, if marked to market on the date of the order, would be substantially loss-making for Baskan Gida. As will appear, such large long-dated orders were in due course placed by Ferrero, in December 2001 and February 2002. Apart from some orders which merely re-graded existing orders, and therefore involved no fresh supplies, the large long-dated orders constituted the entirety of the further business placed by Ferrero with Baskan Gida in the 2001 season, before Baskan Gida went out of business in early 2002.
387. In order to address this allegation of sham, it is necessary to place the 19^th November Letter in its context. By that date, Ferrero had ordered 16,000mt of 2001 crop kernels from Baskan Gida, and had received delivery of approximately 10,000mt. Melih Baskan's invitation to Mr Casale to place orders for a further 19,000 to 21,000mt during the rest of the season would have taken Baskan Gida's share of Ferrero's requirements to about 36,000mt.
388. Mr Casale's evidence about the 19^th November Letter was that he had had no prior discussion with Melih Baskan about Ferrero placing long-dated orders with Baskan Gida, but (although he could not remember them) would have continued to have discussions on the telephone with Melih Baskan about the state of the market from time to time, even during the period when Ferrero was placing no orders with Baskan Gida. He did not regard the market views set out in the letter as surprising coming from Melih Baskan, but he thought that it was highly unlikely that Baskan Gida would be able to provide Ferrero with up to 18,000mt kernels by February, on top of those already ordered and awaiting delivery. Nor did he have any intention of placing aggregate orders with Baskan Gida during the 2001 season significantly in excess of 30,000mt. He acknowledged that it was unusual for Melih Baskan to send out detailed views about the market in writing, no other similar example being found in the documents disclosed. But he recalled that it was not unprecedented.
389. Mr Rosa Brunet had no recollection of being shown the letter at the time, but said that, had he been shown it, he would not have taken much notice of it. He thought that Melih Baskan's attempt to predict the outcome of the 2002 crop as early as 2001 was unrealistically early.
390. Substantial parts of the November 2001 Letter were put to the hazelnut experts, in cross-examination. They agreed that, in their experience, Baskan Gida had a tendency to adopt generally optimistic predictions of market size year on year, and that Baskan Gida was an inveterate taker of short positions in the market, expecting prices to fall towards the end of the season. Furthermore, the experts both agreed that there had been a tendency in the market to underestimate the size of the 2001 crop, and that this had led to the mid-season rise in prices, followed by the late season fall, as the true size of the crop later became apparent. They doubted however whether Melih Baskan was correct in thinking that large scale retention of nuts or long positions by market participants were responsible for the hardening of prices which took place in and after October. While they agreed with Mr Rosa Brunet's view about the premature nature of Melih Baskan's attempt to predict the 2002 crop, they did not altogether rule out analysis of male buds (rather than flowers) on the hazelnut trees as providing at least some slight indication, albeit that the presence or absence of frosts in the first four months of the following year would be far more significant a factor in governing the 2002 crop size.
391. Mr Wardell made a number of textual points about the letter, such as the alternative references to "Casale" and "Mr Casale", a reference to a hazelnut conference in Nepal (obviously a mis-spelling of Naples), and to a supposed ability to predict the following year's crop by reference to male flowers in November. I found none of these persuasive. In a deliberate fake, one might expect to find fewer careless errors than in a hastily put together but genuine document.
392. In my judgment the allegation that the 19^th November Letter was a sham, contrived by Melih Baskan and Mr Casale acting together in furtherance of a conspiracy to defraud the Banks by putting in place uncommercial long-dated unprofitable orders, is misconceived. I consider that the letter was simply a sales pitch by Melih Baskan, designed to encourage Ferrero to place early orders for supplies to be delivered in and after February 2002, so that he could both use those orders as the basis for drawing down large sums under the still to be finalised Facility, long before having to use the funds thereby made available for the purpose of supplying Ferrero with processed hazelnuts, and at the same time achieve a near monopoly supplier status with Ferrero for the rest of the season, something which he had been seeking to achieve throughout 2001, but with no success. More than a month had passed since Ferrero had placed any fresh orders with Baskan Gida, and Melih Baskan needed to find some way of persuading Ferrero to re-enter the market as a large scale buyer, on terms which did not require Baskan Gida to risk pushing up prices further by making immediate large scale purchases in order to fulfil Ferrero's requirements. The market analysis provided by Melih Baskan in the letter offered a reasonable commercial basis for encouraging Ferrero down that route. While I have no doubt that Melih Baskan cannot have thought that Baskan Gida's existing financial resources would have enabled it to supply tonnage to Ferrero at anything like the levels contemplated in the letter, he no doubt expected that, before February 2002, substantial finance from the Banks would have become available, and (although Mr Casale could not recall one way or the other) he may have shared that expectation in discussion with Mr Casale.
393. Ironically, Melih Baskan's market analysis, namely that the high prices would persist into 2002, and then decline significantly, proved in the event to be broadly correct. As the hazelnut market experts both agreed, his prediction of market size turned out to be more accurate than the general market view in late 2001. Melih Baskan's mistake was however as to the time at which prices would fall. In the event, it occurred not in February 2002 but in May.

The Re-Grade Contracts

394. Between the 11^th and 17^th September 2001 Ferrero Italy had placed three contracts with Baskan Gida for rocher kernels in the aggregate tonnage of 2,208mt. They were Baskan Gida contract numbers 5063, 5065 and 5070. On 25^th October 240mt out of the amount then outstanding under contract 5065 (456mt) had been reordered via CAP for reasons which do not matter. By late November the aggregate outstanding under the three contracts (excluding the CAP re-order) was 792mt.
395. On 26^th November Ferrero Italy placed three orders, which were given Baskan Gida contract numbers 5084, 5085 and 5086, for an aggregate equivalent to the tonnage outstanding under the three September orders, but this time for paste rather than rocher. In respect of the September contracts 5065 and 5070 the delivery terms were also changed from CPT to FOT. The prices payable under the renewed November orders for the balance of the outstanding tonnage were arrived at by adjusting prices in the September orders by reference to the change in specification from rocher to paste, and the change in delivery terms. I shall refer to the contracts numbered 5084 to 5086 as "the Re-grade Contracts". They constitute the first group of contracts which the Banks categorise as un-commercial, and designed to assist Baskan Gida in defrauding the Banks. Similar allegations about earlier contracts were abandoned during closing speeches.
396. The Banks' case in relation to the Re-grade Contracts may be summarised as follows:

i) They were at off-market prices, and bound to create a substantial loss for Baskan Gida.

ii) They were surplus to Ferrero Italy's genuine requirements.

iii) They were deliberately renumbered so as to conceal from the Banks that they represented a re-ordering of hazelnuts already ordered in September.

iv) Since by late November 2001 Oltan Gida's performance had far exceeded that of Baskan Gida, any genuine order at this stage would have been given to Oltan Gida.

v) They were, therefore, shams to the knowledge of both Baskan Gida and Ferrero, and can only have been designed and implemented to further Baskan Gida's dishonest purposes, at a time when Ferrero knew that Baskan Gida was desperate for funds.

397. Ferrero's case and Mr Casale's evidence about these contracts was as follows. They came about because of unexpected difficulties which Baskan Gida, in common with other exporters including Oltan Gida, had encountered in relation to the quality of 2001 crop hazelnuts and rocher grade in particular. Whereas Oltan Gida had gone ahead and delivered rocher orders mainly on time, but sometimes of sub-standard quality, Baskan Gida had, so Mr Casale was informed, rejected rocher supplies and in consequence fallen significantly into arrears with deliveries, rather than deliver sub-standard nuts.
398. Ferrero Italy was concerned to obtain delivery before Christmas of sufficient rocher standard nuts to meet its freezing programme. One of the Golden Rules laid down by Soremartec was that all rocher nuts for freezing had to be delivered to Italy by Christmas. One of Ferrero Italy's Italian suppliers, Barbero, had laser processing equipment which enabled it to re-select rocher standard nuts from a paste standard delivery, better than could be expected from Baskan Gida's mainly manual processing facilities. Accordingly Mr Casale invited Melih Baskan to agree a re-grading of the three outstanding September rocher contracts from rocher to Ferrero Italy's paste specification, with a view to their being re-processed in Italy so as to release sufficient rocher standard nuts to complete the freezing programme on time.
399. As to the Banks' allegations, Mr Casale said that it was perfectly normal on a re-grade of an earlier contract to use historic prices with appropriate adjustments, rather the current market prices, reflecting the fact that there existed an outstanding un-performed order which, had it wished to, Ferrero could have required Baskan Gida to perform, or pay compensation in lieu to the extent that the market had since risen. As to the renumbering, Mr Casale said that he was entirely unaware of any improper purpose of Baskan Gida in that respect, nor indeed aware that these were contracts which Baskan Gida might wish to present to the Banks as new contracts, in connection with its obtaining of funding. As to the alternative of ordering from Oltan Gida, Mr Casale said that, whereas Oltan Gida had by then become a significant supplier of rocher nuts to Ferrero Germany (with a different and lower specification) Ferrero Italy still looked to Baskan Gida for the bulk of its rocher supplies, and saw no reason to switch suppliers in relation to these partly fulfilled orders to a supplier with similar quality problems of its own. Mr Casale's reference to Oltan Gida's problems with quality is corroborated by a letter from Oltan Gida to Ferrero in December 2001, abjectly apologising for a number of sub-standard deliveries to Ferrero Italy, and promising only that new processing equipment would solve the problem in the 2002 crop season. Generally, for reasons which will be apparent from my summary of Ferrero's case, Mr Casale denied that the Re-grade Contracts were either surplus to Ferrero Italy's requirements, un-commercial or, still less, a sham.
400. In fact, but this requires the benefit of hindsight, the Baskans delivered the outstanding tonnage under the Re-grade Contracts in full to Ferrero Italy but, as to the majority, after Christmas, and therefore too late to meet the requirements of Ferrero's freezing programme.
401. My conclusions in relation to the Re-grade Contracts are as follows. There is some evidence that Ferrero Italy used Barbero for reselection of rocher nuts for freezing from a paste delivery, but I consider it probable that, again, Mr Casale over-egged the pudding in his rationalisation of the underlying commercial purpose of the re-grading. It appears that only a modest proportion of the tonnage reordered as paste was delivered in time for the requirements of the freezing programme, and it is not clear that any of it was re-graded to extract rocher. Nonetheless the freezing programme appeared to have been fully supported by rocher deliveries, whether from Baskan Gida or other suppliers, without recourse to the deliveries made under the Re-graded Contracts. This part of Mr Casale's explanation only emerged for the first time in cross-examination, but I have concluded that in this respect, his memory played him false, and not that the explanation was a concoction which he knew to be false.
402. Nonetheless, I consider that, more generally, Ferrero's case that the Re-grade Contracts were normal commercial transactions rather than a dishonest sham is to be preferred as being much the more probable. In the absence of evidence that Mr Casale or anyone else at Ferrero was privy to the detailed terms upon which Baskan Gida was negotiating its proposed Facility with the Banks, I consider it fanciful to propose that by agreeing to the Re-grade Contracts, and to the use of newly numbered Baskan Gida contract forms, Ferrero was knowingly assisting Baskan Gida in the furtherance of any improper purpose. Above all, I consider the central attack on the Re-grade Contracts, as being at off-market prices, to be misconceived. Baskan Gida had promised to supply rocher nuts at the prices prevailing in September and had failed to do so within the delivery periods specified in Ferrero's purchase orders. In those circumstances, it seems to me nothing more than common-sense for a permitted down-grade in the specification required of Baskan Gida's delivery to lead to price adjustments based upon the originally negotiated September prices, rather than an agreement by Ferrero Italy to pay substantially increased prices (reflecting the harder market in late November) in relation to a variation of an outstanding order brought about by Baskan Gida's default.
403. The next series of important orders placed by Ferrero on Baskan Gida occurred at the beginning of December 2001, but in order to appraise them in their full context, it is necessary first to describe Ferrero's further engagement in the process of negotiation and finalisation of the Facility between Baskan Gida and the Banks.

Negotiating the Acknowledgement of Assignment – Round One

404. Since signing the Framework Contract on 15^th October, neither Mr Casale nor anyone else at Ferrero had had much contact with the Banks about the progress of the negotiation of the Facility, either directly or via Baskan Gida, subject to one exception. The Banks wished to understand more fully, and to have Ferrero's confirmation of, the relationship between Ferrero Industrial and the Ferrero operating companies. Inquiries by the Banks directly of Mr Casale at the beginning of November had produced an insubstantial response, and polite chasing from Melih Baskan in mid-October did no better. On 28^th November Ismet Baskan sent Mr Casale a draft letter to be written on a Ferrero letterhead addressed to Mr Byles and Mr Lewis, which had been prepared by PRI in accordance with their understanding of the relationship. That must have been unsatisfactory to Mr Casale, because later on the same day, after discussion between Mr Casale and Mr Politis of PRI, Mr Stambler sent him a shorter draft, which Mr Casale copied onto a Ferrero Industrial letterhead and signed, telling the Banks that Ferrero Industrial managed the relationship with suppliers including Baskan Gida, that Ferrero SpA owned Ferrero Industrial, and that purchases of goods were made in the names of the operating companies that received the merchandise. The episode demonstrated the care with which Mr Casale ensured that he committed Ferrero only to statements to the Banks with which, after reviewing drafts, he was content.
405. It will be recalled that at the heart of the proposed Facility, and as the main means whereby the Banks anticipated repayment of their advances, lay an assignment by Baskan Gida of its right to payment for supplies of hazelnuts to Ferrero. The Banks were therefore concerned to put in place documentation demonstrating that the necessary assignments would be notified to Ferrero and to obtain Ferrero's undertaking to pay the Banks. For that purpose, the Banks caused to be prepared draft notices to, and acknowledgments by, Ferrero of the requisite assignments, and invited the Baskans to obtain Ferrero's approval or comment on the drafts, before finalisation of the Facility.
406. On the morning of the 4^th December Ismet Baskan emailed to Mr Casale the then draft acknowledgment of assignment, asking Ferrero to indicate any desired amendments, telling Mr Casale that the Banks would request signature by him for Ferrero Industrial and also by the operating companies, and notifying him that the coming Friday 7^th December was the last day on which Baskan Gida could obtain the Facility.
407. The accompanying draft was headed "Acknowledgement Notice of Assignment of General Terms and Conditions for Sale of Hazelnuts" and was expressed to be given by Ferrero Industrial and five Ferrero operating companies (including Italy, France and Germany), and addressed to KBC. It referred in terms to something described as the "Framework Agreement" and purported to acknowledge receipt of a notice ("the Notice") from Baskan Gida informing Ferrero that Baskan Gida had assigned KBC all its rights, benefit and interest in the Framework Agreement. There followed 9 paragraphs, (numbered oddly 1-8 and 10), described as confirmations, coming from the Ferrero companies to KBC. The first consented to the assignment referred to in the Notice. The third provided that:

"We will not, without your prior written consent, seek to terminate or agree to terminate the Framework Agreement or agree to any variation of the Framework Agreement."

The fifth consisted of an undertaking to pay "any amounts due by us under the Framework Agreement to the bank account referred to in the Notice". The sixth acknowledged that KBC would be relying on the undertakings, while the eighth consisted of an English law clause.

408. Mr Casale must have read the draft Acknowledgement and, without the Notice to which it referred (which was not sent at the same time), he may be forgiven for having thought that the document was somewhat opaque. He clearly had a short telephone conversation with Ismet Baskan about it, because in mid-afternoon on the same day Ismet Baskan emailed him again, with a slightly amended version of the draft Acknowledgment and, for the first time, a copy of the draft Notice. His email continued:

"As you know these letters are prepared by the banks, as you said there are stupid things inside but it does not mean that they are all must. Since the closing date is coming friday and this credit line is very important for Baskan, please read both letters and tell me the things that are not convenient for Ferrero and our relationship, i will make those people to change into a way that you and the other Ferrero companies will sign."

409. In fact, the draft Acknowledgement which accompanied this second email differed slightly from its predecessor, in that wherever the predecessor referred to an assignment of "General Terms and Conditions" or to the "Framework Agreement", Version 2 referred to the Framework Contract. I infer that the slight and appropriate tightening of the language referring back to the Framework Contract was something done at the Banks' end, and was not responsive to any conversation Mr Casale had in the meantime had with Ismet Baskan. Apart from that, Version 2 was exactly as I have described Version 1.
410. The accompanying draft Notice was addressed from Baskan Gida to the same Ferrero companies as were referred to in the draft Acknowledgement and reads as follows:

"Dear Sirs,

We are pleased to inform you that Baskan Gida has agreed to terms on a working capital facility with Bank of Tokyo-Mitshubishi and KBC Bank NV. The closing is expected by mid-December. As a result, we will have access to a reliable source of funding which will enable us to continue to serve your needs in a timely fashion.

According to the terms of the facility, we have agreed to assign all payments from Ferrero SpA, Ferrero OHGMBH, Ferrero France SA, Ferrero Australasia Manufacturing Pty Ltd and Ferrero USA Inc to the arranging banks.

Attached please find the draft form of assignment acknowledgement, which informs KBC Bank NV that you agree to make payments for orders dated on or after December 10^th 2001 to the specified account of Baskan Gida. We are sending you this draft in advance so that you can be prepared to acknowledge the assignment promptly upon our request.

Please let us know if you have any questions regarding the attached.

Thank you for your cooperation.

Yours faithfully"

411. I infer from the text of Ismet Baskan's second email that although in their discussion earlier that day he had understood that Mr Casale was unhappy with Version 1, Mr Casale had not explained precisely why. I also infer that Ismet Baskan obtained a more detailed explanation of Mr Casale's concerns later that day, because at 9.48 the following morning (5^th December) Ismet Baskan sent Mr Casale a further version of the Acknowledgement (Version 3) in very different form. All references to the Framework Contract had been removed, and replaced by references to the General Terms and Conditions for Sale of Hazelnuts (i.e. as per Version 1). More importantly, all the numbered confirmations expressed to be given by the Ferrero companies to KBC had been removed, and replaced by a single confirmation in the following terms:

"We undertake to pay any amounts due by us under each sales contract to the following bank accounts:"

There then followed, under the headings Payments in US Dollars and Payments in Euro, references to KBC's name and address, an unidentified sort code, a reference to an account in the name of Akbank as collecting agent for Baskan Gida with an unidentified account number. No further draft of the Notice was supplied.

412. Shortly after 5 o'clock on the same day Ferrero was sent a yet further version of the draft Acknowledgement in a form designed to be signed by Ferrero Italy, together with an accompanying draft Notice addressed to Ferrero Italy. I shall refer to the draft Acknowledgement as Version 4. Apart from being specific to Ferrero Italy, it differed from Version 3 in omitting any reference to the General Terms and Conditions for Sale of Hazelnuts. It was still addressed to Baskan Gida for the attention of Ismet Baskan and (inexplicably) Alessio Casale. The undertaking to pay to designated accounts was in this version expressed to relate to "each sales contract dated on or after December 10^th 2001 until July 31^st 2001". The draft Notice was amended only so as to include July 31^st 2001 as the end-date for contracts to which it referred.
413. Mr Casale decided to show either Version 3 or Version 4 to Mr Rosa Brunet, and he decided that the documentation should be shown to Ferrero Italy's Treasury Department, specifically to Mr Salomone, for advice as to whether Ferrero Italy should sign it. Mr Salomone was not called to give evidence, but Mrs Di Matteo, who also saw Version 4, did. Her recollection was that Mr Casale brought Version 4 together with the Ferrero Italy version of the Notice to her, and she recognised manuscript annotations made by her on both documents.
414. As the person responsible for ensuring that Ferrero Italy paid its debts to the correct bank accounts of its creditors, Mrs Di Matteo thought it sensible to suggest the addition of the following phrase to the draft Acknowledgement:

"Same account information will appear on each invoice supplied by Baskan Gida to Ferrero Spa."

She discussed this with Mr Salomone, who agreed. I shall refer to that proposed addition as "the Same Account clause". She then so informed Mr Casale and, because the final version of the Acknowledgement contained the Same Account clause, I infer that he passed on that request to Ismet Baskan, not just for Ferrero Italy, but for all Ferrero companies required to sign the Acknowledgment.

415. The final versions of the Acknowledgements and Notices were emailed by Ismet Baskan to Mr Casale late on the following morning, 6^th December. The documents for Ferrero Industrial were in slightly different form from those for the operating companies, of which Ismet Baskan sent an example designed for use by Ferrero France. Taking each in turn, the acknowledgement to be given by Ferrero Industrial continued to be headed, and to make reference to, the "General Terms and Conditions for Sale of Hazelnuts", and the Notice requested Ferrero Industrial to make payments from all the Ferrero operating companies to the specified accounts. By contrast, the documents for use by the operating companies make no references to the General Terms and Conditions, and the Notices merely requested each individual operating company to make payments to the specified accounts. All the Acknowledgments and Notices contained the Same Account clause which Mrs Di Matteo had suggested should be included. None of the documents contained sort codes or account numbers for the specified accounts, but the Notices stated that Baskan Gida would inform the recipients of the account numbers and sort codes once available "after the closing", a reference to the completion of the Facility.
416. On the same day, Mr Casale showed the final version of the documentation to Mrs Di Matteo, and arranged for the Ferrero Industrial Acknowledgement to be signed by himself and Mr Rosa Brunet. The Ferrero Italy Acknowledgement was signed by Mr Rosa Brunet and Mr Salomone. Mr Casale also sent the appropriate versions of the Acknowledgements and Notices to the other Ferrero Operating Defendants, for signature by them, which duly occurred shortly thereafter.
417. Mr Casale was cross-examined at length about his part in the negotiation which I have just described. His evidence about it was unsatisfactory in a number of respects. First, he denied appreciating that completion of the facility was imminent, even though that was clearly spelt out to him in Ismet Baskan's first email sent on 4^th December. His explanation was that he largely ignored the email and concentrated on its enclosure (Version 1 of the Acknowledgement), because Melih Baskan had told him on the telephone in advance to expect it. I have no doubt that Ismet's sense of urgency was expressed in any telephone conversation as clearly as it was in the emails.
418. Secondly, having thus been forced to concede that he read Version 1 carefully, when taken paragraph by paragraph through it by Mr Wardell, he then denied appreciating the plain meaning of substantial parts of it, on the alternative grounds that he did not understand the meaning of the language, or that he did not read it with sufficient care. In particular, he persisted in denying that he understood the meaning of the word 'assignment' where it appeared in numerous places in all the Ferrero Industrial versions of the Acknowledgement. He also denied that he appreciated that the numerous references to the General Terms and Conditions for Sale of Hazelnuts, and to the Framework Agreement, were intended to be references to the Framework Contract.
419. Thirdly, he suggested that well before Mrs Di Matteo's intervention, he and Mr Rosa Brunet had already themselves discussed, and emphasised to Baskan Gida, that Baskan Gida was to have the duty to identify the correct payment account on any invoice. Fourthly, he said that he did not pay much attention to the draft Acknowledgment, regarding it just as a request to Ferrero to agree to pay invoices to any account subsequently stipulated by Baskan Gida from time to time. Finally, he described Ismet Baskan's reference in his second email on 4^th December to Mr Casale saying that "there are stupid things inside" as being a reference not to Version 1 of the Acknowledgment, but to a letter written to him by Mr Politis.
420. In addition to the specifically unsatisfactory parts of his evidence about this negotiation which I have described, Mr Casale's response to a searching cross-examination about the draft Acknowledgements and Notices displayed a degree of prevarication not to be found in the rest of his very lengthy cross-examination, which was unsatisfactory in itself.
421. The gist of Mr Casale's evidence about this round of the negotiation of the Acknowledgement was that he had no real understanding of the meaning of the documents, and gave them insufficient attention, save for ensuring first, that he did not become involved in signing for Ferrero companies by which he was not authorised, secondly that Ferrero did not commit itself to new obligations in a way which would cramp its freedom of action in dealing with Baskan Gida, and thirdly that the Same Account clause should be included in all documents to be signed by any Ferrero company.
422. In my judgment, Mr Casale approached round one of the negotiation of the Acknowledgments with the same meticulous care which he habitually applied to all his work, and which he exhibited to a marked degree in the preparation and giving of his evidence in these proceedings. In particular, as I have already said in relation to the use of the same word in the Framework Contract, I consider that he did understand the general meaning of the word 'assignment' and that the reason why he found it unnecessary to seek legal or linguistic advice about the documentation was that he had a reasonable understanding both of its meaning, as a matter of language, and of its commercial intent.
423. That is not to say that the draft Acknowledgements and Notices were models of clarity. Far from it. Version 1 of the Acknowledgement was little better than meaningless, read on its own, and was not accompanied by the draft Notice in Ismet Baskan's original email. Version 2 was a little better, and made explicit reference to the Framework Contract which, in my judgment, Mr Casale cannot have failed to appreciate. Nonetheless the draft Notice still rather confusingly refers to the consequence of Baskan Gida's agreement that all payments should be made to the arranging banks as being a request to acknowledge that Ferrero should make payments to a "specified account of Baskan Gida" when the accompanying version of the Acknowledgment identified no such account. This was however largely corrected by Version 3 of the Acknowledgement, although at that stage the precise account numbers and sort codes were not identified. Nonetheless, the lack of clarity in the draft documentation by no means justifies the level of incomprehension which Mr Casale asserted in relation to it.
424. Mr Rosa Brunet was also cross-examined, but only briefly, about round one of the Acknowledgement negotiation. His role was a minimal one, being shown Version 3 or 4 without, according to him, seeing the Notice to which it referred, and suggesting that the draft be sent to Ferrero Italy's treasury department for their views. It is surprising that a sophisticated businessman like Mr Rosa Brunet should have thought he could sensibly appraise the draft Acknowledgement without sight of the Notice to which it prominently referred. His explanations were first, that Ferrero did not have a copy of the draft Notice when he was shown the draft Acknowledgment, and secondly, that he did not realise there was meant to be a notice attached to the Acknowledgement. The first explanation was, by the time Mr Rosa Brunet was shown the documentation, wrong in fact, since Ferrero had received a draft of the Notice on the previous day. I find the second explanation difficult to believe. The most that can be said about Mr Rosa Brunet's involvement at this stage, is that he received the document from a trusted subordinate (due to be promoted to the status of director at that very time) and passed it on to Mr Salomone, who he may well have regarded as better qualified that he was to consider its terms.
425. Mrs Di Matteo was also cross-examined about her role in the negotiation of the Acknowledgment. By contrast with Mr Casale and Mr Rosa Brunet, I found her evidence broadly credible. Taking her witness statement and cross-examination together, the gist of her evidence was as follows. She understood broadly what an assignment was, and her Italian translation of it was substantially that to be found in the Italian version of the Framework Contract which Mr Casale had been given. She said that in her experience she could only think of one previous example of an assignment of debts in an international transaction, but that she came across domestic debt assignments about 50 times a year, even though the practice was not popular. Her general expectation was that, if and when an assignment of debts to a bank was finalised, the assignee bank would notify the debtor that it had become the beneficiary of the debts, and that invoices in relation to the debts would require the debtor to pay the bank.
426. When shown the draft Acknowledgement and Notice, her reaction was that it was no more than an advance indication that, if a contemplated facility between Baskan Gida and the Banks went ahead, there would be an assignment, with a consequential notice to that effect from the Banks. She suggested the addition of the Same Account clause in order to ensure that, if any assignment did occur, her department would be able to make sure that invoices coming in from Baskan Gida led to the relevant debts being paid to the correct account. She took the trouble to ensure when reviewing the documents that Ferrero's bankers had the means of making direct payment to KBC, and a note to that effect in her manuscript appears on the draft of the Acknowledgment shown to her.
427. When shown the final versions of the Acknowledgment prior to signature, on 6^th December, she still regarded them as provisional documents, in particular because the Notice referred to closing as being "expected by mid-December", making it clear that the Facility remained in negotiation as between Baskan Gida and the Banks. Nonetheless she accepted that the final version of the Acknowledgment did on its face appear to contain an unqualified obligation upon each Ferrero signatory to make payments in respect of contracts with Baskan Gida entered into after 10^th December 2001 to the designated accounts, even though the documents still lacked account numbers and sort codes. At that stage, Mrs Di Matteo had not seen the Framework Contract.
428. Taking all that evidence together, I consider that round one of the negotiation of the Acknowledgement demonstrated Ferrero, and Mr Casale in particular, taking a thorough, careful and intelligent approach to the invitation from Baskan Gida that they should become involved in the drafting of important documentation required for the purposes of an imminent loan facility by the Banks to Baskan Gida. The result of Ferrero's involvement, led by Mr Casale, with light supervision from Mr Rosa Brunet and valuable assistance from Mrs Di Matteo, was that the originally unsuitable form of Acknowledgment containing commercially unsatisfactory undertakings and provisions affecting Ferrero's dealings with Baskan Gida, and failing to make appropriate distinctions between the different roles of Ferrero Industrial and the operating companies, was refined down to a much shorter pair of drafts with appropriate versions for the different Ferrero companies involved, committing Ferrero to payment into accounts to be designated in due course, once the Facility was completed, as the debtor in relation to debts to be assigned by Baskan Gida to the Banks. The obvious commercial purpose of the proffered documents, which in my judgment Mr Casale, Mr Rosa Brunet and Mrs Di Matteo all appreciated, was to assure Baskan Gida and the Banks that, if the Facility went ahead shortly thereafter, Ferrero would not thereafter refuse to make payments of its debts arising from contracts with Baskan Gida to accounts at KBC of which the details were to be communicated in due course. I consider it also to be clear that Ferrero, and Mr Casale in particular, appreciated that without such confirmation in advance of the completion of the Facility, it would not be likely to be completed, with severe financial consequences for Baskan Gida.
429. Ferrero's participation in the drafting process had achieved all the objectives which Mr Casale identified in his evidence, namely (1) ensuring that the appropriate persons signed different Acknowledgements for each participating Ferrero company, (2) that Ferrero undertook no commercially unattractive or restrictive obligations, such as promising not to change the terms upon which it did business with Baskan Gida (as required in Versions 1 and 2 of the draft Acknowledgment) and (3) ensuring that the treasury department's request for the Same Account clause was included. It is in my judgment incredible that Ferrero's part in those negotiations could have been achieved in the way that it was, if its principal negotiator Mr Casale had been so ignorant of the underlying purpose of the documentation as not to appreciate what the word 'assignment' meant.
430. In cross-examination Mr Wardell put to Mr Casale an altogether more sinister motive for his participation in the negotiation. He suggested that Mr Casale had deliberately engineered a situation where the forms of Acknowledgement and Notice sent to each of the Operating Defendants made no reference either to the Framework Contract, or to any assignment at all, whereas the Acknowledgement sent to and signed by Ferrero Industrial did so in terms. The innuendo was that, in order to assist in Baskan Gida's intended fraud, the Ferrero debtor companies would thereby be able with a clear conscience to continue paying Baskan Gida once the Facility was in place, unaware of any assignment, whereas Ferrero Industrial, which merely arranged contracts but incurred no debts, could thereby safely undertake obligations by reference both the Framework Contract and to an assignment of debts.
431. In addressing that serious allegation, it is necessary for me to ask why Mr Casale should have prevaricated so much about his role in this negotiation, and untruthfully denied an understanding both of the meaning of the word 'assignment', and of the obvious reference to the Framework Contract in the draft Acknowledgements to be signed by Ferrero Industrial. For reasons which will be apparent from a reading of the whole of this judgment, I have concluded that Mr Wardell's allegation that Mr Casale had a sinister intent in his participation in the drafting of these documents ought to be rejected. I am, quite simply, not persuaded that Ferrero or Mr Casale in particular had any inkling at this stage that Baskan Gida intended to defraud the Banks, rather than merely to obtain substantial finance from them at acceptable rates of interest. In my judgment, Mr Casale's lies and prevarication about his understanding of the draft Acknowledgments was the result of a corporate decision made by Ferrero in 2002 to deny any understanding of the meaning of an 'assignment', embarked upon because of its perception in 2002 that, by having paid Baskan Gida on contracts the benefit of which had in fact been assigned to the Banks, Ferrero had exposed itself to a grave risk of having to pay twice for the same goods. Whether it will have to pay twice is a matter for the Italian courts to decide.

The Early December Contracts

432. Between 4^th and 10^th December 2001, Ferrero Industrial arranged for three very large contracts for paste kernels to be placed by each of Ferrero Germany, France and Italy with Baskan Gida for an aggregate quantity of 6,432mt, for delivery during periods ending between March and June 2002. They were the first new orders (as opposed to re-grades) placed by Ferrero on Baskan Gida since mid-October. They were negotiated at exactly the same time as round one of the negotiation of the Acknowledgements, when Mr Casale knew that completion of Baskan Gida's negotiations for the facility with the Banks was imminent.
433. The Early December Contracts are roundly attacked by the Banks as being devoid of legitimate commercial purpose either for Ferrero or for Baskan Gida, from which they advanced the inference that they must have been put in place for an illegitimate purpose to the knowledge of Ferrero, namely the furtherance of Baskan Gida's fraud on the Banks. This allegation has led to the most detailed and minute examination of the background to, negotiation and performance of these contracts, which I must now describe.
434. The Banks' case about the Early December Contracts may be summarised as follows:

i) They were made at a time when Ferrero had no need for the goods ordered, either then, or during the extended periods for delivery.

ii) Any genuine need for the goods would have been satisfied by orders on Oltan Gida, which was up to date with its delivery obligations, rather than on Baskan Gida, which was still seriously in arrears.

iii) The contracts were made at prices which, if marked to market by reference to the prices for supplies of unprocessed nuts available at the time to Baskan Gida, would have resulted in Baskan Gida making very heavy losses. They therefore represented a bet on a future fall in the market during the contracted delivery periods which was both wildly improvident, unprecedented both as between Baskan Gida and Ferrero, and in the market generally. They were also wholly at variance with the Framework Contract.

iv) The contracts were dishonestly backdated.

v) Since, apart from 768mt delivered to Ferrero Italy, none of the quantities ordered were delivered before Baskan Gida went out of business, there was no genuine intention that supplies of hazelnuts should be delivered by Baskan Gida to Ferrero under these contracts.

435. Ferrero's response may be summarised as follows:

i) The contracts were proposed by Melih Baskan to Mr Casale, based upon the Baskans' view that the market would fall during the delivery period sufficiently to render the contracts reasonably profitable at the prices proposed.

ii) Acceptance of Melih Baskan's proposal was consistent with Ferrero's policy of building stocks in excess of production requirements and, although Mr Casale broadly shared the Baskans' view of the market, the prices offered were both attractive in comparison with prices available for immediate delivery, and fair as part of forward contracts.

iii) Ferrero played no part in any deliberate, let alone dishonest, backdating of the contracts, nor had any inkling that Baskan Gida might use them for any improper purpose in connection with the imminent Facility.

iv) Although forward contracts of this type were inconsistent with the normal course of dealing between the parties, they were not unprecedented, and Ferrero was under no obligation to confine its dealings with Baskan Gida to the type summarised in the Framework Contract.

v) As evidence of its wish to obtain the hazelnuts ordered under these contracts, Ferrero did in fact take delivery of all bar 528mt of the aggregate tonnage ordered, albeit from the combination of Baskan Gida, Aksu Gida and Baskan Yuksel, and the Ferrero Italy contract in respect of which there was a shortfall as to 768mt, performed by Baskan Gida itself within the periods specified in Ferrero's Purchase Order.

436. Turning first to the background, Ferrero's stock position (rounded down to the nearest 100mt) was as follows:

Ferrero Germany held 4,700mt, Ferrero France held 3,200mt and Ferrero Italy held 8,000mt. When the quantities ordered but yet to be delivered are taken into account, all three operating companies had amply sufficient stocks to satisfy their production requirements well into the new year, and in the case of Ferrero Germany, through to July 2002.

437. As for the relative performance of Oltan Gida and Baskan Gida, an interim report by Oltan Gida on 8^th December shows that of 17,595mt ordered by Ferrero France, Germany and Italy, only 1,296mt remained to be delivered. By contrast, of the 16,000mt ordered from Baskan Gida, some 5,232mt remained outstanding at the beginning of December, of which 504mt were supplied during the first week of that month. Of the amount outstanding, some 3,240mt were also overdue, in the sense of having not been delivered within the periods specified in the relevant Ferrero purchase orders. Oltan Gida had provided defective quality in some of its deliveries. Although therefore Baskan Gida's performance against orders fell significantly short of that of Oltan Gida, against a similar volume of orders, Baskan Gida's delivery rate during November had been consistently in excess of 500mt per week, reaching weekly amounts in excess of 800mt in early November. This was not as good as the rate achieved during the corresponding period in the previous year, but it was by no means so low as to be indicative of terminal financial difficulties affecting Baskan Gida.
438. I shall describe the making of each of the three Early December Contracts separately. The order by Ferrero Germany, for 2,016mt was the subject of a purchase order dated 6^th December calling for delivery until March/April 2002, pursuant to instructions given by email by Mr Casale to Mr Bolowich on the same day. On the same day Mr Casale faxed to Mr Bolowich the corresponding Baskan Gida contract, numbered 5088 and dated 4^th December 2001. The agreed price was €2,700 per mt FOT.
439. The Ferrero France order was also for 2,016mt, and was made by Purchase Order dated 7^th December, pursuant to instructions from Mr Casale to Ferrero France emailed on 6^th December, and reflected in a Baskan Gida contract numbered 5089 also dated 4^th December, also faxed by Mr Casale to France that day. As for delivery periods, the Ferrero France purchase order, pursuant to Mr Casale's emailed instructions, split the total tonnage into three unequal amounts, for delivery on specific dates in early April, May and June respectively. The Baskan Gida contract prescribed the same divisions of the tonnage and delivery months, but omitted any reference to specific delivery dates. The contract price was again €2,700 per mt FOT.
440. The Ferrero Italy order was for 2,400mt, to the higher Ferrero Italy specification (by comparison with France or Germany). The purchase order was dated 10^th December and the Baskan Gida contract, numbered 5093, was dated 7^th December. Both the Purchase Order and the contract specified a delivery period from December 2001 to March 2002, and a contract price of €2,739 per mt FOT.
441. For this order there are of course no emailed instructions from Mr Casale, since the purchase order was prepared in Alba, Italy in Mr Casale's office, but there survives an email dated 10^th December from Melih Baskan to Mr Casale enclosing a set of price calculations, based upon a re-worked PC dated 15^th October. The email message which ended: "looking forward to hear from you" suggested that the price had not prior to that moment been definitely agreed.
442. Turning to the oral evidence, the only one of Ferrero's witnesses who had any claim to a recollection of the negotiation of these contracts was Mr Casale, although he said that, at the time, he had discussed these orders with both Mr Rosa Brunet and Mr Do before agreeing them with Melih Baskan. He said that the proposal that Ferrero should buy large tonnages from Baskan Gida for long forward delivery came from Melih Baskan, without any prior prompting from Ferrero, in the sense of an indication that it was actively seeking further supplies. The Turkish hazelnut market had not fallen from the price which had, in October, led Ferrero substantially to withdraw from the market. Melih Baskan's proposal was, according to Mr Casale, expressly based upon his (Melih Baskan's) expectation (which he shared) that the market would fall substantially in Spring 2002, so that the price offered, although below what Ferrero would expect to pay for immediate delivery, was based upon Baskan Gida's expectation of being able to purchase the necessary unprocessed nuts in a falling market, at a price sufficiently lower to enable it to make a profit at the price proposed to Ferrero.
443. One of Soremartec's Golden Rules was that hazelnuts sourced from abroad should be delivered into Italy for storage by the end of February in any particular year but, as had occurred earlier in 2001, this was according to Mr Casale a rule as much favoured in the breach as in the observance. Thus, although the delivery periods proposed by Melih Baskan, to enable Baskan Gida to benefit from the anticipated falling market, involved deliveries outside the period specified by Soremartec, Mr Casale said that the offer of a price lower than that currently payable for immediate delivery made the proposal worth discussing with his superiors. He said that they (Mr Rosa Brunet and Mr Do) agreed that it was attractive to obtain at a reasonable price supplies sufficient to fulfil their stock increase policy from a plentiful crop, leaving Baskan Gida with the market risk (or "responsibility" as Mr Casale described it), so that Melih Baskan's offer was accepted for those reasons.
444. Implicit in Melih Baskan's explanation to Mr Casale of the reasons for his proposal was the absence of any intention on the part of Baskan Gida to cover its exposure to Ferrero by purchasing the necessary unprocessed hazelnuts at then current market prices. Although Mr Casale professed no knowledge of what if anything Baskan Gida might have had in stock, or on consignment, he acknowledged that what was being suggested was, in substance, a large scale short speculation by Baskan Gida, against a hoped for fall in the market which both the Baskans and he considered probable.
445. When asked in cross-examination why he had not sought to purchase the same or similar quantities from Oltan Gida, his response was that Baskan Gida had offered them, Oltan Gida had not, and he did not wish to push Oltan Gida for supplies at that time.
446. Mr Casale was pushed hard in cross-examination with the allegation that he had agreed dishonestly to go along with a backdating of each of the three Early December Contracts, in the case of the France and Germany contracts from agreement on 6^th December to a contract dated the 4^th and, in the case of the Italy contract from an agreement on 10^th December to a contract dated the 7^th. His response was first, that he could not be sure of the precise dates upon which he had negotiated orally with Melih Baskan but that, if they had taken place on the 6^th and 10^th, it would not have struck him as suspicious for the contracts to be dated slightly earlier, since he could discern no sinister purpose for the backdating. He resolutely denied that he agreed un-commercial, still less fictitious, contracts with a view to giving Baskan Gida any improper advantage either in the conclusion or operation of the, by then, imminent Facility.
447. There was a battle royal between the hazelnut experts as to the extent to which the pricing of the Early December Contracts was, marked to market against prices current at the date of the contracts, bound to be loss-making. Yet again, the minutiae of the costs analysis involved in processing and exporting the relevant nuts was trawled over in the greatest detail, although there was broad agreement as to the then current prices which Baskan Gida would have needed to pay, in order to obtain, in early December, the nuts necessary to perform those contracts. I find it unnecessary to determine the numerous issues arising from that contest, since it is common ground that, marked to market in early December 2001, the three contracts were all loss-making, and Mr Casale himself acknowledged that he appreciated it at the time, and that it was implicit in the basis upon which Melih Baskan proposed the transactions to him.
448. Mr Solak said that it was unprecedented in his experience for forward contracts in the Turkish hazelnut market to be offered at prices below the then spot price, let alone in the very large quantities represented by the Early December Contracts. In cross-examination he acknowledged however that a livrer contracts below current market price were not uncommon, particularly for export. An a livrer contract means a contract placed in one season for delivery in the following season, and an example was given from Mr Solak's own notes of an a livrer contract made in December 2001 for delivery in October to December 2002. Substantial discounts off spot market prices were identified as capable of being obtained in a livrer contracts which, as Mr Solak acknowledged, involved substantial short speculation in the hope of a fall in prices.
449. I have not been persuaded by the Banks' case that the Early December Contracts involved any knowing participation by Ferrero in Baskan Gida's plans to defraud them. Of course, it is now known that Baskan Gida used these contracts in support of applications for draw-down under the Facility, in furtherance of its fraud. It is a reasonable inference that the Baskans had this illicit purpose in mind when offering to supply Ferrero large quantities on forward delivery terms in early December. But the critical question is whether anything about the circumstances in which they were proposed and agreed leads to the inference that Ferrero was a knowing participant in the achievement of those purposes.
450. Taking each of the Banks' points about the supposedly suspicious nature of these contracts in turn, the fact that Ferrero had no immediate or even medium-term need for the nuts offered to meet its production requirements by no means leads to the conclusion, as alleged, that Ferrero did not need the goods. I accept the uniform evidence of Messrs Casale, Rosa Brunet and Do to the effect that Ferrero had formulated a policy to obtain nuts in excess of its production needs for the purpose of building up stocks in a plentiful season, and that the implementation of that policy was a sufficient commercial basis for Ferrero contracting on forward delivery terms for the large quantities represented by the Early December Contracts, providing that the prices offered were reasonable.
451. Secondly, while Oltan Gida had, in terms of meeting requested delivery dates, established a better track record than Baskan Gida in late 2001, I accept Mr Casale's evidence both that Baskan Gida's difficulties were in his view attributable to a significant extent to quality problems in relation to its heavier burden of rocher deliveries for Ferrero Italy compared with that imposed upon Oltan Gida, and that there was good reason for Ferrero not to seek a competing offer from Oltan Gida in relation to the large offer of forward delivery proposed by Melih Baskan in early December. Prices for immediate delivery were still at a level which had led Ferrero to wish to withdraw from the market, and there is no evidence that Oltan Gida would have been prepared to engage in short speculation of the type implicit in Baskan Gida's offer. Furthermore, the placing of substantial orders for immediate delivery would merely have increased the risk of a further hardening in prices, and cost much more than the price then being offered by Baskan Gida.
452. It is correct that the terms proposed for the Early December Contracts represented a substantial departure from the ordinary course of business between Ferrero and Baskan Gida, and in particular from the basis of trading set out in the Framework Contract and that, if marked to market in early December they would have been loss-making for Baskan Gida. All those factors were I have no doubt apparent to Ferrero at the time, as was the fact that, in offering forward delivery at prices below the spot price, Baskan Gida was taking on its shoulders the risk that the market would not turn down as the Baskans and Mr Casale expected. In fact, the market did turn down in 2002, but not in sufficient time to have turned the contracts from loss-making to profitability within the agreed delivery periods, save for the Ferrero France contract, which permitted delivery as late as June 2002.
453. Fourthly, I consider it probable, although not certain, that the contracts were backdated, by two days in relation to the Ferrero Germany and France contracts, and by the weekend following 7^th December in relation to the Ferrero Italy contract, i.e. by one working day. The case for a backdating of the Ferrero Italy contract is stronger than in relation to the other two, where the appearance may be attributable to Mr Casale just being late in sending ordering instructions to the two operating companies. Mr Casale's expectations, constantly pressed upon him by Ismet Baskan in his emails that week relating to the draft assignment documents, was that the last day for negotiation of the Facility was Friday 7^th December, and the documents themselves expressly contemplated closing on 10^th December, in the sense that that was the start date for contracts in respect of which Ferrero was invited to undertake to pay to the specified accounts.
454. The difficulty for the Banks in making play of this probability is that they were unable to advance and put to Mr Casale in cross-examination any consistent suspicious rationale for the backdating. It was suggested to Mr Casale that he must have had appreciated that the Ferrero Italy contract was backdated from 10^th to 7^th December, so as to enable Baskan Gida to receive payment under it without those payments being caught by the assignment to the Banks. He denied any such perception, and no similar analysis can be applied to the backdating of the Ferrero France and Germany contracts from 6^th to 4^th December. Furthermore, the Facility did not in fact complete, nor was there any assignment of debts, on 10^th December, nor was there any basis on that date for Mr Casale to suspect that the Baskans would dishonestly use that contract as a means of raising money with the Banks, and then (as in fact they did) invoice Ferrero on terms requesting payment otherwise than to the designated accounts. For reasons which I shall have to expand upon when addressing the honesty or otherwise of Mr Casale's repeated trade reference for Baskan Gida on 13^th December, I am not persuaded that anything about Baskan Gida's conduct until that date was sufficient to place Mr Casale on inquiry as to Baskan Gida's integrity, after a long and mutually satisfactory trading relationship spanning many years.
455. A further difficulty with the Banks' case that these contracts were a dishonest sham is the number of people within Ferrero who were informed about them, none of whom are alleged to have taken any part in the alleged conspiracy, or of having acted dishonestly. They included Mr Fontana, (head of Quality control at Ferrero Italy) who signed both the contract and the purchase order for Ferrero Italy, Mr Caroly of Ferrero France and Mr Bolowich of Ferrero Germany, as well as Messrs Altare, Bertone, Cicolin and Gagliasso within Ferrero Italy.
456. Finally, the fact that Baskan Gida went out of business when the bulk of the deliveries promised under the Early December Contracts remained outstanding by no means suggests that the Early December Contracts were shams, or that Ferrero did not desire to obtain the promised supplies. At the time when Baskan Gida ceased trading, none of the delivery periods agreed under those contracts had expired, albeit the period under the Ferrero Italy contract was about to. Over 700mt had by then been supplied to Ferrero Italy, and the bulk of the remainder was in due course supplied, on the same terms as to price, by Aksu Gida and Baskan Yuksel. It follows that I broadly accept as true Mr Casale's account of the negotiation and making of the Early December Contracts. The fact that all three of them were slightly backdated was, in my judgment, insufficient to put him, still less Ferrero, on notice of the improper purpose for which, by then, the Baskans had probably decided to deploy them.

The December Letter

457. The Banks made it a condition precedent to the completion of the Facility that Ferrero should re-confirm its trade reference for Baskan Gida, originally given in the 31^st July Letter. Ismet Baskan passed on the Banks' request to Mr Casale, enclosing a draft of the reference letter and a further copy of the Ferrero Industrial Acknowledgment, with the following text:

"I am pleased to tell you that today we will finally be signing the loan agreement with Bank of Tokyo Mitsubishi KBC Bank NV. I am sure that this step will help us to remain a good supplier to Ferrero Group.

As You will see on the documents, they have only asked that you re-sign the letters that you have already sent, to fit the date of closing, and have included the account numbers for the payments and agent bank name with the rest of Ferrero Group companies. I am attaching them. Please put them both on your stationery.

I will fax you the reference letter to you so you can see it is the same as what you sent on July 2001.

Thanks very much for your help."

Save for the change of date to 13^th December, the draft of the reference letter which Ismet Baskan asked Mr Casale to sign was word for word the same as the July Letter, the text of which I have already set out in full. Having obtained Mr Rosa Brunet's approval, Mr Casale had it copied onto Ferrero Industrial headed paper, signed it and returned it to Baskan Gida, for onward transmission to Mr Byles of BTM. I shall refer to it as the December Letter.

458. The Banks' case is that whatever may have been the position in July, Ferrero's repetition of the trade reference in the December Letter constituted a fraudulent, alternatively negligent, misrepresentation in a number of respects. In its most general terms, the allegation is that Ferrero painted a picture of a continuing mutually satisfactory trading relationship, whereas in truth, it had by December 2001 broken down due to Baskan Gida's default.
459. The Banks in their pleaded case identify the following specific representations as having been false:

i) Ferrero had experienced no significant difficulties with Baskan Gida.

ii) Baskan Gida was able to (and did) provide consistent shipments on a timely basis.

iii) The Ferrero operating companies expected to continue purchasing up to 30,000mt of hazelnuts from Baskan Gida annually.

iv) Contracts entered into between Ferrero and Baskan Gida were on a proper commercial basis.

v) The trading relationship between Ferrero and Baskan Gida had not materially changed since 31^st July, or since 15^th October.

In cross-examination of Mr Rosa Brunet, Mr Wardell sought to add another false representation, namely that Baskan Gida was, as at December 2001, Ferrero's largest single supplier of hazelnuts. Since this had neither been pleaded nor put to Mr Casale in his earlier cross-examination, I acceded to Mr Strauss's submission that this new element of an allegation of fraud could not at that late stage fairly be pursued.

460. It was not disputed that the 13^th December Letter contained the representations which I have numbered (i) to (iv), or that by virtue of its being re-dated from its July predecessor, those representations were put forward as being true as at December 2001, rather than merely as having been true as at July. As for (v), apart from those aspects of the trading relationship which the December Letter specifically confirmed, I consider that it contained no wider or general representation that there had not been any other change in the trading relationship. By December the Banks had been told in terms (in the form of the 45 Day Lie), that the trading relationship had changed from that described in the Framework Contract. Ferrero had by then specifically rejected the Banks' attempt to tie it down to the terms of the Framework Contract. Both the July and December Letters dealt with specific aspects of the trading relationship, and were silent about the rest.
461. It will be necessary for me to return to consider the legal consequences of the 13^th December Letter in due course. In this part of the my judgment I address the factual issues, namely:

i) Were any of the representations untrue as at 13^th December, and if so in what respect?

ii) Were such falsehoods as are thereby identified made fraudulently or negligently (assuming that in this non-English context, Ferrero Industrial owed the Banks something akin to a common law duty of care)?

462. It is convenient to deal with the first and second of the Banks' alleged false representations together, not least since one of the significant difficulties which it is alleged that Ferrero had experienced with Baskan Gida was its failure to make deliveries on a timely basis. In relation to these two representations, the Banks repeated all the matters relied upon in support of their allegation that the 31^st July Letter was untrue. I have rejected those, and need not address them again. In relation to the period after July, the Banks relied upon the substantial failures by Baskan Gida to make deliveries within the delivery periods specified in Ferrero's purchase orders, upon the apparent breakdown in the regular provision of DPRs and PCs since 15^th October, and the apparent discontinuation of the (by then) conventional basis of negotiating orders reflected in the Framework Contract. They relied also upon the fact that Ferrero had been obliged to re-grade a series of rocher orders into less demanding paste orders due to Baskan Gida's inability to provide rocher kernels of sufficient quality under those orders. They invited me to infer that, even if not part of a dishonest conspiracy, the early December forward contracts involved Ferrero ordering on different terms precisely because of Baskan Gida's inability to go on supplying in accordance with the procedure reflected in the Framework Contract. Finally, the Banks pointed to the promotion of Oltan Gida to approximate parity with Baskan Gida as a supplier of Turkish hazelnuts as evidence that Ferrero no longer regarded Baskan Gida as a reliable supplier, due to the difficulties which it had encountered in Baskan Gida's performance.
463. Mr Casale's and Mr Rosa Brunet's evidence on these matters may be summarised as follows. They both firmly denied that they considered that the long established and mutually beneficial close trading relationship between Ferrero and Baskan Gida had broken down, emphasising that they would not have committed Ferrero to purchases in excess of a further 6,000mt from Baskan Gida during the previous week if either of them thought it had broken down. They regarded the trade reference as a description of a long term relationship with Baskan Gida over many years, against which matters arising during the previous few months needed to be looked at in context. They both regarded Baskan Gida's performance during the peak part of the 2001 season (which by common consent, including that of the experts, ran until Christmas) as having been satisfactory, and attributed Baskan Gida's delays and difficulties in fulfilling the rocher orders as being attributable to quality problems which had emerged with the new crop in the course of processing, and which affected all Turkish suppliers, and in particular Oltan Gida, as much as Baskan Gida. As for paste orders, for example by Ferrero France, they attributed delays in delivery of these orders to Ferrero Industrial's insistence that Baskan Gida concentrate primarily on the maintenance of high quality rocher supplies to Ferrero Italy, and to storage capacity limitations in France.
464. As to the failure to make deliveries on a timely basis, the main thrust of Mr Casale's evidence was that he did not take much notice of Baskan Gida's rate of performance against delivery periods specified in Ferrero's purchase orders, being content to leave the timing of deliveries (and the making of the necessary detailed transport arrangements) to his subordinates, while he focused on the rate at which deliveries actually made satisfied both the operating companies' production requirements, and Ferrero Industrial's policy of increasing stocks in excess of production, both of which he said were during the period from August until December 2001 being satisfactorily met by the combination of Baskan Gida's and other suppliers' deliveries. The deliveries were therefore in his view timely for Ferrero purposes, whether or not strictly in accordance with the conventional delivery periods routinely inserted in Ferrero's purchase orders. Mr Bolowich's evidence on this point largely corroborated that of Mr Casale.
465. Mr Rosa Brunet, at one remove from the detail compared with Mr Casale, said that he was not concerned to keep himself aware of Baskan Gida's detailed performance as against the delivery periods in the purchase orders, but was unaware of any significant failures. Nonetheless, he acknowledged that to justify a statement that "consistent shipments … (were being made) … on a timely basis" would be to defend "something that is not to be defended" [day 41 p.83]. He said that, at the time, neither he nor as far as he knew Mr Casale was aware of the extent to which Baskan Gida had fallen behind with its deliveries.
466. As for the allegation that the conventional system whereby contracts were negotiated on the basis of DPRs and PCs as reflected in the Framework Contract had broken down due to Baskan Gida's default, Mr Casale said that the system had been interrupted in mid-October because rising prices had caused Ferrero to withdraw from the market, not because Baskan Gida had declined an invitation to offer supplies on the conventional basis. He said that as at 13^th December he had no perception that the conventional basis had been abandoned for good, even though the Early December Contracts had been negotiated on an entirely different basis, due to the particular state of the market, and the broad consensus between the Baskans and himself as to the likelihood of a fall in prices in 2002.
467. As to these issues, my conclusions are as follows. First, I have no difficulty in accepting Ferrero's evidence that Ferrero did not regard the mutually beneficial trading relationship as having broken down by 13^th December. My conclusion that both the Re-grade Contracts and the Early December Contracts had been made for genuine commercial reasons rather than as participation in any unlawful conspiracy to defraud the Banks leads inevitably to a conclusion that Ferrero continued to regard Baskan Gida as a valued supplier worthy of being given very substantial hazelnut orders, by 13^th December. Looking at the matter in the round therefore, Baskan Gida continued to be a supplier for which Ferrero Industrial believed that it could give a favourable trade reference, on the basis that a beneficial relationship established over many years had not been undermined by anything which had happened since July 2001. Furthermore, nothing in the detailed review of the relationship during that period leads me to the conclusion that a favourable trade reference could not reasonably have been given by Ferrero Industrial at that time. An attempt was made by Mr Wardell in cross-examination of Mr Casale to suggest that Ferrero had deliberately avoided calling the production manager at Ferrero France, because he would have regarded Baskan Gida's delays in supplying paste kernels pursuant to that operating company's orders as wholly unsatisfactory. I regard that assertion as speculative.
468. Turning therefore to the first two specific falsehoods alleged, the reality was in my judgment as follows. First, Baskan Gida had not made all its shipments to Ferrero on a timely basis, if "timely" is to be measured against the delivery periods set out in Ferrero's Purchase Orders. By 13^th December, 22,440mt had been ordered, 11,448mt had been delivered and, of the balance, 2,616mt were overdue, in the sense of not having been delivered within the periods specified in Ferrero's purchase orders. Of the same balance, 6,432mt were the subject matter of the forward orders placed during the previous week. I have included as overdue the quantities which were the subject matter of the November re-grades, because they provided for delivery in November 2001. Even if a November delivery seems unrealistically short in relation to contracts made on 26^th November, they arose as a result of Baskan Gida's default, rather than Ferrero's wish to postpone delivery. They are therefore properly to be regarded as overdue.
469. 2,600mt overdue as a proportion of 22,500 odd ordered or even 11,500 odd delivered may not seem very large, but some of the delays were substantial. Furthermore, Baskan Gida had enjoyed a substantial period since mid-October until early December, when Ferrero was placing no fresh orders, during which to catch up with its deliveries and, although it had made significant progress, it had not fully done so. Yet further, Oltan Gida's delivery performance in relation to a broadly equivalent amount of orders (prior to the Early December Orders) had been substantially better than that of Baskan Gida, albeit at the serious cost of having delivered a significant amount of sub-standard kernels due to the quality problems which undoubtedly were affecting the 2001 crop during that period. It is in that respect notable that Ferrero placed no further orders with Oltan Gida after November 2001, before June 2002, from which it is not difficult to infer that Ferrero's receipt of sub-standard kernels from Oltan Gida represented a real difficulty in those parties' trading relationship.
470. I have described Baskan Gida's rate of weekly deliveries during November as having been satisfactory, albeit less than achieved during the same period in the previous year, and less than Oltan Gida was achieving. In early December its weekly deliveries fell to only 312mt during the week beginning 3^rd December, and only 216mt during the week beginning 10^th December. It is fair to suppose, since transport arrangements would have by then to have been made in respect of that week's deliveries, that Ferrero Industrial had at least the means of knowing what Baskan Gida's delivery rate would be during that second week. Mr Casale said that he saw nothing surprising in a falling off of the delivery rate during late November and early December since, in 2001, it was the Ramadan period and a similar decline in weekly deliveries is observable during the Ramadan period in the previous year, although not to such a low level. I accept that evidence.
471. While Mr Casale may not have found it necessary himself to do a regular and detailed monitoring of Baskan Gida's rate of deliveries against the periods specified in Purchase Orders, at least not to the microscopic extent conducted during this trial, I consider that he significantly understated his degree of awareness at the time of this aspect of Baskan Gida's performance. There is no doubt that his office had the statistics available for such an analysis, and I consider it to be inconsistent with the thorough and careful way in which Mr Casale demonstrably carried out his duties for him to have been as detached from this aspect of the performance of Ferrero's major supplier as he claims. Furthermore, as the author of the 13^th December Letter and the person in Ferrero with the largest responsibility for the accuracy of its contents, I consider that when giving evidence he appreciated his vulnerability in the assertion as to the timeliness of Baskan Gida's deliveries, and therefore carried his assertions of detachment from the detail of Baskan Gida's performance beyond the boundaries of an honest lapse of recollection. By contrast, Mr Rosa Brunet probably was less aware than Mr Casale of the details of Baskan Gida's performance in terms of timeliness, and, with less personal vulnerability, felt able to acknowledge frankly that the assertion in the 13^th December Letter that Baskan Gida achieved a uniformly timely performance could not, once the detailed facts were investigated after the event, be justified. I agree with Mr Rosa Brunet. The statement in the 13^th December Letter that Baskan Gida provided consistent shipments on a timely basis did, as at that date, involve a misrepresentation.
472. The question whether this amounted to a "significant difficulty", experienced by Ferrero in its relationship with Baskan Gida is more difficult. As I have described, the evidence demonstrates that there were significant difficulties in the quality of the 2001 crop that adversely affected the performance of both Ferrero's main hazelnut suppliers in different ways. Baskan Gida was affected by the mould and aflatoxin problems in the Ordu crop, while Oltan Gida was also affected by the cimiciate problems affecting the Akcakoca variety. These problems began to manifest themselves in October and November, having been unnoticed at the beginning of the season. It led to Ferrero relaxing its cimiciate specification. The substance of Mr Casale's evidence, supported by Mr Rosa Brunet, was that whereas Oltan Gida chose to deal with that difficulty by sacrificing quality to timeliness, Baskan Gida appeared to have done the exact opposite, in particular by rejecting deliveries from its suppliers, and then having to find replacements. The results were that Oltan Gida's deliveries gave rise to a quality problem whereas Baskan Gida's gave rise to a timing problem. There is documentary evidence of rejections of Oltan Gida deliveries, in particular of rocher supplied to Ferrero Italy, and compensation being deducted from the cost of supplies in the following season. That evidence was not sufficient to permit any detailed comparative analysis, but Oltan Gida's grovelling written apology, coupled with the fact that Ferrero made no further orders from Oltan Gida after December, until mid 2002, while placing very large orders with Baskan Gida, does provide some corroboration. Even though I have rejected Mr Casale's attempt to distance himself from the detail of that problem, I accept his and Mr Rosa Brunet's evidence that neither of them regarded Baskan Gida as blameworthy for dealing with the quality problem in that way. When it is borne in mind that because of its stocking policy, Ferrero could endure considerable delays in delivery without suffering adverse consequences in terms of interruption to production, whereas a lack of quality in its raw materials threatened its reputation for producing a high quality end product, that attitude seems to me to have been entirely reasonable.
473. Mr Wardell tried to gain an advantage from Mr Do's unchallenged evidence that he was not told at the time of any delivery problem with Baskan Gida, on the basis that this showed that Mr Casale and Mr Rosa Brunet were acting dishonestly vis a vis Ferrero out of personally corrupt motives, and deceiving their superiors. In my judgment it shows the exact opposite, namely that Mr Casale and Mr Rosa Brunet did not regard the delays in delivery as a significant problem, worth reporting to Mr Do.
474. In fact, it appears from the evidence (in particular from Mrs Pay) that Baskan Gida's ever deepening financial problem was a major contributor to its failure to make timely deliveries of the 2001 crop, even if quality problems may also have been a significant contributory factor. Both Mr Casale and Mr Rosa Brunet denied any awareness as at 13^th December 2001 that financial embarrassment in the form of a shortage of working capital played any part in Baskan Gida's under-achievement in terms of timely deliveries. They were aware in general terms that Baskan Gida was seeking a more economical source of working capital than that available from its Turkish bankers, and that, prior to the beginning of the 2001 season, Baskan Gida had obtained significant additional finance (as is now known, from Mr Dahod). They also knew that Baskan Gida was at an advanced stage in obtaining finance from banks unaffected by the Turkish financial crisis, albeit that it had not been put in place prior to the beginning of the season. They had also been told much earlier in 2001 that Baskan Gida did not consider its business with Ferrero to be profitable, but, with good reason, they took that assertion with a pinch of salt. They also knew that, by comparison with the previous year, Baskan Gida's supplies to Ferrero were not assisted in terms of working capital by any pre-payments. There had also been market rumours that Baskan Gida was financially embarrassed but again, Ferrero were in my judgment entitled to place little weight upon them, not least since Baskan Gida had maintained steady, if unspectacular, deliveries of hazelnuts from the start of the 2001 season right through to early December, particularly if (as Mr Casale said in evidence), those rumours came from Baskan Gida's competitors.
475. I have not been persuaded that I should disbelieve the evidence of Mr Casale and Mr Rosa Brunet that they did not in 2001 attribute any underachievement by Baskan Gida in its rate of deliveries to a shortage of working capital, still less to any potentially terminal financial embarrassment. Baskan Gida's continuing search for non-Turkish lenders was consistently presented both to Ferrero, and indeed to potential lenders, as a search for cheaper working capital, rather than a project designed to make good a shortage of it. At its highest, Mr Casale and Mr Rosa Brunet may have suspected that a delay in getting its non-Turkish facility completed may have been a contributory factor, but I accept their evidence that they did not know or believe that it was.
476. A raw material quality problem, believed adversely to affect a supplier's performance is, on a fair reading of the 13^th December Letter, not necessarily to be regarded as a significant problem in the relationship with the supplier, if the customer genuinely believes that the supplier is dealing with the problem responsibly, rather than contributing to it by its own default. I accept the Ferrero witnesses' evidence that this is the attitude which they took towards Baskan Gida's under-performance in terms of timeliness. It did not in fact cause Ferrero any practical difficulties of its own, since first, its freezing programme was satisfied from the aggregate of its rocher deliveries from all its suppliers, and since the delayed deliveries were still received well in time to avoid any production interruption, and to constitute stocks in hand in fulfilment of Ferrero Industrial's stock policy. At the most, it meant that deliveries were to an extent received later than prescribed by the Soremartec Golden Rules, but this had occurred in previous years and I can well understand that from the procurement perspective of Ferrero Industrial, it was envisaged as preferable to obtain good quality hazelnuts late, rather than defective quality hazelnuts in good time.
477. It follows in my judgment that the first representation, that Ferrero had experienced no significant difficulties with Baskan Gida during its lengthy trading relationship, was not in substance untrue. Ferrero's belief was that the under-performance in terms of timeliness was attributable to a factor for which Baskan Gida could not fairly be blamed. It was not therefore a relationship difficulty in the sense referred to in the reference.
478. The third alleged misrepresentation (that Ferrero expected to continue purchasing up to 30,000mt of hazelnuts from Baskan Gida annually) was, in terms of its materiality to the Banks, perhaps the most important of all. By 13^th December the Ferrero Operating Defendants had already ordered 22,260mt from Baskan Gida and more had been ordered by other operating companies, and had received in the form of Melih Baskan's November Letter, an invitation to place orders well in excess of 30,000mt. Its policy for the 2001 season was to continue, but not significantly exceed, the levels of orders placed with Baskan Gida in previous years.
479. The Banks' case is that, by 13^th December, Baskan Gida's under-performance was such that Mr Casale could not honestly have believed that Baskan Gida would be able to deliver 30,000mt during the 2001 season, such that he could not have had any genuine expectation that Ferrero's orders would reach that level. The Banks' pleaded case suggested that this representation would have been untrue even if Mr Casale had believed that Baskan Gida could deliver at that level only with the assistance with finance from the Banks.
480. In my judgment this representation in the 13^th December Letter was true. I consider that Mr Casale and Mr Rosa Brunet probably did have it in mind in expecting to maintain orders to Baskan Gida up to 30,000mt for the 2001 season, that by 13^th December Baskan Gida was on the point of receiving non-Turkish bank finance, for the purposes of providing the working capital necessary not merely to fulfil the early December orders, but further orders at the level which Melih Baskan had invited in the November Letter. But that expectation of a successful outcome to the Baskans' search for a cheaper source of finance in no way undermines the truth of the representation made in the letter. In the event, Ferrero did order 30,000mt from Baskan Gida during the 2001 season, when orders by all the operating companies are taken into account, and took delivery of only slightly less, albeit from the combination of Baskan Gida and its two successor companies.
481. In reality, the Banks' attack on the truthfulness of this representation depended upon their case, which I have in substance rejected, that Ferrero knew by mid-December 2001 that Baskan Gida was in terminal financial difficulties, and that Ferrero had placed the early December orders by way of participation in the Baskans planned fraud against the Banks, rather than for any genuine commercial purpose.
482. For the same reason, I also reject the Banks' case that the implied representation that contracts entered into between Ferrero and Baskan Gida were on a proper commercial basis was untrue. Of course, the Re-grade Contracts and the Early December Contracts were placed on a quite different basis from that reflected in the procedure recorded in the Framework Contract, but that did not of itself render any of those contracts improper or un-commercial from Ferrero's perspective, even if, as is probably the case, Baskan Gida had an improper purpose in mind in relation to the early December contracts, since I have concluded that Ferrero was entirely unaware of it.
483. In conclusion therefore, the only misrepresentation to be found in the 13^th December Letter was that Baskan Gida made its shipments on a timely basis, the letter containing otherwise a fair description of relevant facts, and of Ferrero's beliefs and expectations in relation to its business with Baskan Gida. The question remains whether Mr Casale made the representation on behalf of Ferrero Industrial fraudulently, negligently or innocently.
484. The question whether Mr Casale acted dishonestly inevitably raises an issue as to motive. Plainly, it was in Ferrero's interests that Baskan Gida should obtain, rather than be deprived of, its long-sought non-Turkish finance. For reasons which will become apparent when I deal with the events of 2002, I have concluded that Ferrero had a powerful desire to maintain Baskan Gida as a major supplier of Turkish hazelnuts, not merely because of its own deserved reputation for producing high quality nuts meeting Ferrero's exacting specifications, but also because the Baskans' very long experience of the market had involved building up supply chains, stretching back through crackers, manavs and growers, whose own activities Ferrero assumed contributed in no small measure to the high quality of the end product. Ferrero had, of course, no easy means of access to, or means of supporting, the entities and persons in the supply chain behind Baskan Gida, but could preserve the chain as well as Baskan Gida by assisting in the achievement by Baskan Gida of its own commercial objectives. When faced with an invitation to provide a trade reference, Ferrero therefore had an understandable motive for painting its supplier in as rosy a hue as was consistent with the objective facts, but not, in my judgment a sufficient motive for telling an outright lie.
485. I bear in mind my conclusion that Mr Casale knew in December 2001 more than he has been prepared to admit in this trial about Baskan Gida's under-performance in terms of timeliness, and the fact that Mr Casale has, upon specific occasions during the giving of his evidence on oath, been prepared to lie in relation to specific matters where he and/or Ferrero perceive that they have a serious vulnerability.
486. Mr Casale's evidence was that when presented with the draft 13^th December Letter, he reminded himself that it was a straight repetition of what he had been prepared to sign for in July, asked himself whether anything had happened to shake his confidence in Baskan Gida as a reliable supplier, or Ferrero's intention to continue to use Baskan Gida provided its quality standards were maintained, concluded that nothing had happened to that effect, and therefore signed after obtaining Mr Rosa Brunet's approval. I accept that evidence. In my judgment Mr Casale failed to focus on the fact that, read objectively, a reference to timely deliveries would ordinarily be taken as meaning timely, by reference to delivery periods specified in orders and contracts, rather than, as he intended, merely by reference to Ferrero's production requirements and stocking policy. I have come to the conclusion that this inaccuracy in an otherwise fair and genuine trade reference is to be attributed to negligence rather than dishonesty on Mr Casale's part. A careful appraisal of the December Letter, by reference to the meaning which the Banks might be expected to place upon the statement about timeliness, ought to have made him realise that his intended meaning was not that which, objectively, that statement conveyed.
487. There is an issue whether the Banks received the 13^th December Letter before the facility was granted. No copy of it (other than one known to have been received much later) is to be found in the Banks' files, and no mention is made of it in the internal audit investigation which followed the discovery of the fraud, although the July Letter is mentioned.
488. The only Banks' witness who spoke to its receipt was Mr Byles. Although he could not recall seeing it at the time, he said that he did recall being satisfied that the condition precedent to the grant of the Facility, namely an up to date confirmation by Ferrero of its July trade reference, had been received. He relied upon an exchange of emails between the Banks' lawyers, Lovells, Baskan Gida's lawyers White & Case and himself on 21^st December (the last Friday before Christmas) which started with a note from White & Case that the condition precedent remained outstanding, and which concluded at 8.52 p.m. with an email from Lovells to White & Case stating that the condition had been satisfied. There was no evidence of any kind from KBC on the issue.
489. Mr Byles speculated that for Lovells to have confirmed that the condition precedent was satisfied, either he or they must have seen the December Letter, or perhaps been told by PRI that it had been received from Ferrero. This is thin evidence for receipt by the Banks of a document upon the detail of which they now allege that they relied at the time, but on a narrow balance I am just persuaded that, although not studied with any care, it was either received or, more likely, that PRI confirmed that Ferrero had signed it.

Negotiating the Acknowledgement of Assignment – Round Two

490. Ismet Baskan's email enclosing the December Letter also enclosed a further form of Acknowledgement of Assignment for signature, pointing out, correctly, that the names of the Ferrero operating companies had been added to the Acknowledgment in the form which had already been signed by Ferrero Industrial and, incorrectly, that the sort codes and account numbers for the specified accounts had been completed. In fact, Ismet Baskan attached the wrong document, in which the account numbers and sort codes remained blank. Furthermore, it omitted the Same Account clause which, at Mrs Di Matteo's suggestion, had been added during round one of the negotiations.
491. There followed a flurry of communications, both oral and electronic, between Ferrero Industrial and Baskan Gida during the rest of that day during which, I have no doubt mainly at Mr Casale's instigation, the various amendments and additions in the earlier draft were put right so that, just after 6.30 that evening, Mr Casale faxed to Baskan Gida from Ferrero Industrial a signed Acknowledgement by Ferrero Industrial addressed to Baskan Gida as agent for KBC and post-dated 17^th December 2001, in the following terms:

"Dear Sirs:

ACKNOWLEDGMENT OF NOTICE OF ASSIGNMENT OF GENERAL TERMS AND CONDITIONS FOR SALE OF HAZELNUTS

We refer to the Framework of General Terms and Conditions for the Sale of Hazelnuts dated 15^th October 2001 between ourselves and Baskan Gida Sanayii Ve Pazarlama A.S.

We hereby acknowledge receipt of the notice from Baskan Gida Sanayii Ve Pazarlama A.S., of which the attached is a duplicate, under which it is informed us that it has assigned to KBC Bank NV its rights in the Framework of General Terms and Conditions, and provide the following information:

We undertake to pay any amounts due by us under each sales invoice dated on or after December 17^th 2001 to the following bank accounts:"

There then followed two bank accounts, both at KBC in London with sort codes and account numbers, being identified as "Favour of: Akbank T.A.S. as Collecting agent for Baskan Gida Sanayii Ve Pazarlama A.S. Finally, the document concludes:

"Note: Same account information will appear on each invoice supplied by Baskan Gida to Ferrero."

492. In fact, on 13^th December when this document was signed by Ferrero Industrial and sent to Baskan Gida, there had still been no completion of the Facility, and, above all, no assignment of debts by Baskan Gida to the Banks. None of that occurred until, at the earliest, the following day. Nor had Ferrero been sent a notice of assignment of the type referred to in the Acknowledgement, nor was any duplicate of it attached to the Acknowledgement. In short, Ferrero Industrial signed an acknowledgement of a notice which it had not received, of an assignment which had not yet taken place. Furthermore, the acknowledgment made no reference to debts, but merely to Baskan Gida's rights under the Framework Contract, which had neither been made by, nor sent to the Ferrero operating companies, which, by contrast with Ferrero Industrial, were the companies which became debtors under contracts for the supply of hazelnuts.
493. One of Ismet Baskan's emails during 13^th December (sent at 4.05 p.m.) drew Mr Casale's attention specifically to the fact that the Assignment of which Ferrero was being asked to acknowledge having received notice was an assignment specifically permitted by clause 9 of the Framework Contract. Mr Casale continued, unconvincingly in my judgment, to deny that he made any connection between the Acknowledgement and the Framework Contract at the time, or that he had studied this particular email coming from Ismet Baskan in any detail. Again, I consider that to be wholly inconsistent with Mr Casale's careful approach to documentation required of Ferrero in connection with Baskan Gida's Facility.
494. In relation to round one of the negotiation for the Acknowledgement, Mr Wardell suggested in cross-examination, and Mr Casale denied, that he had quite deliberately arranged for the Acknowledgement to be framed in such a way that it could be signed by Ferrero Industrial without any consequences at all, knowing as he did that the operating companies rather than Ferrero Industrial would be the debtors in relation to any assignment, and that they were not being sent copies of the Acknowledgment to sign. Again, in my judgment this paints Mr Casale too much in the role of a Machiavelli. The evidence of the second round of the negotiations suggests that Mr Casale supplied to Baskan Gida the names of the individuals at each of the operating companies to whom notices of assignment and Acknowledgements should be sent for their signature. These were Mr Rosa Brunet for Ferrero Italy, Mr Bolowich for Ferrero Germany and Mr Aimaro for Ferrero France. Their names can be seen used by the Banks in communications sent direct to the operating companies on 21^st December, to which I shall refer in due course. In my judgment Mr Casale fully expected that notices of assignment and Acknowledgements would be sent to each of the operating companies. In insisting that the Ferrero Industrial Acknowledgment should be given on its own behalf, and not on behalf of the operating companies, Mr Casale was doing no more than he had done in round one of the negotiations, and for no sinister purpose.

Completion of the Facility

495. The Facility actually completed on 19^th December, but the main documents regulating the Facility were dated 14^th December 2001. They were:

i) The Facility Agreement between Baskan Gida and the Banks.

ii) A charge of the account ("the Collection Account") of Akbank at KBC into which all repayments under the Facility were to be made ("the Account Charge") made between Baskan Gida, KBC and Akbank.

iii) The Master Assignment Agreement between Baskan Gida and KBC.

iv) A Master Pledge Agreement between the same parties.

v) A Master Trust Receipt, also between the same parties.

On the 19^th and 20^th December respectively, there were also signed:

vi) Tenancy agreements between Baskan Gida and SGS, in relation to warehouses to be used by SGS for the custody of pledged hazelnuts.

vii) A Collateral Management Agreement between SGS, KBC and Baskan Gida, regulating SGS's role and responsibilities in relation to the custody, documentation and reporting arising from the pledge of hazelnuts in the warehouses rented to it.

A significant element of the documentary structure of the Facility was the Framework Contract which had been signed between Baskan Gida and Ferrero Industrial on 15^th October 2001.

496. It will be necessary for me to describe the fine detail of parts of this documentary structure when addressing the allegations of contributory negligence later in this judgment. For present purposes it is necessary only to describe how, in accordance with this structure, the Facility was intended to operate in practice. The Facility Agreement committed the Banks to lending to Baskan Gida in Dollars or Euros up to specified limits for the purpose of financing the purchase, processing and export of hazelnuts by Baskan Gida to specified buyers ("Buyers"), including the Ferrero companies. The limits were US$30 million until the end of December 2001, US$35 million during January 2002, US$25 million during February to June inclusive and US$20 million during July, the Facility expiring at the end of July.
497. Advances were to be made by BTM (as Issuing Bank on behalf of the Banks) issuing letters of credit to Akbank in Istanbul, and Akbank making payments to Baskan Gida. Repayments were intended to come from the assigned proceeds of sale of hazelnuts to Baskan Gida's Buyers, by payment being made direct to Akbank, acting as collecting agent for Baskan Gida, into the Collection Account of Akbank at KBC, with KBC acting as agent for the Banks.
498. Security for repayment was to be provided by:

i) A charge on the Collection Account in favour of the Banks (by the Account Charge).

ii) An assignment of Baskan Gida's debts owed by Buyers (in the case of Ferrero, by the Ferrero operating companies), and an assignment of Baskan Gida's rights under the Framework Contract against Ferrero Industrial, in both cases to the Banks, under the Master Assignment Agreement.

iii) A Turkish law pledge of hazelnuts acquired for processing by Baskan Gida, once stored in warehouses let to and operated by SGS, pursuant to the Master Pledge Agreement and the Collateral Management Agreement.

iv) An English law declaration of trust by Baskan Gida in favour of KBC, as agent for the Banks, in relation to hazelnuts received from pledge from time to time for processing, pending delivery to Buyers.

v) Promissory notes by Baskan Gida, guaranteed by senior members of the Baskan family.

499. All the Facility documents, other than the Master Pledge Agreement, the warehouse rental agreements and the Framework Contract were expressed to be governed by English law.
500. Conditions precedent for any lending under the Facility Agreement (as set out in Schedule 1 thereto) included the provision to KBC by Baskan Gida of its audited accounts for the period ending 30^th June 2001, duly executed documents including both Notices and Acknowledgements of Assignment to and from each Buyer and (in relation to the Framework Contract) from Ferrero Industrial.
501. The machinery for obtaining advances under the Facility was triggered in relation to any particular advance sought by Baskan Gida by its delivery to BTM (as Issuing Bank) and KBC (as Agent) of a signed Utilisation Request, together with a copy of a signed purchase order from one of the listed Buyers, identifying the quantity and quality of hazelnuts required, the price payable, the delivery date and method of delivery, and the payment date. Subject to a right to refuse in the event of any default by Baskan Gida, BTM would then issue a letter of credit to Akbank for the amount requested. The amount payable was limited in every case so as to ensure that the aggregate amount of that and any earlier letter of credit would not lead to the aggregate advances exceeding 90% of the total amounts payable but as yet unpaid by Buyers under qualifying purchase orders and 100% of the purchase price payable by Baskan Gida in respect of its purchases of hazelnuts required for delivery to Buyers.
502. Draw-down by Baskan Gida under letters of credit issued to Akbank by BTM required Baskan Gida to provide, in addition, inspection certificates and warehouse receipts from SGS in relation to pledged hazelnuts, together with an appropriate pledge agreement, and suppliers' invoices addressed to Baskan Gida in respect of the hazelnuts to be pledged, together with copy invoices to the appropriate Buyer, certified as delivered to the Buyer for hazelnuts at a price exceeding 111% of the amount drawn under the letter of credit, such invoices to be endorsed with a notice of assignment to KBC and a direction to pay to the Collection Account, in a prescribed form set out on page 59 (in Schedule 6) of the Facility Agreement. The prescribed form of notice of assignment to be endorsed on every invoice by Baskan Gida to a qualifying buyer was as follows:

"We confirm that pursuant to a master assignment agreement between KBC Bank NV ("the Agent") and ourselves dated [date], we have assigned to the Agent all of our rights, benefit and interest in this contract, including any money payable to us under or in connection with it.

We hereby irrevocably and unconditionally authorise and instruct you to pay any amounts becoming due and payable by you to us under or in connection with this contract to

[Insert Collection Account details]

or to such other accounts as the Agent may nominate from time to time.

This authorisation and instruction may not be revoked or varied without the prior written consent of the Agent."

I shall refer to this form of notice as "the Prescribed Endorsement". It is specified in substantially the same form in the Schedule to the Master Assignment Agreement.

503. The Banks' security over the hazelnuts between purchase by Baskan Gida and onward sale to Buyers was constituted by the combination of the Master Pledge Agreement, individual pledges signed in relation to each deposit of hazelnuts with SGS, the terms of the Collateral Management Agreement, and the Master Trust Receipt. I shall have to consider each of these in more detail when addressing the Banks' proprietary claim. For the purposes of my findings of fact, it is necessary to outline the functions imposed upon SGS, under the Collateral Management Agreement. By way of background, the structure put in place reflected the understanding of the Banks, on advice, that the effectiveness of a pledge of goods in Turkey depended critically upon being able to identify defined goods being held in the possession of the pledgee or its agent. Loss of possession, in particular to the pledgor, gave rise to a threat (the extent of which was the main point at issue between the parties' Turkish law experts) of loss of security. The Banks recognised that the need for Baskan Gida to process hazelnuts between purchase from its suppliers and delivery to Buyers meant that there would be an inevitable loss of possession to Baskan Gida as pledgor during the processing period, and the perceived vulnerability of any pledge expressed to persist during that period was meant to be buttressed, to the limited extent possible, by the Master Trust Receipt.
504. By the Collateral Management Agreement, SGS were appointed as custodians of relevant hazelnuts by Baskan Gida on SGS's standard Warehousing General Terms and Conditions. The agreement envisaged that three warehouses would be leased for that purpose to SGS, at Giresun, Hendek and Bulancak. In the event, only one warehouse, namely Warehouse 2 at Giresun was actually used for the storage of pledged hazelnuts.
505. By Article III and Appendix II, SGS undertook to provide stated inspection and management services in relation to the pledged hazelnuts on behalf of KBC as Agent for the Banks. Those services included:

Inspection

( Supervision of discharge from truck into SGS controlled warehouse.

( Tally of goods entering and leaving the SGS warehouse.

( The provision of inspection certificates on request by KBC or Baskan Gida, as to quantity and quality of goods stored.

Management

( Receipt and storage of goods in SGS warehouse.

( Issuing warehouse receipts, signing pledge agreements on behalf of KBC.

( Regular stock control.

( Receiving, verifying and acting upon release instructions issued by KBC.

( Authorising and checking physical handling of operations and delivery from SGS warehouse, including supervision of FOT deliveries.

( Regular weekly reports detailing tonnage and description of stock in store, and stock in process.

506. Although the description of services to be performed by SGS in Appendix II to the Collateral Management Agreement might be interpreted as involving SGS in supervising the processing of hazelnuts released from the warehouse and their delivery FOT Giresun to Buyers such as Ferrero thereafter, it became apparent from Mr Aslan's evidence that SGS's supervision of any particular consignment of hazelnuts ended once they were weighed upon leaving Warehouse 2. Mr Aslan regarded it as no part of his duty to ascertain where they were to go (i.e. whether into the processing factory or elsewhere), still less did he have any responsibility to monitor the processing or the delivery of processed hazelnuts FOT Giresun to Ferrero or anyone else.
507. In addition to imposing an express trust of hazelnuts during processing, between Baskan Gida and the Banks, the Master Trust Receipt contained a series of irrevocable undertakings by Baskan Gida in connection with their use of hazelnuts released from SGS warehouse, and the terms and proceeds of any sale thereof. Again, the effectiveness or otherwise of this document in creating proprietary rights in relation to the hazelnuts is something I shall review later in this judgment.
508. The Facility Agreement regulated the process whereby Baskan Gida obtained release of pledged hazelnuts from the SGS warehouse in Schedule 10, containing Operating Procedures. The relevant paragraph reads as follows:

"Release of Hazelnuts

No release of Hazelnuts from SGS will be permitted without the written authorisation of the Agent to SGS. Prior to such release the Borrower will send to the Agent a release request, in a format acceptable to the Agent, accompanied by a duly completed schedule to the Master Trust Receipt and a copy of the adjusted Invoice which must domicile payment to the Collection Account. Upon receipt of these documents, the Agent will send written release authorisation to SGS."

The expression "domicile payment to the Collection Account" was an oblique reference to the requirement that each Adjusted Invoice contained the Prescribed Endorsement.

509. Consistent with the 45 Day Lie, the Facility Agreement provided for repayment of advances made for the purchase of hazelnuts for on-sale to Ferrero 45 days after advance. In relation to purchases for on-sale to any other Buyer, 60 days was allowed for repayment. In the event, all the advances made under the Facility were drawn down upon the basis that they were to finance Ferrero business. In the event, the 45 day window between advance and repayment proved to be sufficient for Baskan Gida comprehensively to defraud the Banks of the bulk of their advances, before a failure to repay started bells ringing at the Banks.

Notifying Ferrero of the Assignments

510. It will be recalled that all that the Banks had obtained prior to the completion of the Facility and the execution of the Master Assignment Agreement were acknowledgements from Ferrero Industrial and the operating companies in what Mrs Di Matteo correctly described as an essentially provisional form, coupled with a post-dated acknowledgement from Ferrero Industrial of the purported receipt (which had not by then occurred) of a notice of an assignment of the Framework Contract which had not yet taken place.
511. Since it was a condition precedent for any advance under the Facility that Baskan Gida should provide copies of all the requisite notices of assignment, and signed Acknowledgements, the Banks left it to Baskan Gida to carry out this part of the process. On Monday 17^th December Baskan Gida signed a form of Notice of Assignment of Framework Contract letter for sending in identical form to Ferrero Industrial and to each of the operating companies, and copies of it, in English and Turkish, were sent to Baskan Gida's attorneys White & Case for onward transmission to Lovells, the Banks' solicitors. They passed it on to Lovells on 18^th December, together with a copy of the Ferrero Industrial Acknowledgement, dated 17^th December, but in fact signed on 13^th December by Mr Rosa Brunet, as I have described.
512. Contrary to the impression thereby created, Baskan Gida's Notice of Assignment letter had not been sent to any Ferrero company, and never was. On the contrary, Baskan Gida merely arranged for Acknowledgements purporting to come from each of the operating companies to have signatures forged on them, before passing them to White & Case who forwarded them on to Lovells on 20^th December, in the case of all the operating companies other than Ferrero Germany, and on 21^st December for the latter.
513. It is a matter for speculation why Baskan Gida thought fit to forge the Acknowledgements which they needed from the Ferrero Operating Defendants. Nothing in the two rounds of negotiations between Baskan Gida and Ferrero Industrial, through Mr Casale, during the course of which the Acknowledgments reached their final form, could have given rise to any expectation or concern that the Ferrero Operating Defendants might refuse to sign such documents. The versions forged by Baskan Gida were all in a form which Mr Casale had already had approved, and in particular contained the Same Account clause requested by Mrs Di Matteo. There survives a fax from Mr Politis of PRI to Mr Byles and Mr Lewis dated 21^st December in which, on Baskan Gida's behalf, he objects to a suggestion that the first draw-down should only follow the Ferrero companies' further signature to revised Acknowledgements correcting a typographical error in the account number, stating that "Baskan feels that we have strained the relationship enough on this subject".
514. In my judgment, Baskan Gida probably forged the Acknowledgments required from the Operating Defendants simply to save time. Having failed to obtain Mr Casale's agreement to sign on behalf of all the relevant Ferrero companies, Baskan simply avoided by the forgeries having to wait until the relevant officials from companies situated all round Europe complied with a routine request, a process which, as is evident from the time it took Ferrero France to reply to Mr Lewis's fax, might well have been overtaken by the Christmas holidays.
515. In the meantime, someone on the Banks' side doing due diligence as to satisfaction by Baskan Gida of all conditions precedent noticed that the Collection Account details for payments in Euros specified in the Acknowledgement letters contained a typographical error in the account number. Lovells sent a draft letter for Mr Lewis at KBC to send directly to each of the Ferrero Defendants, addressed the individuals already identified for the purposes of any such communications by Mr Casale, namely himself at Ferrero Industrial, Mr Rosa Brunet at Ferrero Italy, Mr Bolowich at Ferrero Germany and Mr Aimaro at Ferrero France. Mr Lewis duly faxed letters supplying the correct account number to each of those individuals on 21^st December in the following terms:

"Dear Sirs

Baskan Gida Sanayii Ve Pazarlama A.S.

ACKNOWLEDGEMENT OF ASSIGNMENT OF GENERAL TERMS AND CONDITIONS FOR SALE OF HAZELNUTS AND ACKNOWLEDGEMENT OF MASTER ASSIGNMENT AGREEMENT ("THE ACKNOWLEDGMENTS")

Thank you for the Acknowledgment which we have now received.

Unfortunately, there was a typographical error in the account for Euro payments.

The correct account details for Euro payments are:

[the correct details are then given]

Please ensure that all Euro payments made in accordance with your undertakings set out in the relevant Acknowledgment are made to the above account. We should be grateful if you could countersign a copy of this facsimile and fax it to us."

516. It is common ground that all four Ferrero Defendants received copies of Mr Lewis's fax. Only Ferrero France sent a countersigned copy, and only on 10^th January 2002.
517. In his witness statement Mr Casale attributed his own Christmas holiday as the reason why he did not deal promptly with Mr Lewis's fax to Ferrero Industrial. He added that, coming directly from the Banks rather than Baskan Gida, he preferred not to enter into direct communication, and received Mr Rosa Brunet's approval not to reply. In Italian criminal proceedings early in December 2003 to which I shall have to return, Ferrero Industrial alleged that Mr Lewis's fax had been left unanswered because Ferrero Industrial did not anticipate having to make any payments for hazelnut supplies by Baskan Gida. Mr Rosa Brunet in his evidence said that he could not recall seeing Mr Lewis's fax until after matters became contentious in 2002.
518. In my judgment the inconsistencies between Ferrero Industrial's various explanations for not replying to Mr Lewis's fax arise because, apart from Mr Casale's absence for his Christmas holidays from 21^st December onwards, the remainder are reconstructions after the event. I consider that the failure of all the Ferrero companies except for Ferrero France to reply to his fax is because no-one applied any pressure for replies, the Banks having decided before Christmas not to insist on a satisfactory reply as a condition precedent to beginning substantial lending under the Facility. I draw no adverse inferences as to the integrity of Ferrero, Mr Rosa Brunet or Mr Casale from that failure.

The Re-numbered December Contracts

519. On 19^th December, at Melih Baskan's request to Mr Casale, he and Mr Rosa Brunet signed four newly numbered Baskan Gida contracts on behalf of Ferrero Italy, each of which related to the amounts outstanding for delivery under four existing contracts. The aggregate amount outstanding, to which these four newly numbered contracts related, was 1,224mt.
520. The Banks' case is that by signing these re-numbered contracts Ferrero Italy through Mr Casale dishonestly assisted Baskan Gida in deceiving the Banks, by using the re-numbered contracts for the purposes of drawing down finance under the Facility, the deception being that the contracts represented fresh orders from Ferrero, rather than stale orders, against which Baskan Gida had long since failed to make timely delivery.
521. In order to make the parties' cases about the re-numbered December contracts intelligible, it is necessary first to describe them, and their predecessors, in a little detail. The telephone agreements between Mr Casale and Melih Baskan which led to the original purchase orders and contracts took place between 9^th September and 1^st October 2001. They led to purchase orders dated respectively the 11^th, 13^th and 21^st September and 1^st October, and Baskan Gida contracts numbered respectively 5063 (dated 9^th September), 5065 (dated 13^th September), 5071 (dated 21^st September) and 5074 (dated 1^st October). All four contracts were for rocher kernels. The purchase order delivery periods were as to the first two by the end of September and as to the remaining two by the end of October. All four Baskan Gida contracts were for delivery 'prompt from origin'.
522. In late November, the amounts outstanding under Baskan Gida contracts 5063 and 5065 were re-graded to paste. The outstanding deliveries under all four contracts were made the subject of new Baskan Gida contracts numbered 5097, 5096, 5094 and 5095 respectively, each bearing dates the 19^th December, with delivery periods December 01-February 02, December 01-February 02, December 01-March 02 and December 01-February 02 respectively. No new purchase orders were prepared.
523. Following receipt of copies of each contract signed for Ferrero Italy by Mr Casale and Mr Rosa Brunet, Baskan Gida used all four of them as the basis for its first utilisation request on 21^st December for a draw-down of €2,739,614.40 under the Facility. This led to BTM issuing a letter of credit to Akbank, against which Baskan Gida made its first and second draw-downs under the Facility on 3^rd and 9^th January 2002.
524. The only way in which an uninitiated reader of the four re-numbered contracts might have appreciated that they related to an outstanding balance from earlier contracts and orders was that each of them continued to identify by its number the Ferrero Italy purchase order pursuant to which the hazelnuts had originally been requested, in September and October 2001. But again, the uninitiated monitor of the Facility at the Banks had no means of knowing, without calling for sight of those purchase orders, the dates upon which they had been placed.
525. The Banks' case is that Baskan Gida made those outstanding deliveries the subject of newly dated re-numbered contracts because the Baskans wished to draw down substantial funding under the pretence of having received fresh orders, and to use the Ferrero signatures on the contracts as if they qualified as Ferrero purchase orders under the provisions in the Facility Agreement regulating utilisation requests, which expressly required a signed purchase order to be submitted. Submission of the relevant Ferrero Italy purchase orders would have demonstrated to the Banks that the utilisation request lodged on 21^st December was based on Ferrero Italy orders on average more than three months old, specifying delivery periods which had long since passed.
526. The Banks allege that Mr Casale must have appreciated when agreeing that he and Mr Rosa Brunet should sign the re-numbered contracts on 19^th December that they would be likely to be used by Baskan Gida to give a false impression to the Banks of the type which I have just described, in connection with the raising of finance. In support of the allegation of dishonesty, the Banks also repeat in relation to these contracts similar allegations to those made in relation to the Early December Contracts such as that Ferrero had no need of the hazelnuts to which they related; that, if marked to market on 19^th December 2001 they were loss-making and that, if all that was being done was to re-number and re-date old contracts, this made no commercial sense. The Banks also pointed to the fact that Ferrero did not include copies of the re-numbered contracts in its own disclosure, the inference being the person or persons responsible for agreeing to having them signed had quietly buried them.
527. Mr Rosa Brunet's evidence was that he had some recollection of having been asked to sign the re-numbered contracts by Mr Casale, with the explanation that Baskan Gida wanted Ferrero to confirm the amounts outstanding under the existing orders. He said that having been told by Mr Casale that the outstanding amounts were correctly recorded in the new contracts, he signed them without any further inquiry, still less checks of his own, not regarding it as his job to become immersed in matters of detail that were the responsibility of Mr Casale and his subordinates. He said that he gave no thought to the use which Baskan Gida might wish to make of the re-numbered contracts, but if they had been used for the obtaining of finance, he would not have thought there was anything wrong about having outstanding orders re-confirmed by fresh contracts.
528. For his part, Mr Casale was reluctant to admit that he knew by 19^th December that the Facility had been completed, rather than merely that it was imminent. Nonetheless he accepted in cross-examination that he probably did think at the time that Melih Baskan wished to use the re-numbered contracts in connection with raising finance from the Banks. Nonetheless, he said that it did not occur to him for one moment that the Baskans had any improper intent to mislead by the use of the re-numbered contracts, having no doubt at that stage as to their integrity. He added that he could not see how any pretence that the contracts related to fresh orders could have been made to the Banks by the use of re-numbered contracts which contained reference to the original Ferrero Italy purchase order numbers. His evidence as to why Ferrero had no copies of the re-numbered contracts to disclose was not entirely consistent, as between his August 2005 witness statement and his cross-examination, but in the context of an enormously complicated case I do not regard that discrepancy as being of much significance.
529. In my judgment, the question whether I should conclude that Mr Casale or Mr Rosa Brunet knowingly participated in the Baskans' intended dishonest use of the re-numbered contracts by signing them on behalf of Ferrero Italy depends critically upon the anterior question whether anything in the Baskans' conduct to date was sufficient to put them on inquiry as to their integrity. An honest person with such a suspicion would, when asked to sign the re-numbered contracts have stood back and asked whether the Baskans could put them to a dishonest use when raising finance from the Banks. From that perspective, questions to Melih Baskan as to why, for example, the outstanding deliveries could not simply be made the subject of an exchange of letters referring to them as such would have been unlikely to have produced a satisfactory answer. By contrast, to a person without such prior suspicion, the fact that the Baskans wished to make outstanding deliveries the subject of fresh contracts, even in connection with imminent or available Bank finance would not in my judgment have, of itself, been inherently suspicious, coming from a family with whom Ferrero had a longstanding relationship of mutual trust and respect. Businessmen are not ordinarily expected to approach requests from counterparties with a predisposition to ask whether the request derives from a dishonest motive. In the present case there is no evidence that Ferrero knew anything of the terms of the Facility, let alone whether the presentation of outstanding orders as the subject matter of fresh contracts would of itself have made the difference between obtaining or not obtaining finance for their performance. In the event, perhaps ironically, it was only because of a concession by the Banks, in accepting signed contracts rather than purchase orders as the basis for the utilisation request, that the Baskans were able to use the re-numbered contracts for the purpose of raising finance at all.
530. I am therefore disposed to give Mr Rosa Brunet and Mr Casale the benefit of the doubt in relation to their participation in the signing of the re-numbered contracts. I have little difficulty in that regard in relation to Mr Rosa Brunet, whose seniority would have entitled him to pay little attention to the detailed content of documents which he was asked to sign by a trusted junior who had his own subordinates under him. My conclusion has been less straightforward in relation to Mr Casale. In the end, the absence of a motive for either of them to have played a knowing part in a deliberate deception of the Banks at this stage is sufficient to lead to a conclusion in their favour.

2002

January – Dealings between Baskan Gida and Ferrero

The 3^rd January Bolowich Email

531. On 3^rd January 2002 Mr Bolowich sent a short email to Mr Rosa Brunet and Mr Casale, about an invoice just received by Ferrero Germany in respect of a delivery made on 30^th December, pursuant to Contract 5072, which had itself been made on 21^st September 2001, with a delivery period (according to the Ferrero Germany purchase order) expiring at the end of January 2002. The body of the email reads as follows:

"Today we are getting the original invoice 01kdz 0596 with the documents.

1. The invoice date is 31.10.01, however the lot was shipped on 30.12.01.

2. The invoice states "our bank is Pamukbank". Do we have to assume that this still correct? (issued on 31.10.).

3. The phytos. certificate states "control date 31.10.01" and "in 20 days must be exported".

4. The certificate also has a stamp with "fumigation has not been done".

However I have received from Akdeniz a fumigation certificate with the corresponding invoice.

As a "team player" I merely wish to avoid possible problems.

Best regards

Bolowich"

532. Points 1, 3 and 4 of Mr Bolowich's email are self-explanatory. He was concerned about a request to pay on an invoice which preceded delivery by two months, in respect of goods with an out of date phytos. certificate and a doubt as to whether there was a fumigation certificate. The mischief upon which the Banks relied is in point 2, which in his evidence Mr Bolowich explained as deriving from his recollection of the Ferrero Germany Acknowledgment dated 6^th December 2001, with its incomplete payment instructions, referable to sales contracts dated on or after 10^th December 2001, and its concluding note: "Same account information will appear on each invoice supplied by Baskan Gida to Ferrero OHGMBH." Mr Bolowich said that his reference to being a team player was because, shortly beforehand, he had been offended by being asked by Mr Rosa Brunet whether he was a team player. It is in that context understandable that Mr Bolowich retained a lingering memory of this otherwise routine email.
533. There is no surviving written response by either Mr Rosa Brunet or Mr Casale, and in cross-examination both Mr Bolowich and Mr Casale assumed (without being able to remember) that Mr Casale must have responded on the telephone, instructing Mr Bolowich to pay to the account shown on the invoice, i.e. Pamukbank. Mr Rosa Brunet could not remember anything about the matter.
534. Having at a late stage decided not to pursue a case of dishonesty against Mr Bolowich, the Banks claimed that this email shows an honest Ferrero employee pointing out to dishonest employees that Baskan Gida was backdating invoices, the inference being that it was designed to enable Baskan Gida dishonestly to avoid Ferrero Germany paying the money to the Banks' Collection Account.
535. I am wholly un-persuaded by this argument. First, Contract 5072 long pre-dated 10^th December (the contract trigger date in the Ferrero Germany Acknowledgment) and was not used by the Baskans as the basis for any draw-down from the Banks under the Facility. As will appear, it was, later in January 2002, replaced by a new contract for the balance then outstanding (which did not include this consignment) namely Contract 5100, which was so used.
536. The Ferrero Industrial Acknowledgment signed on 13^th December did request payment to the Collection Account of invoices dated on or after 17^th December, but also contained the Same Account clause at its end. Although it was uncommon for Ferrero to receive invoices dated long before a particular delivery, I am not persuaded that Mr Bolowich's email should have made Mr Casale think that the invoice to which Mr Bolowich was referring had been dishonestly backdated. Mr Casale suggested the obvious alternative explanation in cross-examination (although as he acknowledged he could not in fact recall what had happened), namely that there had been some reason for a delay in the delivery. Furthermore, Mr Bolowich's email does not in terms explain that his concern was whether the money should have been paid to the Banks' Collection Account, and he could not recall whether he had spelt out the nature of his concern to Mr Casale when he received his instructions on the telephone to pay Pamukbank. I have little doubt, had Mr Casale asked Melih Baskan about this, that he would have been told that, since these hazelnuts had not been funded by the Facility, payment was not due to be made to the Banks in respect of them. That is exactly what happened when Ferrero Australasia paid an invoice into the Collection Account in respect of a contract which had not been funded by the Banks, later in January. In those circumstances the suggestion that this event shows either an active dishonest participation by Mr Casale in a conspiracy, or that it should have put Mr Casale on notice of the Baskans' fraud, falls well short of the mark.

Contracts 5088 and 5089

537. Contracts 5088 and 5089 were two of the three large early December Forward Contracts, made respectively by Ferrero Germany and Ferrero France. Contract 5088 was for 2,016mt and the contract delivery period was March to April 2002 inclusive. Contact 5089 was for the same quantity, for delivery as to 600mt in April, 720mt in May and 696mt in June 2002 respectively. The Baskans wanted to use those two contracts for the purpose of drawing down money under the Facility, but their delivery dates were too far forward to give rise to a repayment stream within the specified 45 day period. They therefore wished to obtain signed revisions to those two contracts with altered delivery dates permitting (but not requiring) delivery within 45 days.
538. On 16^th January Mr Casale received from Melih Baskan copies, for signature by Ferrero Germany and Ferrero France respectively, of revised versions of Contracts 5088 and 5089. The revision of Contract 5088 specified a delivery period from December 2001 through March 2002. The revision of Contract 5089 specified a delivery period of "from December 2001 to June 2002". By chance, Mr Bolowich was visiting Alba that day, so Mr Casale obtained his signature to the revised Contract 5088, and returned it that day to Melih Baskan. On the following day Mr Casale returned a revised Contract 5089 appearing to bear a signature and stamp of Ferrero France. The delivery period on the version apparently signed by Ferrero France reads "from December 2001 through June 2002" ("through" being a replacement for "to" on the version originally sent by Melih Baskan on the previous day). I shall refer to those two versions as the "To version" and the "Through version" respectively.
539. Surviving email records show the passage of the To version from Baskan Gida to Mr Casale at Ferrero Italy, and on to Ferrero France, where an unsigned copy was retained. The Through version, appearing to be signed by Ferrero France, survives only in a fax sent by Mr Casale back to Baskan Gida, and then transmitted on to the Banks, and retained in their records, with an annotation to link it to the relevant Ferrero France purchase order.
540. Neither Mr Casale nor Mr Bolowich could recall the episode at all, and no witness was called from Ferrero France.
541. Basing himself on the absence of any copy of the Through version in Ferrero France's disclosure and the absence of any transmission details showing it having been sent, after signing, by Ferrero France back to Ferrero Italy, Mr Wardell submitted that Mr Casale must have forged the Ferrero France signature, and applied a Ferrero France stamp to the Through version before sending it back to Baskan Gida, either because Ferrero France had refused to cooperate or because he did not want to risk a refusal by sending an altered Through version to replace the earlier To version. More generally, Mr Wardell submitted that Mr Casale's obtaining (or forging) of Ferrero Germany and Ferrero France's signatures on revised versions of these two contracts demonstrated him both knowing of and participating in the Baskans' fraudulent designs on the Banks, by falsifying contract details to make them appear to comply with the requirements of the Facility.
542. Mr Casale's evidence was that, although he could not recall any involvement in this episode, he did not forge the altered version of Contract 5089, did not have a Ferrero France stamp, but accepted that Melih Baskan must have asked him to arrange for Ferrero Germany and Ferrero France to countersign copies of those two contracts. His assumption (rather than recollection) was that he had arranged for that to happen without being appraised of any change in the delivery periods, and that, because he did not regard delivery periods as important, he instructed Ferrero Germany and Ferrero France to sign them, without realising that the dates had indeed been changed.
543. My conclusions on this episode are as follows. First, the evidence comes nowhere near proving that Mr Casale forged the Through version of Contract 5089 appearing to be signed by Ferrero France. In my judgment it was both signed and stamped by Ferrero France, even though that company retained the earlier, slightly different To version in their records unsigned, and despite there being no surviving evidence of the Through version having been transmitted back to Ferrero Italy. It seems to me fanciful to suppose that Ferrero France would have objected to signing either the To or Through versions of Contract 5089 on Mr Casale's instructions. Ferrero Industrial and Mr Casale in particular were in charge of these matters, and the responsible officers of the operating companies habitually acted upon his instructions, just as Mr Bolowich did in relation to Contract 5088. I can see no reason why Mr Casale should have had a Ferrero France stamp in his office, and I do not regard Ferrero's failure to call Mr Caroly (who had in any event left Ferrero France before this matter came on for trial) as justifying an inference of forgery.
544. As for the more general suggested inference that this episode shows Mr Casale acting as a dishonest conspirator, I regard that as thoroughly improbable. This event occurred before Mr Casale had any reason to suspect the Baskans of dishonesty. Even if Mr Casale had recognised that the delivery period was being changed, and had asked Melih Baskan for an explanation, I would expect Melih Baskan to have satisfied him with the explanation that the Facility terms prevented Baskan Gida using the Banks' funds in relation to these contracts unless the delivery periods were adjusted as requested. Ferrero had no reason to object to Baskan Gida being given the option to make earlier delivery. I see no reason why a variation to adjust the delivery periods in that way should have been regarded by Mr Casale as an improper, still less dishonest request by the Baskans, providing only that they made no secret agreement between themselves that Ferrero and Baskan Gida should remain bound by the original delivery periods, to the exclusion of the new ones.
545. There is some very slight evidence to that effect, in that when these consignments were reordered from Aksu Gida in February 2002, the original delivery periods were reinstated. Since both Ferrero France and Ferrero Germany would at that stage probably have been writing new orders based upon the original purchase orders of December 2001, this is not particularly surprising, all the more so since neither Ferrero France nor Ferrero Germany retained signed copies of the 16^th/17^th January revisions on their files. That evidence therefore falls well short of establishing a case of a secret bargain between Melih Baskan and Mr Casale to adhere to the previously agreed delivery periods, sufficient to render the revised versions signed on 16^th and 17^th January instruments of fraud. In relation to Contract 5088, it is difficult to see how that could have happened without Mr Bolowich's knowing involvement, and the Banks quite properly no longer maintain any allegation of dishonesty against him.

January and February – The Implementation of the Fraud

546. Having lured the Banks into the Facility by the end of December 2001, large amounts of money began flowing from the Banks to Baskan Gida from the beginning of January 2002. By 26^th February an aggregate in excess of €22.8 million had been drawn down in eleven tranches, pursuant to letters of credit issued pursuant to six utilisation requests.
547. Getting the money into Baskan Gida was only the first step in the fraud. If the Facility was allowed to operate in accordance with its terms, the working capital thereby obtained would flow back to the Banks by means of the purchase and processing of hazelnuts, their sale to Ferrero, and payment by Ferrero to the Banks pursuant to the Prescribed Endorsement which Baskan Gida was required to insert in its invoices, requiring payment to the Collection Accounts. Furthermore, even if that process was circumvented (as it was), the Baskans would not themselves thereby have obtained a benefit from their fraud if Baskan Gida was not thereby rescued from insolvency (which it was not), since Baskan Gida's numerous creditors, including both the Banks, its Turkish bankers and the Dahod Investors would ultimately have had recourse to the money, or its proceeds in the form of hazelnuts, in any enforced liquidation of Baskan Gida.
548. The third necessary element in a fraud designed by the Baskan family to use the Banks' money as the means of rescuing their hazelnut business from Baskan Gida's insolvency was therefore to find a way of spiriting Baskan Gida's assets, including hazelnuts purchased with the Banks' money, into some other corporate vehicle within which they would be protected both from the claims of Baskan Gida's creditors and from personal claims against the Baskans, followed by execution against their assets. This required the construction of a phoenix scheme designed to frustrate the underlying principle of Turkish insolvency law, namely that the assets of an insolvent company should be distributed pari passu among its creditors, so that they could be enjoyed in the future by the Baskans, and by such associates as they found it necessary or convenient to involve in the scheme.
549. Once it was apparent that money would be forthcoming from the Banks under the Facility, the Baskans lost no time at all in formulating the necessary phoenix scheme ("the Scheme"), and in bringing on-board Mr Abidali, Indo-Med and the Dahod Investors as necessary participants in it. As designer of the scheme the Baskans chose a Turkish lawyer Mr Alpay. The essential features of the Scheme were as follows:

i) A phoenix company had to be identified, having the outward appearance of being owned and controlled entirely independently of Baskan Gida and its Baskan family shareholders, to which Baskan Gida's factories and warehouses would be let, its workforce, plant and equipment transferred, together with a sufficient opening stock of hazelnuts for it to be able to carry on Baskan Gida's former business without substantial outside bank lending for working capital purposes.

ii) A pretext for the lawful transfer of those assets and (in the event) pre-payment of rentals had to be identified, sufficient to prevent the transfer from being set aside at the instance of Baskan Gida's creditors or liquidator.

iii) Baskan Gida's suppliers and customers had to be persuaded to transfer their business to the phoenix company.

550. The Baskans and their advisers identified Aksu Gida as the suitable phoenix company. It was at the time wholly owned by a Mr Latif Aksu, who was employed by Baskan Gida as its domestic sales manager. Domestic sales formed only about 5% of Baskan Gida's business at the time. Although Mr Aksu was not a manager of Baskan Gida's workforce, for reasons which were not explained (and which I infer had something to do either with tax or Turkish employment legislation) Baskan Gida's workforce had some years previously been transferred to Aksu Gida as employer, but then made available on contract to Baskan Gida. Aksu Gida had no other business than the provision of employees to Baskan Gida, and in particular no hazelnut business of its own. Similarly, Mr Aksu had no experience of hazelnut exports, raw material acquisition or processing, and he spoke no English at all, or any other European language with which he could speak to export customers.
551. Since Aksu Gida was not a significant creditor of Baskan Gida, and had no assets with which to purchase Baskan Gida's plant, equipment, stock in trade and goodwill, some means had to be found to justify the transfer of all those assets from Baskan Gida to Aksu Gida. A suitable pretext was found, in the form of a retrospective purported reconstruction of the nature and terms of the Dahod Investment. It will be recalled that the Dahod Investment (as described by Mr Abidali, Mr Dahod and the contemporary documents) was in substance a joint venture for the conduct of the hazelnut export business to Ferrero, under which the Dahod Investors provided the capital, Indo-Med managed the application of the capital, and Baskan Gida did the work, Indo-Med's and the Dahod Investors' reward being a US$140 per mt profit share on hazelnut exports to Ferrero, the joint venture and associated Dahod Investment being agreed to persist for the whole of the 2001 season (i.e. to July 2002).
552. Mr Alpay must have regarded this as a less than suitable basis for a demand that Baskan Gida transfer its assets to Aksu Gida, because he arranged for part of it to be mis-described as an overdue loan, pursuant to an agreement for repayment backdated to 5^th November 2001 ("the Protocol Agreement") supported by three backdated promissory notes, supposedly made on 15^th November 2001 ("the Promissory Notes").
553. The false appearance was thus to be created that Indo-Med (through which the Dahod Investment had been channelled) was by mid-January 2002 an unpaid creditor of Baskan Gida in the sum of some US$10 million, with an apparent commercial reason for putting pressure on Baskan Gida to repay, either in cash, or if cash was unavailable, in kind.
554. The next stage was to arrange for Indo-Med to pretend to exert pressure for repayment in the form of debt recovery and threatened bankruptcy proceedings against Baskan Gida in the Turkish courts, and for those proceedings to be compromised by an agreement whereby Baskan Gida would transfer assets (mainly hazelnuts) to Indo-Med's order in settlement of the debt.
555. The final stage was for Indo-Med to transfer its rights under that settlement agreement and under the Protocol Agreement to Aksu Gida by way of subscription in kind for an 80% shareholding in Aksu Gida and the provision of working capital, so that Baskan Gida could apparently lawfully transfer its assets to Aksu Gida pursuant to an agreement to do so obtained by the application of creditor pressure.
556. None of the Baskan family participants in the creation of the Scheme, still less Mr Alpay, gave evidence in these proceedings, but it is an easy inference from the subsequent implementation of the Scheme, which I shall describe in detail, that it was pre-planned at least in substance by 5^th January 2002, even if its fine tuning continued thereafter.
557. The effect of the Scheme as I have described it was that, following the transfer of Baskan Gida's plant, equipment, stock in trade and goodwill to Aksu Gida, and the lease of the necessary factory and warehouse premises with which Aksu Gida was to carry on the business, it would appear to the outside world that Baskan Gida had succumbed to creditor pressure, in paying its most pressing debts in kind under threat of bankruptcy proceedings to an independent company Indo-Med, which had then arranged for the transfer of the assets to its 80% owned subsidiary so that the hazelnut business could thereafter be carried on by a new Turkish company apparently neither owned nor controlled by the Baskan family. Nonetheless, the Scheme would not achieve its desired purpose (namely the preservation of the hazelnut business for the Baskan family's benefit) unless first, Baskan Gida's suppliers and customers could be persuaded to transfer their business to Aksu Gida and, secondly, the Baskan family retained concealed beneficial ownership of Aksu Gida, or the means of resuming ownership once the dust generated by the collapse of Baskan Gida, and the non-payment of its major creditors, had settled. The Baskans needed also to retain de facto control of Aksu Gida throughout. This is because it is, on the face of it, difficult to see how the Baskans would have been able to persuade Ferrero to take on Aksu Gida as its main Turkish hazelnut supplier if its only credentials were that it was a company with no previous experience of the sourcing, processing or export of hazelnuts, owned as to 20% and managed by a man with no such experience, and no English or other European language with which to communicate with Ferrero, and with no track record of having a supply chain of the type which had in the past contributed to Baskan Gida' reputation as the supplier of the highest quality hazelnuts. The same difficulties might be thought to attend any attempt to persuade Baskan Gida's supply chain to transfer their business to Aksu Gida, for similar reasons, save for the language barrier.
558. The mechanics by which the Scheme was implemented, so as to be fully up and running by the end of February 2002 before the Banks had any inkling of it, are not the subject of significant factual dispute. Even the backdating of most of the principal documents used to implement the Scheme is not now denied, although previously dishonestly concealed by the Scheme's participants when seeking with some success to defend it from Baskan Gida's creditors before the Turkish courts. The main issues which I have to decide about this part of the case concern the manner in which the Scheme was presented first to Mr Abidali and the Dahod Investors and secondly to Ferrero, the true nature of the deal between the Baskan family, Mr Abidali and the Dahod Investors in relation to the ownership and control of Aksu Gida, the question whether the Scheme in fact involved unlawful acts, either in its implementation or use, and the question whether either Mr Abidali or Ferrero knew (in the relevant sense) that it did.
559. The last two questions arise at two distinct levels. The first looks at the Scheme on its own, the potential unlawfulness being the deprivation from most of Baskan Gida's creditors of their rights to share in its assets pursuant to Turkish insolvency law. The second level treats the Scheme as part of a wider conspiracy to defraud the Banks, the alleged unlawful acts being the dishonest obtaining and use of the Banks' money, contrary to the terms of the Facility Agreement, for the purchase of the hazelnuts to be transferred to Aksu Gida under the Scheme, and without any intention that the money borrowed would ever be repaid. It also involved the dishonest obtaining of hazelnuts free from the Banks' rights as pledgee, and the dishonest breach of the trust which then attached to them.

Introducing the Scheme to Mr Abidali

560. The Baskans introduced the Scheme to Mr Abidali during a visit by him to Turkey which lasted from the 5^th to the 9^th January 2002. By the beginning of his visit he knew that the Baskans had succeeded in obtaining a very large pre-export finance facility from the Banks, not least because he had been sent a press release by PRI crowing about its part in the transaction and stating that "PRI secured US$50 million in pre-export financing from two major international banks in London at a cost less than half of what the Company was paying to Turkish banks." In reply, Mr Abidali had sent PRI a short congratulatory email. Although he denied knowledge of it, contemporary documents show that Indo-Med had also paid KBC's fee for the Facility on Baskan Gida's behalf, debiting it to the inter-company running account. I reject Mr Abidali's evidence that he did not know what the payment to KBC was for.
561. In response to a vigorous cross-examination from Mr Wardell, along the lines that this information on its own must have given him severe concern about the future of the Dahod Investment (since it was meant to constitute the whole of Baskan Gida's working capital funding for its Ferrero business), Mr Abidali vigorously asserted first that he had no such concern, nor any worry about Baskan Gida's financial position, and secondly that, as far as he knew, this further finance related to Baskan Gida's large non-Ferrero business. While I accept that, prior to his visit, Mr Abidali may have had no particular concern for the future of the Dahod Investment, I reject his evidence that he thought that the Banks' money was to be applied solely for non-Ferrero business, for reasons which I shall explain later.
562. Mr Abidali's stated purpose for visiting Baskan Gida in January 2002 was to obtain documents in connection with a dispute being litigated in the American courts between Indo-Med and its American agent which, according to Mr Abidali, had sought to arrogate to itself the business connection with Baskan Gida which had been afforded to it on an agency-only basis. I have no reason to doubt that this formed a reason, even the main reason, for Mr Abidali's visit.
563. Nonetheless, while at Giresun Mr Abidali said that Melih and Cevat Baskan asked for a private meeting with him, and gave him the following information which I recite from Mr Abidali's first witness statement, to which he adhered in cross-examination:

"Cevat Baskan spoke little English so Melih Baskan interpreted. Cevat Baskan said that Baskan Gida was having problems with one of its Turkish bankers, Yapi Kredi. Apparently Yapi Kredi had persuaded Baskan Gida to make an investment in a biscuit factory to which Baskan Gida supplied hazelnuts. The biscuit factory had run into financial difficulties, a dispute had erupted between Baskan Gida and Yapi Kredi relating to the investment and Yapi Kredi was not providing facilities to Baskan Gida as previously promised.

Cevat Baskan said that because of these problems Baskan Gida wished to pay off its smaller creditors, the largest of which, as a result of the advances to Baskan Gida under the Dahod Agreement, was Indo-Med (and indirectly Mr Dahod and the Dahod Investors). He said that Baskan Gida was not in a position to repay the advances in cash but was looking for a way to repay in stocks of hazelnuts or other assets. This might be done by making stocks or assets available to Indo-Med or to another company which might be in a position to continue to use the finance which Indo-Med and Mr Dahod had provided to carry on a trade in hazelnuts. He suggested that I consider this and discuss it with Mr Dahod."

564. Mr Abidali's evidence was that, although this information gave him cause to discuss with Mr Dahod how the Dahod Investment should be preserved, it did not occur to him that a dispute between Baskan Gida and one of its Turkish banks had any adverse implications for Baskan Gida's ability to continue trading generally. He did not understand the proposal to amount to a wholesale transfer of Baskan Gida's assets and business or that, if implemented, it would lead to Baskan Gida being unable thereafter to pay its creditors in full. He said that the Banks were not even mentioned, and that he gave no thought to their position. If, as I consider likely, the Scheme was introduced to him broadly in the manner described in the passage quoted above, I find his evidence as to how he regarded its implications as wholly unreal. A proposal to pay a smaller creditor in kind, with assets sufficient to enable it to take over the debtor's trading business in the face of a dispute with one of its larger bank creditors, and with insufficient funds to pay in cash, clearly implies commercial insolvency at the very least.
565. After agreeing in principle with Mr Dahod that his investment should be protected in the manner suggested by the Baskans, Mr Abidali said that he then returned to Turkey on 13^th January for further discussions. It is clear from surviving documents that it was during Mr Abidali's second visit that serious drafting of the requisite documents began, and I accept Mr Abidali's evidence that he was first introduced to Mr Alpay during this visit, for a serious discussion about the detail of the scheme. There is some doubt when Mr Abidali first met Mr Aksu. In Further Information served on 11^th February 2004 verified by Mr Abidali's statement of truth, it is said that he was introduced to Mr Aksu "in or about the second week of January 2002", and the Further Information makes a brave but incredible attempt to portray an arm's length commercial discussion between Mr Abidali and Mr Aksu (who shared no common language), interpreted by Mr Melih Baskan, under which Mr Abidali discussed using an investment in Aksu Gida as a means of carrying on a Turkish business with the hazelnuts proposed to be transferred to Indo-Med in settlement of the Dahod Investors' debt. Since the Further Information suggests that the discussion commenced with Mr Aksu having "the Protocol Agreement dated 5^th November 2001" explained to him, and since it is clear from surviving documents that the Protocol Agreement was not signed and dishonestly backdated until, at the earliest, 19^th February 2002, I am satisfied that Mr Abidali's attempt to portray Latif Aksu as a serious player in the formulation of the Scheme, rather than as a puppet, was a complete fabrication. The account continues disingenuously to suggest that it was Mr Aksu who identified suitable local lawyers for Indo-Med to instruct, and Mr Aksu who arranged for drafts to be prepared, in Turkish and English, to record the terms of the agreement for Indo-Med's subscription for a shareholding in Aksu Gida. I am in no doubt that these steps, like the rest of the Scheme, were masterminded by Mr Alpay and his team, on the Baskans' instructions.
566. Although work had probably started on drafting the Scheme documents before Mr Abidali's arrival in Turkey on 14^th January, that is the earliest date on which they can now be traced as having been work in progress. The reason is that Mr Abidali (probably on advice from Mr Alpay) decided to instruct Mr Trotter to assist Indo-Med in connection with the documentation, and Mr Trotter's file on the subject opens with a request from Mr Abidali to courier to him in Istanbul a form of authority to enable him to sign documents on Indo-Med's behalf. Mr Trotter's file, with its evidence of the backdating of the main Scheme documents (so clear that the backdating has been admitted) only became available to the Banks when, at their request, Indo-Med's liquidator waived legal professional privilege. Until then, Mr Abidali had maintained that the documents had been signed on the dates they bore, for example in affidavit evidence in Turkish court proceedings later in 2002 which I shall describe in due course.
567. The first document in terms of fictional time was the Protocol Agreement. A first draft was prepared by Mr Alpay and forwarded by Mr Abidali to Mr Trotter, using Baskan Gida's fax machine, on 16^th January. Mr Trotter's instructions were that it had been prepared by Turkish lawyers whose first language was not English, and that he was requested to "anglicise" the English versions of the drafts. The toing-and-froing of the drafting process does not matter, save to note that the Protocol Agreement was still in draft when returned by Mr Trotter to Mr Abidali on 19^th February, and signed by Mr Abidali on behalf of Indo-Med and by Melih Baskan on behalf of Baskan Gida then or on some unknown date thereafter, before it was dishonestly relied upon in Turkish proceedings in June 2002.
568. In its final form it was expressed to have been entered into on 5^th November 2001 and made between Baskan Gida (as "Debtor") and Indo-Med (as "Creditor"), and recited that the Debtor was indebted to the Creditor in the sum of US$10 million. The Agreement purported to record "the repayment arrangements agreed between the Creditor and the Debtor". Clause 1 provided for US$2 million to be repaid in cash in accordance with three numbered drafts to be provided and signed by the Debtor in the amounts of US$700,000, 700,000 and 600,000 respectively, and maturing on 10^th January, 20^th January and 10^th February 2002. Clause 2 provided that the balance of US$8 million should be repaid "by the delivery by the Debtor in accordance with the directions of the Creditor of Hazelnuts to the value of US$8 million … not later than close of business on 15^th February 2002." Clause 3 contained a Turkish law and jurisdiction clause. My summary of the drafting history of the Protocol Agreement demonstrates that by the time it was signed, the deadline for the performance of every obligation mentioned in the document had already passed.
569. Three promissory notes with the amounts and maturities described in the Protocol Agreement survive in notarised form purporting to have been signed on 15^th November 2001. Again, it is apparent from the detailed drafting history of the Protocol Agreement that the amounts and maturities had not been finally decided upon until 18^th January 2002, at the earliest. It is not known upon what date, after 18^th January they were in fact signed, and then notarised, but this probably occurred before they were relied upon in the bankruptcy proceedings at the end of January.
570. The next document in fictional time order was the Shareholders Agreement, between Aksu Gida, Indo-Med and Latif Aksu. The chronology of its drafting history is virtually the same as that of the Protocol Agreement. It was therefore signed on or at some unknown date after 19^th February 2002, and backdated to 15^th January. It provided for Indo-Med to subscribe for 80% of the total issued shares of Aksu Gida for a cash payment of US$1.6 million, to make a further US$400,000 available to Aksu Gida by way of loan repayable on demand, and to introduce, at its discretion, up to US$8 million of working capital, in kind or in cash.
571. Next, there survives a written Assignment purportedly dated 16^th January 2002 and made between Aksu Gida and Indo-Med, reciting Baskan Gida's obligation under the Protocol Agreement to satisfy US$8 million of its debt to Indo-Med by delivery of hazelnuts or other goods (described as "the Rights"), and purports to assign the Rights from Indo-Med to Aksu Gida, in satisfaction of Indo-Med's obligation under the Shareholders Agreement to provide working capital in the amount of US$8 million to Aksu Gida. This Assignment was still in draft as late as 23^rd September 2002, as appears from a letter from Mr Trotter to Mr Abidali on that date, enclosing a draft. It was not signed until late October 2002.
572. As an additional fictitious preliminary to the commencement of bankruptcy proceedings by Indo-Med against Baskan Gida, a purported letter of demand bearing date 17^th January 2002 was created for Mr Abidali to send on behalf of Indo-Med to Melih Baskan. It reads:

"Dear Melih

Subsequent to our discussions in Giresun, please arrange to transfer all our remaining assets by way of either goods or cash to M/S Aksu Gida with immediate effect.

I regret this step that we are forced to take in view of the prevailing circumstances.

Yours sincerely

Shabbir Abidali"

It is not known upon what date this disingenuous letter was created, or signed by Mr Abidali. On the evening of the 17^th January 2002, he, Melih Baskan and Mr Trotter had a convivial dinner together in London, preparatory to Mr Abidali and Melih Baskan flying to Boston for further discussions with Mr Dahod.

573. The next document in the timeline cannot by its nature have been backdated. It was the bankruptcy application made by Indo-Med (through a Turkish lawyer Mr Zenginpeduk acting under a power of attorney) against Baskan Gida. It is to be noted that this application, being the first public manifestation of the scheme to the outside world, took place only after the Baskans obtained Ferrero's agreement to transfer its custom from Baskan Gida to Aksu Gida, on 25^th January.
574. The bankruptcy application was based upon non-payment by Baskan Gida against the three Promissory Notes which, I infer, must have been created and backdated between the 18^th and 28^th January, even though the Protocol Agreement, pursuant to which they were to have been signed, had not at that stage even been finalised. In strict form it was a debt claim with a threat of bankruptcy proceedings if the debt was not paid (rather like an English statutory demand), but since the parties have described the claim as a bankruptcy claim, I shall adopt that description for convenience.
575. The bankruptcy application was followed in terms of fictitious time by an immediate settlement of the claim, purportedly recorded in a Second Protocol Agreement, purportedly made on 28^th January 2002 between Baskan Gida and Indo-Med. It recited the first Protocol Agreement, as having been "executed between the parties on 5.11.2001", Baskan Gida's failure to honour the Promissory Notes, the bankruptcy proceedings and their settlement. It contained an admission of debt by Baskan Gida in the sum of US$2 million, a promise to pay it by transferring goods to the value of the debt to Aksu Gida, against the withdrawal of the bankruptcy proceedings, and provided for Baskan Gida to pay Indo-Med's costs. This document was still in draft as late as 23^rd September 2002, as appears from a letter of that date from Mr Trotter to Mr Abidali enclosing a revised draft, and signed at some unspecified date thereafter, probably in late October.
576. The documentary structure which I have described was designed to give the impression that Baskan Gida, having agreed in November 2001 to repay a US$10 million debt owed to Indo-Med by a mixture of promissory notes and hazelnuts, and having failed to pay on the due dates, was forced by the commencement of bankruptcy proceedings by an angry creditor to agree to a settlement under which the whole of the US$10 million was to be paid in kind rather than in cash, mainly in hazelnuts, and transferred to Aksu Gida by way of Indo-Med's satisfaction of an obligation to provide working capital and a subscription for a controlling shareholding. The objective of creating a new company apparently 80% owned and controlled by Indo-Med, and 20% owned and managed by Latif Aksu, to the exclusion of the Baskans, for the purpose of carrying on Baskan Gida's former hazelnut business was thereby largely achieved, at least in appearance, leaving only the transfer of the necessary stock in trade, plant, machinery and leasehold possession of the factory and warehouse premises to complete the exercise. The junior staff were already employed by Aksu Gida.
577. In fact, that documentary structure was a sham. First, the supposed debtor obligations apparently recorded in the November 2001 Protocol Agreement and Promissory Notes were a complete fiction, the truth being that there was something akin to a joint venture between the parties pursuant to which the Dahod Investors through Indo-Med had provided finance for the entire crop season which was months away from being due for repayment. Secondly, the appearance of creditor pressure reflected in the bankruptcy application and its prompt settlement were the precise reverse of the truth, since the initiative for repayment had come from Baskan Gida's owners, rather than from Indo-Med. Thirdly, the appearance that Aksu Gida was to be owned, controlled and managed independently from the Baskans was also a fiction, as will appear from my detailed findings of fact as to what subsequently occurred. In short, subsequent correspondence between Mr Abidali and the Dahod Investors on the one hand, and the Baskan family on the other hand, clearly demonstrates that, from the moment of the implementation of the Scheme in February 2002, the true understanding between those parties was that the Dahod Investors' and Indo-Med's beneficial interest in the hazelnut business thereafter carried on by Aksu Gida (and later by Baskan Yuksel) continued as it had always been to be that of a financial investor with a US$140 per mt profit share, rather than that of an owner, controller or manager. Furthermore, a wealth of evidence including but by no means limited to correspondence between Mr Abidali and Melih Baskan during 2002, demonstrates that the Baskans remained in de facto control, and continued to exercise day to day management, of the affairs of Aksu Gida, Baskan Yuksel and the hazelnut export business, notwithstanding the contrary picture presented by the documents, and ineffective attempts by the Baskan family in particular to conceal their continuing involvement.
578. My description of the fictitious documentary timeline has taken me a little beyond the real chronology of events. Following Mr Abidali's visit to Giresun in mid-January, it was necessary for him and Melih Baskan to discuss the Scheme with Mr Dahod, for which purpose they both went to see him in Boston, breaking their journey in London for a brief meeting and dinner with Mr Trotter. Face to face discussions with Mr Dahod took place in and near Boston on 18^th and 19^th January, about which he and Mr Abidali gave evidence, broadly to the effect that the Scheme was explained to him in the same way as it had been explained to Mr Abidali, and his approval duly obtained, upon the basis that it appeared to be the best way of protecting his investment in the hazelnut business from the unwelcome consequences of the Baskans' dispute with Yapi Kredi. Mr Dahod said that he approved the Scheme without asking to see, or being shown, any documents relating to it.
579. The nature and extent of Mr Dahod's knowledge and understanding of the Scheme is only of secondary importance in these proceedings, since Mr Dahod is not sued as a co-conspirator. Nonetheless, since his subsequent email correspondence with Mr Abidali is one of the most important pieces of evidence demonstrating that the Scheme was a sham, and since Mr Werbicki on Mr Abidali's behalf relied heavily in his submissions upon Mr Dahod's apparent corroboration of his client's account of events, it is necessary for me to make such findings of fact as are practicable about the way in which the Scheme was introduced to Mr Dahod. For this purpose, I have not been significantly assisted by the evidence of either Mr Abidali or Mr Dahod, for the reasons set out in my summaries of their shortcomings as witnesses. The most reliable guide as to what happened is to be found in their private email communications at the time.
580. The first is an email from Mr Abidali to Mr Dahod on 21^st January, describing his sleepless night on the plane back to London from Boston, thanking the Dahods for their hospitality and continuing:

"As per our discussions, once Melih is back in station we will

(1) Finalise the accounts for the trading period up to the end of 2001.

(2) Consider a formula and proposal for participating in the business beyond the current Ferrero boundaries.

(3) Work on a format for the new season to have a more structured conduit for our partnership.

If there are any other points that come to mind, please do not hesitate to let me know. As advised Melih and the others in the sales team will be in London (Inshallah) over the coming weekend to format our new marketing strategy."

581. On its face, that email suggests first that the Baskans, Mr Abidali and the Dahods were to continue in a form of partnership in the hazelnut trade as they had been during the 2001 season, secondly that the discussion had been about extending that partnership to hazelnut sales to Baskan customers other than Ferrero, thirdly that for the 2002 season (which would not start until August 2002) something more formal would be put in place for the continuation of that three way partnership, and fourthly that Melih Baskan would, throughout, remain the leader or at least a principal member of the sales team for the business. The email is irreconcilable with Mr Abidali's and Mr Dahod's evidence that, on 18^th and 19^th January Mr Dahod had been presented with and approved a scheme whereby, through Indo-Med, he was to become the 80% owner and controller of his own hazelnut business, separate and independent from the Baskans, free to trade with whatever customers his new company should think fit, and freed from any partnership or profit sharing arrangement with the Baskan family, whether in relation to the Ferrero business or otherwise.
582. In cross-examination Mr Dahod tried lamely to suggest that Melih Baskan's involvement in the further matters for discussion related only to the first of the three points itemised in the email. Subsequent email correspondence, in particular Mr Dahod's email of 23^rd January, Mr Abidali's email to Mr Dahod of 8^th February and his further email on 22^nd April together make it clear that Melih Baskan's input was required on all three outstanding matters.
583. In closing submissions Mr Werbicki tried valiantly to persuade me that the phrase in point (3) "our partnership" related only to a partnership between the Dahods and Mr Abidali, rather than one involving the Baskans. Again I reject that submission. The emails which I have already mentioned, together with further private exchanges between Mr Abidali and Mr Dahod in May and October 2002 (which I shall describe in their due chronological place) leave me in no doubt whatsoever that, throughout 2002, Mr Dahod regarded himself as being in a continuing profit sharing partnership with the Baskan family after the implementation of the Scheme, despite his attempts in his evidence loyally but untruthfully to assist Mr Abidali by suggesting otherwise.
584. Returning therefore to the face to face meetings with Mr Dahod on 18^th and 19^th January, the possibilities seem to me to be twofold. Either Mr Dahod was told of the structure of the Scheme as pretended by the documents which I have described in detail, but assured that in fact the profit sharing partnership with the Baskans was to continue behind a façade which pretended the contrary. Alternatively he was dealt with on a strict "need to know" basis, and left in the dark as to the pretence to be generated by the Scheme documents, and told only that there was to be a reorganisation of the Baskans' corporate vehicles with which he need not concern himself, and which would in substance leave the profit sharing partnership, and the continued involvement of the Baskans as owners and controllers of the business, intact. On balance, I find that the second of those alternatives probably represents the truth. My reasons follow.
585. The Dahod Investors' participation in the Baskan family's hazelnut business had by January 2002 led to an investment in excess of US$12 million, but had generated only two documents associating the Dahod name with that business, namely the email setting out the heads of terms and the short summary presented to the spiritual leader of the Dawoodi Bohra community for his approval, which I have labelled the Dahod Agreement, but which was by no means a full description of the terms of the joint venture. Secondly, the Scheme documents do not mention the Dahod Investors at all. They remain entirely undisclosed beneficial owners behind the screen of Indo-Med. Thirdly, although two of the main Scheme documents (namely the Protocol Agreement and the Shareholders Agreement) were in anglicised draft by 18^th January, Mr Trotter appears to have been instructed to send those drafts only to Mr Abidali at his Indo-Med offices in London on 18^th January, rather than (as he could easily have done) to Mr Abidali in Boston, for discussion with Mr Dahod. The clear implication is that Mr Abidali and Melih Baskan decided to keep to themselves the terms of the sham Scheme, and to describe to Mr Dahod only the commercial realities. Finally, that impression is confirmed by Mr Dahod's description of the corporate changes which took place early in 2002 as "the major reorganisation of Baskan" in a discussion paper prepared by him about the first year of the joint venture on 24^th October 2002. Even as late as July 2003, when Mr Dahod signed a letter of representation requested by Indo-Med's auditors Chantery Vellacot, he described the Dahod Investors as having funded only the purchase of hazelnuts (during the relevant year of Indo-Med's account ending in December 2001) and having simply assigned the benefit of those advances to Indo-Med during fiscal year 2002 when "Baskan faced difficulties and were unable to repay the funds".

Selling the Scheme to Ferrero

586. It is common ground that Melih Baskan travelled to Alba and met Mr Casale and Mr Rosa Brunet on 25^th January, to introduce them to the Scheme. There is an issue whether Mr Franko was also in attendance because, although both Mr Casale and Mr Rosa Brunet say that he was not, he did accompany Melih Baskan on his onward journey to London, where they were both collected by Mr Gersappe of Indo-Med. On balance, I doubt whether Mr Franko did attend the Alba meeting, but it does not matter.
587. In order to address the question how Melih Baskan's presentation of the Scheme must have struck Mr Casale and Mr Rosa Brunet, it is necessary to bring up to date their experience of Baskan Gida's performance. After a decline in deliveries in December which, as in 2000, got worse towards the end of the month, deliveries picked up thereafter, amounting in the weeks beginning 31^st December, 7^th January, 14^th January and 21^st January to 96mt, 504mt, 504mt, and 744mt respectively. Broadly speaking therefore, the delivery rate had by 25^th January recovered to the levels achieved in October and November the previous year. Like Mr Abidali, Mr Rosa Brunet and Mr Casale both knew that the Facility had by then been implemented, but knew nothing about its amount or detailed terms, beyond that which might be deduced from a careful reading of the Notices and Acknowledgments of Assignment. Beyond notifying the Ferrero Defendants of the typographical error in the Akbank designated account number, the Banks had done nothing to bring to the attention of any Ferrero company the detailed terms of the Master Assignment, the Master Pledge or Trust Receipt, or the arrangements for storage of hazelnuts in an SGS controlled warehouse.
588. It was both pleaded and put to Mr Casale in cross-examination that he knew by this time from discussion with Ferrero's employees on-site at Giresun (or from Melih Baskan) that hazelnuts purchased by Baskan Gida for export to Ferrero were stored at Giresun in a warehouse controlled by SGS, such that they could not be accessed by employees of Baskan Gida or Ferrero. In their defence the Ferrero Companies originally pleaded that in early 2002 Mr Casale had been orally informed both (i) that somebody's permission had been needed before hazelnuts could either be processed or taken out of the warehouse and that this had caused a short delay in production at the Giresun factory; and (ii) that part of the temperature controlled warehouse in Giresun was under the control of someone other than Baskan Gida, whose permission was needed in order to release hazelnuts for processing, and could not be accessed. It was denied that Mr Casale knew that SGS was in control of Warehouse 2, or that hazelnuts stored there were the source of deliveries to Ferrero.
589. By amendment, the admission in part (ii) of the above summary of Ferrero's case was deleted, and in cross-examination Mr Casale adhered to what thereby became Ferrero's pleaded case. He explained that the information described in (i) above had been given to him by Mr Scavino but it had not led him to understand that someone, let alone the Banks, were exercising security rights in relation to hazelnuts to be sold in due course to Ferrero. Mr Casale said that he assumed that Baskan Gida would be unlikely to wish to store hazelnuts destined for Ferrero for any significant length of time, because, once processed and made available for delivery FOT, Baskan Gida could obtain almost immediate payment.
590. Mr Aslan's evidence was that he wore an SGS uniform, that he maintained a rigorous control over Warehouse 2 which prohibited entry without his authority, and that the warehouse door had SGS's logo fixed on it by means of a relatively small notice. He was in daily attendance, and a not infrequent visitor to the staff canteen for lunch.
591. In my judgment it is probable that Ferrero's original pleading, rather than the amendment, correctly summarises Mr Casale's knowledge at the material time. As he himself said, there would have been no reason why SGS's occupation and control of Warehouse 2 would not have been noticed by Messrs Altare and Scavino, or reported to him. Had it been, he accepted that he would probably have assumed that some third party was exercising security rights in relation to hazelnuts stored there, but not that this necessarily related to hazelnuts later to be exported to Ferrero. Furthermore, he said, reasonably in my judgment, that he would have assumed that, before any nuts were released, any relevant terms for the release of the security would have been complied with.
592. The Banks' case in relation to the 25^th January meeting in Alba is that, however precisely the Scheme was explained by Melih Baskan to Ferrero, Mr Casale and Mr Rosa Brunet must have known or at least suspected that Aksu Gida would be a front company owned or at least controlled by the Baskans, and that the Scheme therefore involved a fraud by the Baskans on Baskan Gida's creditors. Further, the Banks say that since Mr Casale and Mr Rosa Brunet knew about the assignment to the Banks of the Framework Contract and of Baskan Gida's rights under all outstanding contracts with the Ferrero operating companies, they must have appreciated that a scheme under which outstanding contracts would be performed by Aksu Gida, with payment for deliveries therefore being made to Aksu Gida, would specifically involve a fraud on the Banks. It is therefore alleged that by agreeing to participate in the Scheme by transferring outstanding orders to Aksu Gida, without first informing the Banks or discussing the proposal with them, Ferrero thereby agreed to participate in those frauds, not least because, without Ferrero's willing participation, the Scheme could not have proceeded.
593. Ferrero's case is that, as presented by Melih Baskan, the Scheme appeared to be a perfectly lawful and transparent restructuring of the hazelnut export business for the purpose of insulating it from the damage which might otherwise be caused to it by unconnected financial difficulties experienced by the Baskans, that it represented a satisfactory way of preserving for Ferrero's benefit the supply chain which had provided high quality hazelnuts for many years and that, since it involved the use of the same factories, staff, equipment and middle management as had generated that high quality to date, Ferrero could both rationally and legitimately go along with the proposal. Their case is that nothing about Melih Baskan's proposal gave Ferrero any inkling that a fraud was in contemplation, and that the position of the Banks was neither mentioned nor considered.
594. The direct evidence of what Melih Baskan said at the meeting on 25^th January came from Mr Casale and Mr Rosa Brunet. No document recording the meeting has survived, and their evidence was that no notes were taken. Mr Wardell invited me to treat that evidence as incredible. It is certainly surprising. In my judgment, manuscript jottings may have been made, which never found their way into any formal file. I am not persuaded that a formal minute was taken, still less that it was deliberately destroyed. Indirect evidence came from Mr Do's account of a brief meeting shortly thereafter at which Mr Rosa Brunet and Mr Casale explained the Baskans' proposal, and obtained his consent to Ferrero's participation in it. An even more indirect source of evidence, but recorded in a nearly contemporaneous document, is to be found in a brief description of the outcome of the meeting given by Mr Abidali (who no doubt got it from Melih Baskan) to Mr Dahod by email on 8^th February. It is convenient to start with that contemporary evidence, which consisted of the following reply to Mr Dahod's emailed question on the previous day:

"What's the status with Baskan? I hope the meeting with Ferraro (sic) was a success."

Mr Abidali replied:

"Baskan – Ferrero received very positively an all systems go. Initiated ahead of schedule, 6^th feb due possibility that wind of their ideas may be getting out. Staying on top of it."

595. My summary of the direct evidence of the meeting comes mainly from Mr Casale's witness statement and cross-examination, since he provided more detail than Mr Rosa Brunet. Their accounts were broadly the same, subject to certain points of detail which I shall mention.
596. Mr Casale said that Ferrero was given no advance explanation of the purpose of the meeting, save for being told on the telephone by Melih Baskan that he wished to come and talk to him and Mr Rosa Brunet in Alba. He described Melih Baskan as "very down emotionally" … and that "he did not sound like he was having an easy time telling us this information". Melih Baskan did most of the talking, while Mr Rosa Brunet and Mr Casale listened.
597. Melih Baskan's message was that the Baskan family had some time previously been put under pressure by Yapi Kredi to take over a biscuit factory, the pressure being that if they did not, Yapi Kredi might withdraw Baskan Gida's existing credit lines. The factory had been acquired for some US$15 million with a loan from Yapi Kredi for which the Baskan family had given personal guarantees. The biscuit factory transaction had led to financial difficulties for the family, as a result of which Yapi Kredi was demanding payment under those guarantees. Mr Casale's witness statement continued as follows:

"He (Melih Baskan) said that the Baskans were trying to resolve the problem with Yapi Kredi but that, in the meantime, they might have to withdraw from the hazelnut business so as to avoid the problems with the biscuit business affecting the hazelnut business and also to be free to try to resolve their problem with Yapi Kredi.

If this happened, the Baskans were considering that they would step back from the hazelnut business and rent their factory to a company called Aksu Gida, headed by Mr Latif Aksu and Mr Ishak Franko …

Melih Baskan asked us whether, if this happened, we would be prepared (in principle) to receive from Aksu Gida the outstanding quantity of hazelnuts ordered with Baskan Gida. He assured us that the business would continue to be based on the structures that had been developed over the previous 10 years. In particular, other experienced Baskan Gida personnel, including the laboratories and production staff, would work for Aksu Gida so as to ensure satisfactory service … Melih also told us that there would be a continuity of the supply chain in that Aksu Gida would get hazelnuts from the same suppliers as Baskan Gida to meet the outstanding orders with the Ferrero Operating Companies on the same terms as to quality, quantity and price as before. He gave us the strong impression that Aksu Gida would buy hazelnuts from Baskan Gida's old suppliers as and when needed to meet Ferrero's orders.

Melih made it clear to me that the move to Aksu Gida was only one option and that he was optimistic that it would not come to pass. … He emphasised that the Baskans were doing everything to resolve the problems with Yapi Kredi. Nevertheless, he said he felt that it was important for him to be open with Ferrero. He was not asking Ferrero for any commitment at this stage. He was just explaining the situation and the options.

What I was told was that the Baskans would be stepping back from the hazelnut business, which they would rent to Aksu Gida, although Melih Baskan did not say that the Baskan family would have no further involvement in it. Indeed, we were told or later found out (I cannot recall which) that one of Melih's cousins, Resat Baskan (Purchasing), would continue to have some involvement in Aksu Gida's business. Resat Baskan was subsequently killed in a car crash sometime around spring 2002.

The impression that Melih Baskan gave was that the lease/transfer, if it happened, was not intended to be permanent. We believed that one of his objectives in this meeting was to try to reassure us that Ferrero should continue to receive supplies from the supply chain so that Ferrero would not take its business to other suppliers."

598. I have thus far quoted selectively from Mr Casale's main witness statement, as amended in chief in one important respect. As originally written, it had described Yapi Kredi's pressure as having been applied to Baskan Gida, rather than to the Baskan family to purchase the biscuit factory, and that the consequential financial problems were those of Baskan Gida, rather than merely the family. In that respect, Mr Casale's original statement was consistent with what Mr Abidali had been told by Melih Baskan. It is also consistent with Baskan Gida's accounts, which record that the biscuit factory company was one of its subsidiaries, although there is no reason to suppose that Mr Casale saw those accounts.
599. In cross-examination Mr Casale said that Melih Baskan had only said, on 25^th January, that the biscuit factory was a separate business from the hazelnut business, and managed by different people. He said that the question whether the biscuit factory was in separate ownership from Baskan Gida was not discussed, and that his thinking went no further than an assumption that the troubles with the biscuit factory affected the Baskan family, and might therefore affect the hazelnut business.
600. In cross-examination Mr Casale emphasised his understanding that Melih Baskan was describing a possible withdrawal of the Baskan family from the hazelnut business which was intended only to be temporary, while they sorted out the Yapi Kredi problem. He said that the question who were to be the owners of Aksu Gida was not discussed, nor whether it was to be owned or controlled independently from the Baskans during the period of their temporary withdrawal from the hazelnuts business. He also emphasised in cross-examination that Melih Baskan was not at the meeting seeking an answer from Ferrero to the question whether it would transfer outstanding orders to Aksu Gida, but simply giving advance information of a possibility which might occur in the future. Both in his witness statement and in cross-examination Mr Casale said that there was no reference to the Banks at the meeting from start to finish.
601. Mr Rosa Brunet's evidence was broadly confirmatory of that of Mr Casale, albeit from a less sure or detailed recollection of events. In his witness statement he described the nub of Melih Baskan's message as follows:

"As a result of their difficulties with Yapi Kredi, Melih said that the Baskans were not sure that they could continue supplying Ferrero with hazelnuts. He said that, if the problems could not be solved, they were thinking of renting the business to another company run by a Latif Aksu, which would then supply Ferrero with the outstanding amounts under existing contracts. … Melih assured us that the factory, workforce, supply chain and quality standards would remain the same. It would be the same quality hazelnuts and the same people would be involved in selecting and processing them."

602. Mr Rosa Brunet's recollection from the outset was that the Yapi Kredi problem had been described by Melih Baskan as one affecting the Baskan family, rather than Baskan Gida in particular, but that, as a result Yapi Kredi might try to interfere with Baskan Gida's business.
603. Mr Rosa Brunet's cross-examination included the following revealing passage:

"Q. Melih Baskan went on to say everything would be the same as it was before.

A. He said: we will maintain the same labour force, the same structure and that we will continue as before doing business."

The repeated use of 'we' in that answer struck me as a telling indication that, although Mr Rosa Brunet said that ownership of Aksu Gida was not discussed, his understanding from Melih Baskan was that the Baskan family would remain in de facto control of Aksu Gida throughout.

604. Mr Do's evidence about what Mr Casale and Mr Rosa Brunet told him about their meeting with Melih Baskan adds little of substance to their own accounts, save that Mr Do's impression was that Melih Baskan:

"had asked Mr Casale whether, if the lease to Baskan Gida (he must have meant Aksu Gida) went ahead or the business was transferred in some other way, they would now place new orders with Aksu Gida."

605. There was little with which the Banks could challenge the Ferrero witnesses' account of what they were told by Melih Baskan, other than by reference to the background and to probabilities. For clarity, I shall express my findings about what Melih Baskan said first, before addressing the major issues about what, in consequence, Mr Casale and Mr Rosa Brunet may have understood, believed or suspected about the nature and purpose of the scheme proposed to them.
606. It was, as I have said, essential for the Baskans to sound out Ferrero in relation to the Scheme since, without Ferrero's cooperation in placing orders with Aksu Gida, it would not work. It is to be noted that, prior to 25^th January, the Baskans had done nothing further than plan the Scheme and introduce it to Mr Abidali and the Dahod Investors.
607. In order to obtain Ferrero's cooperation it seems to me inherently likely that Melih Baskan would have been at pains to emphasise, to the extent consistent with the achievement of the Scheme's purpose, the extent to which it would ensure 'business as usual' in relation to Ferrero's hazelnut requirements. Although Mr Casale knew Mr Latif Aksu and Mr Franko as Baskan Gida employees, based respectively in Giresun and Istanbul, Melih Baskan could not reasonably expect Ferrero to feel any real assurance that business would continue as usual unless Baskan family management remained in place, even if behind a screen, so as to avoid giving the obvious impression that Aksu Gida was merely a front for the Baskans. In particular, I would have expected it to have been obvious to Melih Baskan that if he was to persuade Ferrero that its hazelnut requirements would continue to be sourced from the supply chains established by the Baskan family over many years, it would be necessary to encourage Mr Casale and Mr Rosa Brunet to assume that the Baskans would, behind the scenes, still be using their longstanding relationship with those suppliers to ensure this happened.
608. In my judgment, that is the impression which Melih Baskan gave to Mr Rosa Brunet and Mr Casale at the 25^th January meeting. Their recollection that he spoke of the Baskan family withdrawing temporarily from the hazelnut business while at the same time saying that "we will maintain the same labour force, the same structure and that we will continue as before doing business" hints at, but underplays, a message conveyed by Melih Baskan to the effect that, although the Baskan family would appear outwardly to have withdrawn during the temporary period necessary to deal with Yapi Kredi, they would, in the background, continue to ensure that business with Aksu Gida would, for Ferrero, be business as usual.
609. I consider also that, although no doubt Melih Baskan presented the Aksu Gida proposal as one which was, at that stage, a possibility rather than a certainty, he did nonetheless make it plain to Mr Casale and Mr Rosa Brunet that his family wished to obtain a reliable indication of Ferrero's attitude towards cooperation with the scheme. Being then uncertain of Ferrero's attitude, Melih Baskan would not have wished to present the proposal as being definite, lest the family burn its boats with its main customer before discovering that customer's attitude. For that purpose, I can well imagine that Melih Baskan portrayed the Yapi Kredi problem as something which might, rather than would, require the family's apparent withdrawal from the hazelnut business. Nonetheless, the fact that Mr Rosa Brunet and Mr Casale almost immediately sought Mr Do's views upon whether Ferrero should cooperate, that Melih Baskan must by early February have received a positive indication to that effect from Ferrero, in order to pass it on to Mr Abidali and that, by 28^th January the Baskans pressed the starting button by getting Indo-Med to issue the bankruptcy proceedings against Baskan Gida, demonstrates in my judgment that Melih Baskan made it clear to his listeners on 25^th January that he was seeking a positive Ferrero response which, shortly thereafter, he must have obtained. To the limited extent that Mr Casale's evidence suggests otherwise, I reject it.
610. Finally, I have no reason to doubt that Melih Baskan volunteered no information to his listeners as to the intended ownership or de jure control of Aksu Gida, and that neither Mr Rosa Brunet nor Mr Casale asked for that information. For Melih Baskan there was nothing to be gained by telling Ferrero more about the proposed scheme than it needed to know, and any emphasis upon the planned pretence of an Indo-Med/Aksu ownership of Aksu Gida would detract from the 'business as usual' assurance which I have concluded that he both needed to, and did, give. As practical businessmen, I have no reason to think that Mr Rosa Brunet or Mr Casale would have wanted to pursue questions about the ownership of Aksu Gida, provided that they were reasonably satisfied, as I infer that they were, that the Baskan family would remain in de facto control throughout.
611. The issue as to the effect which Melih Baskan's presentation had upon Mr Rosa Brunet and Mr Casale, and upon Mr Do when reported to him, is one of the central questions of fact in the case. I must first summarise their own evidence about it, starting with Mr Casale. In his witness statement he said that nothing in Melih Baskan's presentation gave the slightest indication that the proposed lease of the factory or transfer of assets or business to Aksu Gida would be anything other than legitimate, that it did not occur to him that Aksu Gida might be a front company for the Baskans, or that the Baskan family would own or control Aksu Gida. When pressed hard in cross-examination with the proposition that it was obvious from Melih Baskan's presentation that Aksu Gida was going to be used by the Baskans as a front to defeat the claims of Baskan Gida's creditors, Mr Casale's response was to say that it never occurred to him to suspect such behaviour of a company with which Ferrero had worked for ten years, doing business amounting to millions of dollars, without a single dollar being lost.
612. In cross-examination, Mr Casale's final position in relation to his understanding whether Aksu was to be independent from the Baskan family was that his impression was that it was, but that he did not know that it was. Mr Wardell suggested that it was inconceivable that honest managers of Ferrero's hazelnut business would have agreed to have transferred that business to an independent Aksu Gida without carrying out detailed due diligence, and deciding the matter formally, at board level. Mr Casale's response was only that he, Mr Rosa Brunet and Mr Do were honest, that they were the decision-makers in relation to that question, and that they did decide to transfer the business to Aksu without due diligence or a board meeting. Elsewhere, Mr Casale supported the rationality of such an approach by evidence that he knew Mr Aksu (albeit he could not speak to him in any common language) as a Giresun employee and Mr Franko as a manager based in Istanbul and that, bearing in mind that the factory, the equipment and the employees, including the laboratory staff, would all remain the same, this was a sufficient assurance of business as usual to make due diligence unnecessary. In cross-examination he accepted that all he knew about Latif Aksu was that he was an employee of Baskan Gida, but not his rank or his job description. He did not appear to have known what Mr Franko's job was either.
613. Throughout his evidence, Mr Casale maintained that he never gave a moment's thought to the question what effect the Aksu Gida proposal might have upon the Banks as creditors of Baskan Gida. He said (and I accept) that question was not discussed. In cross-examination it was put to him that he must have wondered how the proposal would impact upon the Banks' Facility, which, as he knew, Baskan Gida had only recently obtained. His answer was that, as at 25^th January, Ferrero was paying invoices in the usual way, and nothing had occurred to suggest that there was any difficulty with the Banks. He emphasised that Melih Baskan's account concerned problems with Yapi Kredi, rather than with any other banker to Baskan Gida. When asked how he thought Aksu Gida was going to provide working capital to fund the purchase of nuts for export to Ferrero he said that he had no idea. When asked why a supposedly independent Aksu Gida would wish to take on Baskan Gida's unperformed speculative forward contracts with Ferrero, when the market had if anything risen since early December, his response was only that the delivery periods were from March to June, so that there was plenty of time for the market to fall, and that any entrant to the hazelnut business would think it worth taking on the risk inherent in those contracts in order to obtain Ferrero as a major customer.
614. Mr Rosa Brunet's evidence as to the understanding which he gained from Melih Baskan's presentation was much the same as that of Mr Casale, and his responses to a similar cross-examination were also similar. He was, in the end, agnostic as to the intended ownership of Aksu Gida, saying that this information had neither been volunteered nor requested, so that he did not know whether Aksu Gida was or was not to be owned by the Baskan family. He did not share Mr Casale's recollection of understanding that the Aksu Gida arrangement was intended to be temporary, and his explanation for the absence of due diligence or rigorous questioning was more heavily based upon his perception that what was being described was, at that stage, a mere possibility.
615. As to Mr Do, he said that he obtained an abbreviated description of Melih Baskan's presentation at a short meeting with Mr Rosa Brunet and Mr Casale, that he understood Baskan Gida to be threatened with financial trouble, and that his main reaction was that he could not see how or why Ferrero could or should object to the proposed transfer, in particular if the same factory, plant and staff were to be used for the transferred business. He said that he gave no thought to the question who might be the owners of Aksu Gida, or whether the Baskans might be able to resume ownership or control in due course. As to the suggestion that the proposal smacked of impropriety, his main point was that, as presented to him, it appeared to be an open and transparent transaction, rather than something carried out in secret, hidden from Baskan Gida's creditors.
616. The starting point in my analysis of this issue is that, contrary to Mr Casale's denial and Mr Rosa Brunet's agnosticism on the point, they were given to understand by Melih Baskan, and did understand, that during the temporary transfer of the hazelnut business to Aksu Gida, the Baskan family would remain in de facto control. I have already described why, in my judgment, it was essential for Melih Baskan to convey that message to Ferrero. I consider it inconceivable that Ferrero would have agreed to place large orders with Aksu Gida without further questioning or due diligence, if that assurance had not been both given and understood. Nor would the meeting otherwise have generated not a single surviving document within Ferrero. Mr Do said that, from his perspective, uncertainties whether Aksu Gida would deliver the goods could be dealt with on a wait and see basis, since Ferrero paid only after delivery and was therefore exposed to no credit risk, while having stocks sufficient to cover its production needs in the meantime. While I recognise the theoretical force of that proposition, I am unable to believe that Mr Rosa Brunet and Mr Casale would simply have ventured into the uncharted territory of buying hazelnuts from a brand new company with no track record, controlled and managed only by two barely known employees of Baskan Gida, with the more senior of whom they could not even communicate, without making much more rigorous inquiry, both as to the ability of management to maintain quality standards, and as to the preservation of the supply chain. I reject Mr Rosa Brunet's explanation for the absence of due diligence, namely that the Scheme was then presented only as a possibility. Although that is how it was presented, no due diligence was later carried out by Ferrero before it became an actuality.
617. It follows that the proposal presented to and understood by them involved the following elements. First, its objective was to enable the Baskan family to shield their cherished hazelnut business from attack by a major creditor, namely Yapi Kredi. For this purpose, it matters not whether they envisaged that Yapi Kredi was a creditor of the Baskan family or (or also) of Baskan Gida itself. Even if Yapi Kredi was not understood by Ferrero to be a direct creditor of Baskan Gida, its hazelnut business was vulnerable to an attack by Yapi Kredi on the Baskans, since Baskan Gida was itself a major Baskan family asset, available by way of execution of any judgment against the Baskans personally. In fact I very much doubt if any of Messrs Casale, Rosa Brunet or Do gave significant thought to the distinction between debts of the Baskans and debts of Baskan Gida. As practical businessmen employed by a group for which Baskan Gida presented no credit risk, I would have thought such analysis unlikely.
618. Secondly, the proposal involved the Baskan family at one and the same time appearing to withdraw from the hazelnut business while continuing to remain in de facto control of it. This necessarily involved an element of subterfuge, not different in kind from that employed by a businessman who, fearing ruin, transfers his assets to his wife on the understanding, known only to them, that when the dust has settled they will be restored to him and that, in the meantime, she will deal with them in accordance with his instructions.
619. Thirdly, the proposal therefore necessarily involved an element of secrecy, at least in the sense that Yapi Kredi was unlikely to be told of the Baskans' assurance to Ferrero that they would continue to have de facto control of the hazelnut business while parked with Aksu Gida.
620. Fourthly, the scheme necessarily involved, if implemented, Baskan Gida defaulting on unperformed or partly performed hazelnut export contracts with Ferrero. The whole point of Melih Baskan's inquiry on 25^th January was whether Ferrero would then permit Aksu Gida to complete the performance of those contracts on the same terms. This was not a proposal which Melih Baskan described as at all likely to be implemented during the short period between the end of one hazelnut export season and the beginning of the next, when nothing would be outstanding for delivery. Mr Casale's description of his presentation expressly involved an assumption that there would be outstanding deliveries from Baskan Gida to Ferrero, at the point when the proposal was implemented.
621. The significance of this fourth point is that Mr Casale's careful study of the Notices and Acknowledgements negotiated in December 2001 might have demonstrated to a careful person in his position that those necessarily unperformed or partly performed contracts, as at the date of the implementation of the proposal, would fall within the contracts assigned to the Banks in connection with the Facility, such that a breach of them by Baskan Gida could destroy or diminish the value of Baskan Gida's assigned rights to payment thereunder, leaving the Banks (pro tanto) deprived of at least one of its means of obtaining payment of any advances made in connection with any of those contracts. While a decision to take that course after the obtaining of such an advance might be thought to be at worst unkind, a decision to do so followed by the obtaining of further advances, and without informing the Banks of that decision, or seeking their cooperation, would be dishonest.
622. Taken singly or together, the elements of subterfuge, secrecy and potential dishonesty in the scheme which I have concluded Melih Baskan presented to Ferrero on 25^th January are, for the most part, not of that sophisticated type which takes a trained insolvency lawyer to discern. Apparent alienations of assets while retaining covert control together with a promise that ownership will be restored in due course are as old and familiar as the hills. Furthermore, it must have been obvious to Mr Casale and Mr Rosa Brunet that Yapi Kredi were not to be informed either that the transfer of the business to Aksu Gida was temporary, or that the Baskan family were to remain in de facto control. Indeed, the extent to which Melih Baskan went in blaming Yapi Kredi for all the family's troubles (including forcing them to borrow the money for the purchase of the biscuit factory in the first place) suggests that Melih Baskan went out of his way to de-sensitise his listeners from any sense of moral disapprobation at the intended deception of Yapi Kredi which the scheme necessarily involved.
623. The more difficult question is whether the potential of the Scheme to deceive or otherwise injure the Banks was apparent to Mr Casale and Mr Rosa Brunet at the time when it was presented to them. In favour of a conclusion that it was apparent to them is the fact that they knew by then that the Facility was up and running, and therefore (if they thought about it) that the Banks were likely to be creditors of Baskan Gida, although they knew nothing of the personal guarantees given by members of the Baskan family.
624. In favour of a negative conclusion to that question are the following considerations. First, the Baskans had only recently established a new long term relationship with the Banks which Mr Casale knew had taken many months to negotiate, and which was designed to secure finance for the hazelnut business free from the difficulties of using Turkish bank lending. In the absence of any suspicion of premeditated fraud, it would be counter-intuitive to think that the Baskans would almost immediately put that relationship at risk merely because of a difficulty with one of their Turkish banks. Secondly, neither Ferrero, Mr Rosa Brunet or Mr Casale owed any duty to the Banks, who they may fairly be supposed to have assumed had constructed a sophisticated security structure for their lending, together with a duly careful and diligent system for its monitoring and implementation. In fact, as will appear, they had not, but this cannot have been apparent to Ferrero at the time. Thirdly, Ferrero and its relevant decision-makers were in principle entitled to consider their own commercial interests when presented with the proposal by businessmen who had proved trustworthy over many years, without worrying themselves about how that proposal might affect other stakeholders in Baskan Gida.
625. Finally, I have accepted subject to exaggeration Mr Casale's evidence that Melih Baskan put forward a proposal which might rather than definitely would happen, and that the contingency was explained (untruthfully) by Melih Baskan as based upon the family not yet knowing whether the Scheme would be necessary as part of its strategy for dealing with Yapi Kredi. In truth the only contingency by then affecting the Baskans was whether or not Ferrero would cooperate with it, but Mr Rosa Brunet and Mr Casale would have had no reason to know that. Presentation of a proposal as a possibility may reasonably be supposed to dull the sensitivities of the recipients as to its full implications.
626. I have come to the conclusion that Mr Rosa Brunet and Mr Casale did understand that the proposal put to them by Melih Baskan on 25^th January necessarily involved an element of subterfuge and consequential deception of Yapi Kredi as a creditor of the family (whether or not also of Baskan Gida itself) which was probably unlawful. Putting assets beyond the reach of creditors by a pretended disposal of them to a third party while facing financial ruin is unlawful in most civilised countries. Although I accept Mr Do's evidence that he believed that the transaction was transparent, and therefore lawful, in my judgment that resulted from the very brief explanation given of it to him by his colleagues. It was transparent only in the sense that no-one at Ferrero had any reason to suppose that the lease and asset transfer to Aksu itself would be concealed. It was covert in the sense that the Baskan family were to remain in control, and because the scheme represented a temporary expedient rather than a permanent transfer. I do not thereby mean that Mr Rosa Brunet or Mr Casale deliberately misled Mr Do in that respect. It was simply that the brevity of their explanation to him failed to disclose the subterfuge of which they had been made aware of by Melih Baskan.
627. By contrast I have concluded that I should accept the evidence of the Ferrero witnesses that the possible implications of the proposed scheme for the Banks never entered their minds on 25^th January. Since they owed no duty to the Banks, the question whether or not it should have entered their minds does not arise.

Ferrero Decides to Cooperate

628. Although it is nowhere documented, the Ferrero witnesses all said that the outcome of the short meeting between Mr Rosa Brunet, Mr Casale and Mr Do to which I have referred was that they decided that, if the Baskans went ahead with their proposal, Ferrero would order from Aksu Gida any hazelnuts outstanding on contracts with Baskan Gida at that time, and on the same terms. Notwithstanding the suggestion in cross-examination to the contrary, it was in my judgment a perfectly rational commercial decision, having regard to three factors. The first was that Ferrero had sufficient supplies to meet immediate production needs, and believed that there was a plentiful supply of hazelnuts still available on the market if, contrary to expectations, Aksu Gida also defaulted. The second was the 'business as usual' description of the way in which Aksu Gida business would be run, supervised by the Baskans. The third was that because Ferrero never pre-paid, it would incur no credit risk in ordering outstanding supplies from Aksu Gida. By contrast with those favourable factors, a refusal to cooperate would merely have meant that if Yapi Kredi's claims against the Baskans did adversely impact upon Baskan Gida's ability to deliver on its outstanding orders, Ferrero would then simply have no ready supplier in place to deliver comparable quantities to a reliable specification.
629. The documents do not disclose, and Ferrero's witnesses did not choose to reveal, when this favourable decision was communicated to the Baskans. In cross-examination on a different issue, Mr Casale revealed that he warned his subordinates Mr Altare and Mr Scavino in early February that a transfer of the business to Aksu Gida might happen and that, if it did, Ferrero would cooperate with it by transferring outstanding orders to Aksu Gida. His best recollection was that this occurred at the time when Ferrero placed its last three contracts with Baskan Gida, which occurred on the 5^th to 6^th February. It is a fair inference that Mr Casale so informed Melih Baskan at the same time if not earlier, and this precisely accords with Mr Abidali's email to Mr Dahod on 8^th February recording that Ferrero had received the proposal "very positively", and that therefore it was "all systems go".
630. In my judgment it is probable that Mr Casale told Melih Baskan that Ferrero would cooperate with the proposal if it went ahead, by 28^th January, the first working day after 25^th January, and the day upon which, at the Baskans' bidding, Indo-Med by its Turkish attorney issued the bankruptcy application against Baskan Gida. That application was an official document, with grave consequences for the credit of Baskan Gida if it came into the public domain. I consider it unlikely that the Baskans would have risked burning the Baskan Gida boat in that way if they had not by that stage been confident of Ferrero's cooperation in the scheme. Whether or not by coincidence, that was also the date upon which Melih Baskan asked for the last set of contract renewals between Ferrero and Baskan Gida, to which I now turn.

The 28^th January Contract Renewals

631. On 29^th January 2002 Mr Bolowich signed two further contracts on behalf of Ferrero Germany with Baskan Gida, numbered 5100 and 5101, both dated 28^th January. To the uninitiated, they look like new contracts for the delivery FOT to Ferrero Germany of 576 and 552mt respectively of paste kernels, with delivery periods in each case from January to March 2002 inclusive. It is common ground that Mr Bolowich signed them on Mr Casale's instructions, following a request to do so from Melih Baskan on 28^th January.
632. In fact, contracts 5100 and 5101 constituted re-confirmations of outstanding deliveries contractually due on two earlier contracts, numbered 5072 and 5076, dated respectively 21^st September and 8^th October 2001, for 1,392mt and 1,344 mt respectively. The difference in amount between the old and the new contracts had in the meantime been delivered. Unlike the original contracts replaced by the 19^th December re-confirmations, the deliveries outstanding under contracts 5072 and 5076 were not overdue, since the corresponding Ferrero Germany purchase orders specified delivery periods expiring at the end of January and February 2002 respectively. They were thus old but not 'stale' contracts in the sense that the predecessors to the 19^th December re-confirmations had been. The prices under contracts 5100 and 5101 were of course the same as those in 5072 and 5076 because, although the market had by then risen substantially, they represented no new commercial bargain, apart from a modest postponement of the contractual delivery periods.
633. Baskan Gida used contracts 5100 and 5101 as the basis for its fifth utilisation request on 1^st February 2002, for a sum of €2,375,676, and for its tenth draw- down in the same sum, on 8^th February 2002.
634. With the exception of 120mt under contract 5101, Baskan Gida did not default on either of these contracts. On the contrary, the remainder of the tonnage ordered under them was delivered to and paid for by Ferrero Germany before Baskan Gida went out of business. As to the 120mt outstanding, this was re-ordered by Ferrero Germany from Aksu Gida on 14^th February, and the remaining five lorry loads delivered within the following three days.
635. The Banks levy the same charges of fraud and dishonesty against Ferrero in respect of the 28^th January contracts as they do in respect of the contracts countersigned on 19^th December and 16^th/17^th January, with the added complaint that, by this time, Ferrero knew of the Baskans' plan that Baskan Gida should default on its obligations and be replaced as the supplier of any outstanding deliveries by Aksu Gida.
636. It will be recalled that I acquitted Mr Rosa Brunet and Mr Casale of any dishonesty in relation to the earlier countersigned contracts primarily because nothing in the Baskans' conduct by then had been sufficient to put either of them on inquiry as to their integrity. By 28^th January the position had changed. Both Mr Rosa Brunet and Mr Casale had together been presented with a scheme by Melih Baskan which, if implemented, involved to their knowledge both subterfuge and an element of deception as against one of the Baskans' Turkish banks. Although I have concluded that nothing on 25^th January actually caused either of them to ask themselves whether any deception of the Banks might also be involved in the proposal, the presentation to Mr Casale of re-numbered contracts with postponed delivery dates should have caused Mr Casale to consider, for the first time, whether the Baskans had any improper purpose in mind when using those re-numbered contracts to obtain finance from the Banks.
637. Mr Casale acknowledged, indirectly in his witness statement, and directly in response to cross-examination, that he appreciated when asked to obtain Ferrero Germany's signature to contracts 5100 and 5101 that the Baskans would probably use them to raise finance from the Banks under the Facility. He was, perhaps surprisingly, given unusually gentle treatment by Mr Wardell in cross-examination, on the question whether the events of 25^th January made Melih Baskan's request to sign these re-numbered contracts three days later suspicious, but the allegation was nonetheless just about sufficiently put. It may be that the Banks had it well in mind that, with the benefit of hindsight, Baskan Gida's purpose in seeking to raise further finance on these contracts was not to default upon them, leaving Aksu Gida to perform them free of any assignment, but rather to perform them and defraud the Banks by raising invoices which misdirected Ferrero's payments otherwise than to the Collection Account. Nothing in Melih Baskan's request that Ferrero Germany sign these re-numbered contracts could have put Mr Casale on notice of that intended fraud.
638. Mr Casale's evidence was that he could remember very little about the request to sign these two contracts. He said he had no recollection of there being new contract numbers on the documents, and could not remember whether Melih Baskan's request had been made directly to him, or whether it had reached him via Mr Bolowich. He could recall nothing in the request which actually made him suspicious of the Baskans' underlying purpose.
639. I am, again on balance, not persuaded that Mr Casale actually appreciated that the Baskans had a fraudulent intent behind their request that Ferrero sign these renewals, or that he had formed such a level of suspicion that he deliberately refrained from enquiry in order to shield himself and Ferrero from the unpalatable truth. As I have said, there is no evidence that anyone at Ferrero appreciated that the invoices later to be sent in respect of deliveries by Baskan Gida under these two contracts would direct payment elsewhere than to the Collection Account, and knowledge of an intent to defraud of that kind was not put to Mr Casale. As for knowledge that the Baskans intended that Baskan Gida should default on these two renewed contracts, that flies largely in the face of the objective facts, since, apart from 120mt, they were fully performed by Baskan Gida, and even that small outstanding amount was delivered very shortly thereafter by Aksu Gida. I consider it harsh to attribute to Ferrero knowledge of an intended default by the Baskan Gida in relation to these contracts, which was not in the event perpetrated, to any significant degree.
640. Nonetheless, the combination in Mr Casale's mind of his knowledge of the scheme proposed on 25^th January, and the Baskans' continued raising of finance from the Banks on the strength of renewed contracts with Ferrero must in my judgment have caused Mr Casale to start to wonder about the Baskans' integrity vis-à-vis the Banks. It must have generated some degree of suspicion, however small. It was the first of a series of events, the remainder of which I shall shortly describe, which must have caused that suspicion to grow during the following month, until it reached a belief that the Baskans were, vis-à-vis the Banks, probably up to no good, in March.
641. Whatever Mr Casale's level of suspicion at the end of January, it is clear that he asked no questions of the kind which might have had dispelled it, whether of the Baskans or of the Banks.

The Early February Contracts

642. Between 5^th and 7^th February 2002 the Ferrero Operating Defendants placed fresh long-dated orders with Baskan Gida for an aggregate of a further 5,904mt of paste kernels. They were negotiated between Mr Casale and Melih Baskan on 5^th February, and the three Baskan Gida contracts numbered 5103, 5104 and 5105 were all given that date. They have given rise to similar allegations by the Banks against Ferrero to those made in relation to the Early December Contracts, with which they share the same features, namely being for large tonnages, for delivery during a long forward period, and at prices which, if marked to market on the date they were made, represented a very substantial loss for Baskan Gida.
643. Taking them in order, contract 5103 reflected an order by Ferrero France for 2,016mt paste kernels for delivery between February and June, at a price of €2,750 per mt. Contract 5104 reflected an order by Ferrero Germany for 2,880mt paste kernels for delivery from February to 15^th May, priced at €2,795, and contract 5105 represented an order by Ferrero Italy for 1,008mt paste kernels for delivery between February and April, priced at €2,795mt. The slightly lower price for the Ferrero France order probably reflects its lower specification.
644. Not one lorry load of any of these three orders was in the event delivered by Baskan Gida before it went out of business slightly less than two weeks later. The whole of the tonnage was made the subject of new orders placed with Aksu Gida, by Ferrero Germany and Italy on 14^th February and by Ferrero France on 19^th February. In fact (for reasons which I will later describe) Aksu Gida made no deliveries under these three orders either, and they were replaced by orders placed with Baskan Yuksel by Ferrero Italy on 7^th May (the order being undated but the contract bearing that date) and by Ferrero France and Germany on 10^th May. In the event, Baskan Yuksel delivered 1,824mt to Ferrero France against 2,016mt ordered, and 984mt to Ferrero Germany against 2,640 ordered, by early July and early June respectively. Nothing was delivered to Ferrero Italy in the 2001 season, because the 7^th May contract between Baskan Yuksel and Ferrero Italy provided for delivery in September 2002, i.e. it was an a livrer contract for delivery in the following season.
645. The immediate background to the Early February Contracts may be summarised as follows. First, the Baskans had by then committed themselves to the implementation of the Scheme. The evidence for this consists of the bankruptcy application which they caused Indo-Med to make to the Turkish courts at the end of January, together with the documentary records of a stock-take purportedly made by Aksu Gida of hazelnuts to be transferred by Baskan Gida to Aksu Gida under the Scheme, dated 31^st January to 1^st February 2002. It is of course possible that the Aksu Gida stock certificates were all dishonestly backdated, like the Scheme documentation itself, but this cannot have been true of the bankruptcy application. The third piece of evidence to the same effect is Mr Abidali's email to Mr Dahod telling him on 8^th February that it was "all systems go" for the Scheme.
646. Secondly, the Baskans had by 5^th February obtained from the Banks most but not all of the money which they planned to put to fraudulent use. By 5^th February Baskan Gida had made five utilisation requests, in the aggregate sum of €20,829,290, and nine draw-downs in the aggregate sum of €18,453.614. Baskan Gida was to make a further draw-down of the full amount contained in the fifth utilisation request, on 8^th February, in the sum of €2,375,676, based on contracts 5100 and 5101.
647. The only use of contracts 5103, 4 and 5 made by the Baskans for the purpose of stealing money from the Banks was that they based their final utilisation request (the sixth) on 7^th February 2002, for €7,244,640 on contract 5104, and a final draw-down, against the letter of credit issued on that request, and backed by fictitious invoices purportedly issued under contract 5104, in the sum of €1,992,276, on 26^th February. Contracts 5103 and 5105 were not used at all.
648. The background from Ferrero's perspective was that weekly deliveries continued during the last two weeks in January and the first week in February at an apparently satisfactory and uniform rate of 744mt. The Ferrero Operating Defendants' stock positions as at 5^th February 2002 were approximately 2,650mt for Ferrero France, 6,724mt for Ferrero Germany and 9,274mt for Ferrero Italy. This was sufficient to meet those operating companies' production requirements for most or all of the rest of the 2001 season, but that left scope for at least 6,000mt to be ordered for the provision of a buffer stock for the beginning of the following season. Existing orders to Baskan Gida by the Ferrero Operating Defendants by the beginning of February amounted to 22,440mt, against which deliveries had been made by 5^th February in the aggregate of 14,448mt, leaving 7,992mt outstanding. Little of that was overdue, since the delivery periods for the whole of the Early December Contracts for 6,432mt had not yet passed, and a number of the earlier contracts in respect of which deliveries were outstanding had since been renewed with later delivery dates. Nonetheless, on the assumption that Baskan Gida continued to deliver at the weekly rates then being achieved, it would take at least ten weeks to complete all outstanding contracts. By contrast, virtually no deliveries were outstanding from Oltan Gida, and no new contracts had been placed with it since November 2001.
649. As for market prices, the price (converted to Euros) which Baskan Gida would be likely to have to pay in early February for paste kernels to TS3075 before processing to achieve Ferrero's specifications were actually higher than the prices negotiated for processed hazelnuts for export in the Early February Contracts. The result was, marked to market as at 5^th February, the prices charged in those contracts to Ferrero would have been insufficient to enable Baskan Gida even to purchase the hazelnuts necessary to fulfil those orders in the Turkish market, before incurring any processing or export costs. As for market movement, the Turkish lira price for the hazelnuts which Baskan Gida would need to purchase to fulfil the Early February Contracts had fallen a little since early December (for standard 2 from TL3,550 per mt to TL3,250) while due to exchange rate fluctuations, the Euro price had risen slightly, from 2,746 on 5^th December to 2,872 on 5^th February, using the tables derived from Mr Solak's contemporary notes. Whether viewed from a Euro or Turkish lira perspective, the fall in the market which both Melih Baskan and Mr Casale expected to occur in 2002 had yet to materialise.
650. The debate about the genuineness or commerciality of the Early February Contracts followed much the same course as the similar debate, which I have described, about the Early December Contracts, but with the added impetus to the Banks' cross-examination of the Ferrero witnesses' knowledge of what Melih Baskan had told them on 25^th January, about the prospect of Baskan Gida's early default.
651. The Ferrero witnesses' evidence about the making of the Early February Contracts was as follows. Mr Casale said that, as in December, he had been approached by Melih Baskan with an offer to make large supplies over long delivery periods, at prices which, while favourable at current market rates and within Ferrero's budget, reflected their common expectation that an excess of supply over demand would soon cause prices to fall. Mr Casale said that he asked Melih Baskan whether negotiations with Yapi Kredi were such as to make it at all likely that Baskan Gida would be able to deliver the amounts proposed, and was told that, at that stage, negotiations were still proceeding, so that it remained uncertain whether the Baskans would need to implement the scheme described at the meeting on 25^th January, but that if they did, Aksu Gida would be able to make the proposed deliveries in place of Baskan Gida.
652. Mr Casale then discussed the proposed terms with Mr Rosa Brunet, and they briefly discussed the matter with Mr Do. The view was taken that orders should be placed as Melih Baskan had suggested, for the following reasons. First, building further stocks accorded with Ferrero Industrial's policy to build up buffer stocks for use early in the following season. Secondly, the prices offered were both within budget and fair having regard to the parties' expectations of the market. Thirdly, regardless whether Baskan Gida was replaced by Aksu Gida, placing substantial orders would help preserve the supply chains, whether accessed by Baskan Gida or by Aksu Gida, upon which Ferrero depended for the obtaining of the highest quality hazelnuts. It was not Ferrero's understanding that the placing or withholding of the Early February Orders would make a difference to whether the Scheme had to be implemented or not. Their understanding was that this depended entirely on the progress of the negotiations between the Baskans and Yapi Kredi.
653. In support of the Banks' case that the Early February Contracts were not genuine (i.e. that Ferrero had no need for, or intention to take delivery under them) the Ferrero witnesses were pressed in cross-examination with the undoubted fact that the delivery periods agreed under these orders were even more clearly outside the ambit of the Soremartec Golden Rules than had been the periods agreed in the Early December Contracts. Again, the replies, which I broadly accept, were that these rules were favoured as much as in the breach as in the observance, as had occurred in the previous year, in particular in relation to the FKB Contracts. Furthermore, the Ferrero Italy purchase order for the subject matter of Contract 5105 was signed by Mr Fontana, head of Quality Control for Ferrero Italy, and it states the delivery period in clear terms. In fact, it appears that the later conversion of contract 5105 into an a livrer contract for the following season did take place under pressure from Soremartec or from quality control at Ferrero Italy, when it was re-made with Baskan Yuksel in early May. In the meantime, Ferrero did take delivery of the bulk of the subject matter of contract 5103 (only 192mt being outstanding by 9^th July), and of slightly more than one third of the amount ordered under contract 5104. Furthermore, Ferrero continued to place orders with other suppliers in every month between February and July inclusive 2002, including, in particular, a 1,200mt order on Oltan Gida in June.
654. In response to the question why, in early February, Ferrero did not place the bulk of its orders on Oltan Gida, which had by then supplied virtually all that had been ordered, in sharp contrast with Baskan Gida, Mr Casale referred to the quality problem experienced with Oltan Gida in late 2001, and to his preference to maintain the supply chain behind Baskan Gida, even if Aksu Gida became the intermediary between that supply chain and Ferrero.
655. Finally, in response to cross-examination based upon the very low prices reflected in the Early February Contracts (if marked to market as at that date), Mr Casale pointed to the fact that within the delivery periods specified, the market did fall substantially. Mr Rosa Brunet and Mr Do said that, from their position of relative detachment compared to that of Mr Casale, they had no recollection of appreciating that the prices negotiated between Mr Casale and Melih Baskan for the Early February Contracts were particularly low.
656. Two features stand out in relation to the negotiation of the Early February Contracts, which call for special attention. The first is that, when it is borne in mind that over 5,500mt of the subject matter of the Early December Contracts remained outstanding by early February, and that the market had yet to begin to fall significantly, the addition of a further combined tonnage of almost 6,000 in early February at similarly low prices increased Baskan Gida's uncovered short speculation with Ferrero to a tonnage in excess of 11,000, a truly remarkable level for a business for which, as Ferrero by then knew, faced possibly terminal difficulties due to the dispute with Yapi Kredi. As Mr Haas-Rickertsen was constrained to accept in cross-examination, the February orders really did represent the last desperate throw of the dice by the Baskans, and this must in my judgment have been apparent to Mr Casale, even if not to his more detached superiors.
657. The second feature is that the combination of deliveries outstanding by early February with the aggregate of the Early February Contracts constituted an order book for Baskan Gida and/or Aksu Gida in excess of 12,000mt, which it would be likely, on the track record thus far achieved in that season, to take the whole of the rest of the 2001 season to perform.
658. The following factors, relevant to the genuineness or commerciality of the Early February Contracts, emerge from that analysis. The first is that, from the Baskans' perspective, it made eminent commercial sense to seek to tie in Ferrero as a principal customer for Aksu Gida in the first months of its trading and, which is the other side of the same coin, for the Baskans to seek to avoid driving Ferrero into the arms of competitors during the necessarily vulnerable early part of Aksu's life.
659. The second is that the Baskans no doubt appreciated that there was a price to pay for obtaining that tie, namely a significant reduction in price for future deliveries from the current market price, all the more so when dealing with a counterparty with the same market view, namely that it was likely to fall. Thirdly, and most importantly, that commercial analysis would only have made sense to either of the negotiators, on the basis of a common assumption that the Baskans with their particular market view and large appetite for short speculation, were to be in control of Aksu Gida and, in all probability, the ultimate beneficial owners of it. I cannot conceive that Mr Casale would have thought that any wholly independent newcomer to the hazelnut export business would have been prepared to take on an 11,000mt short speculation against the market as the dominating aspect of the first four months of its trading activity. The only explanation for the apparent indifference of Mr Casale and his superiors to the question whether Baskan Gida or Aksu Gida would perform these large contracts is that they regarded both companies to be emanations of the same controlling mind and will, namely the senior members of the Baskan family.
660. None of those considerations lead to the conclusion that the Early February Contracts were not genuine, or that they are inexplicable as the product of a commercial negotiation. On the contrary, it is readily understandable why Ferrero should have been prepared not merely to have been tied in to taking on a new company as its principal supplier, but to have positively supported an arrangement under which a connected successor to Baskan Gida would start business with a full order book with which to soak up the stream of hazelnuts likely to come from the Baskan's established supply chain, so as to preserve that chain for the future as the best source of high quality Turkish hazelnuts then known to be available. In fact, the preservation of the supply chain which had until then provided Ferrero with the best quality supplies came across in the oral evidence time and time again as Ferrero's principal motive for behaviour which, from time to time, otherwise appears to be difficult to explain. Ferrero also had an established policy to avoid becoming over- dependant upon any single supplier. The threat to Baskan Gida's survival made support for Aksu Gida as its successor an obvious way of avoiding over- dependence on Oltan Gida, which was the only competing Turkish supplier with which Ferrero wanted to do large scale business at the time.
661. The more difficult question is whether that motive was a sufficient stimulus to lead Ferrero into knowing participation in a dishonest and unlawful design against the Banks, whereby the Baskans sought to achieve their own objective of remaining the principal hazelnut exporters in Turkey by using hazelnuts purchased with the Banks' money as Aksu Gida's initial working capital.
662. The effect of Ferrero placing purchase orders and/or signing contracts for almost 6,000mt of hazelnuts in early February was in fact to put into the Baskans' hands documents which, to the Banks or to any other third party, appeared to recognise a bona fide intention on Baskan Gida's part to supply, and on Ferrero's part to pay Baskan Gida for, a quantity of goods justifying the advance of many millions of Euros by way of working capital finance, spread over several months. Yet Mr Casale, Mr Rosa Brunet and Mr Do all knew that the Baskans did not regard Baskan Gida as having bona fide undertaken any such commitment, since the plan presented on 25^th January, if it took place, necessarily involved a large scale default by Baskan Gida. The sooner the plan was implemented, the larger would be the scale of the default.
663. The knowledge, shared by Mr Casale and Mr Rosa Brunet, that the Scheme probably involved an element of deception of Yapi Kredi certainly ought to have generated in their minds the question whether the planned large scale default by Baskan Gida in complying with the Early February Contracts was to be shared with the Banks, if and when those contracts or the purchase orders which accompanied them were used for the raising of finance. Neither Mr Casale nor Mr Rosa Brunet suggested that they asked this question, either of the Baskans or of the Banks. On the contrary, their evidence was that the question simply never occurred to them. Mr Casale in particular said that he gave little thought to the extent to which Baskan Gida really needed working capital funding for its business with Ferrero, because of Ferrero's readiness to pay in full on delivery.
664. In my judgment Mr Casale and Mr Rosa Brunet probably did ask themselves (although not necessarily each other) how, if at all, the Baskans proposed to deal with the Banks in relation to the obtaining of working capital finance on the back of the Early February Contracts. They did at least suspect that the Banks might be no more made privy to the Baskans' plans than would Yapi Kredi, if for no other reason than by doing so, the Baskans would risk Yapi Kredi learning on the banking grapevine that they were to be in control of Aksu Gida.
665. In my judgment Mr Casale and Mr Rosa Brunet simply ignored these potential difficulties, when committing Ferrero to a course of action which they genuinely believed to be in Ferrero's best interests. I think it most unlikely that they believed that the Baskans would use the Early February Contracts or, for that matter, the late January renewals in the specifically fraudulent way which they actually did. I have described how the late January renewals were used. As for the Early February Contracts, the Banks used contract 5104 as the basis for the final utilisation request, leading to the issue of a letter of credit in the sum of €7.244 million, but in the event only drew down €1.992 million under it, by means of pro-forma invoices which pretended, contrary to the truth, that Baskan Gida rather than Aksu Gida was to make deliveries under that contract. By that time, namely 26^th February, Baskan Gida had already handed over the further conduct of the export business with Ferrero to Aksu Gida.
666. The main reason why I doubt whether anyone at Ferrero believed or even specifically suspected that such a use was to be made of the late January and Early February Contracts was because of the obvious risk to which that conduct would expose Ferrero, in terms of having to pay twice for the same goods. There is no evidence whatever that either Mr Rosa Brunet or Mr Casale had any corrupt motive for exposing Ferrero to risk, let alone to liability, while assisting the Baskans. On the contrary, when at the beginning of March the possibility that Ferrero might be so exposed first came to Mr Casale's attention, in circumstances which I shall describe, I consider that it came to him as a complete surprise.
667. In my judgment, the thinking of both Mr Rosa Brunet and of Mr Casale at the beginning of February was that it was sufficient for them to conclude, as they did, that supporting the project to rescue the Baskans' hazelnut business and their supply chain from the threat posed by Yapi Kredi was in Ferrero's best interests, and that other entities having commercial relationships with the Baskans could be left to look after themselves.

The Split Invoices – Round 1

668. It will be recalled that Baskan Gida's first utilisation request on 21^st December 2001 had been based upon Contracts 5094 to 5097, and its first draw-down, made on 3^rd January 2002, was supported by Contracts 5094, 5096 and 5097. By the time that Baskan Gida became liable to make repayment of that first draw-down in early February, it had already invoiced Ferrero Italy under Contracts 5096 and 5097, directing it to pay not to the Collection Account, but to separate Baskan Gida accounts, while at the same time sending false copy invoices to the Banks purporting to direct payment to the Collection Account. The Banks therefore expected to receive payments from Ferrero Italy at the rate of €2,555 and €2,556 per mt (being the prices agreed under Contracts 5096 and 5097 respectively).
669. In order to avoid sending warning signs to the Banks of its invoicing fraud, it was therefore necessary for the Baskans to arrange for an income stream to be paid from Ferrero Italy to the Collection Account, in respect of other deliveries. Baskan Gida was at this time making deliveries under Contract 5093. This had been used as the basis of its second utilisation request on 8^th January 2002, and draw-downs starting on 15^th January, but the 45 day period for repayment in relation to those was not yet imminent. The difficulty for Baskan Gida was that the price under Contract 5093 was €2,739 per mt, so that a direction to Ferrero Italy to pay amounts falling due under those invoices would generate the wrong income stream, and thereby put the Banks on enquiry that something might be amiss.
670. The Baskans resolved this difficulty by use of what have come to be known as the split invoices. In relation to each 24 mt delivery to Ferrero Italy, Baskan Gida sent two invoices. The first, priced at €2,555 per mt (when simulating Contract 5096) contained the prescribed endorsement directing Ferrero Italy to pay to the Collection Account at KBC for Euros. The second, priced at €184 per mt, directed payment to Baskan Gida's account at Garanti Bankasi in Turkey. In relation to each delivery, both invoices identified the same contract number (5093) and the same batch number, but used different invoice dates. The aggregate of the two invoices precisely matched the price payable under Contract 5093, of €2,739. A similar split was engineered so as to create the impression of payments under Contract 5097 (namely €2,556 per mt).
671. Ferrero Italy duly paid the amounts demanded under the split invoices, thereby obligingly creating the appearance of an income stream to the Euro Collection Account at KBC precisely matching what the Banks expected to receive by reason of the assignment of invoices under Contracts 5096 and 5097.
672. The first set of split invoices sent to Ferrero Italy consisted of six pairs, faxed on 11^th February, together with copies of other invoices for Ferrero France and Ferrero Germany. The split invoices were all signed off by Mr Rosa Brunet, in Mr Casale's absence, on 12^th February. The process was repeated a week later, with a larger batch of eleven pairs, both sent to Ferrero Industrial and signed by way of approval for payment by Mr Casale on 18^th February.
673. I have already described, on the basis of Mrs Di Matteo's evidence, the procedure for approving and paying invoices within Ferrero Italy, beginning with signed approval from Mr Casale's department at Ferrero Industrial, which Ferrero Italy treated as its purchasing department, passing to its import department, its accounts department and finally to the treasury department for payment. Nothing in Ferrero's documentary disclosure suggested that the receipt, approval and payment of these unusual pairs of invoices caused any surprise, or led to any inquiries being made at the time. The only witnesses who gave evidence about them were Mr Rosa Brunet, Mr Casale and Mrs Di Matteo.
674. Mrs Di Matteo's evidence, which I accept, was that by the time these invoices reached the treasury department they would have been treated as having been scrutinised and approved by purchasing, imports and accounts, so that it would have been no part of treasury's job to compare contract and batch numbers, or do anything which would have alerted them to the fact that they were pairs of invoices for single deliveries. She said that treasury occasionally received invoices for unusually small amounts, but understood those to involve price adjustments in relation to earlier payments.
675. Mr Rosa Brunet's evidence, which I also accept, was that so far as he could recall he signed the first five pairs as a matter of ordinary routine, having had them put in front of him for signature by someone in Mr Casale's department, in Mr Casale's absence. He could recall nothing having happened at the time which alerted him to their unusual nature, still less to their having formed part of the means whereby Baskan Gida was perpetrating a fraud on the Banks. There is no evidence that Mr Rosa Brunet was involved in approving the second group of pairs, received on 18^th February, for payment.
676. Mr Casale's evidence was to much the same effect as that of Mr Rosa Brunet, but in order to appraise it in its full context, it is necessary first to describe his involvement in two almost simultaneous important events which occurred on 13^th February, in between Ferrero's receipt of the two groups of split invoices. For clarity however, I state my conclusions about Ferrero's reaction to receiving the first group, in Mr Casale's absence, on 11^th February.
677. I consider it inconceivable that the junior staff in Ferrero Industrial's office who first reviewed these six pairs of split invoices would have failed to notice that, in relation to each pair, most unusually Baskan Gida was sending two different invoices for the same 24 mt batch, the effect of which was to request Ferrero Italy to pay more than 90% of the price for each batch to one bank and less than 10% of the price to another. The relevant employee (whose identity cannot now be ascertained) would have ascertained that the batch to which each pair of invoices related had both been ordered and delivered. Each invoice identified the original Ferrero Italy purchase order (number 1121188), and Ferrero Industrial had readily available means of knowing that delivery had by then been made FOT at Giresun. Since the aggregate claimed under each pair of invoices was the correct price for that batch, I think it probable that the relevant employee would simply have included all six pairs in a larger group of invoices and other documents for Mr Rosa Brunet's signature, on the assumption that, for reasons which he neither knew nor thought it necessary to ascertain, Baskan Gida wished to split the payments to be received from Ferrero in relation to those deliveries between different bank accounts. Had Mr Casale been in the office, rather than (as he was in fact) in Bulgaria, I would have expected the relevant employee to have mentioned the pairs of split invoices to him as something unusual, but consider it probable that he thought it insufficiently important to trouble Mr Casale about while he was away, or, still less, to bring them to the attention of the more senior Mr Rosa Brunet. I would not expect him to have studied or understood the small print on the prescribed endorsement, which described Baskan Gida as having:

"assigned to the Agent all of our rights, benefit and interest in this contract, including any money payable to us under or in connection with it."

The Ferrero Industrial employee would in my judgment have been unlikely to be concerned with the very small print of a foreign language endorsement which, however relevant to other departments, had nothing to do with the question whether Ferrero was liable to make the aggregate payments represented by the six pairs of split invoices. I think it highly unlikely that the relevant employee would have taken the trouble to check back against earlier invoices and earlier orders, so as to discover that the €2,555 per mt payable under the larger of the invoices in each pair precisely reflected a price payable (but already paid) in respect of a different purchase order.

678. I consider that, at the highest, similar thought processes would have occurred in the minds of the staff in the export and accounts departments considering the first group of split invoices. Again, they would have appeared unusual, and accounts would no doubt have satisfied themselves that the aggregate amount demanded under each pair represented the full extent of Ferrero's liability for payment. By then, the invoices were signed off by Mr Rosa Brunet, and I consider that this would have been sufficient comfort for the staff in accounts for them to pass the invoices on to treasury for payment, without further inquiry.
679. In reaching those conclusions I have not ignored the fact that Ferrero called no witnesses from the accounts department of Ferrero Italy, or anyone junior to Mr Casale at Ferrero Industrial. While adverse inferences may be drawn from the failure of a party to call an apparently available witness, the presentation of these pairs of split invoices to members of Ferrero's staff who had, at that stage, no reason to be suspicious of the Baskans' integrity does not in my judgment give rise to any sufficient prima facie case to justify the drawing of an adverse inference in this context.
680. The highest that the case can be put against Ferrero staff other than Mr Rosa Brunet and Mr Casale is that the small print in the English language form of prescribed endorsement on the larger of each of the pairs of invoices does not, if thought about, lie easily with a splitting between different banks of the amount payable for a single delivery under the same contract. But again, I consider it unlikely that this foreign small print would have been studied closely by relatively junior accounts staff. Even if it had, the most suspicious imaginable person would have thought only that the Baskans were for some unexplained reason directing payment of less than 10% of the invoice to a prima facie inappropriate recipient. This is hardly the stuff of fraud, and the fact that these invoices were promptly paid strongly suggests that no alarm bells rang in Ferrero to the effect that, quoad that small percentage, Ferrero might thereby have exposed itself to a risk of having to pay twice. It is only the possession of a very large quantity of hindsight that reveals the true dishonest purpose behind the splitting of the invoices.

13th February 2002 – Framework Contract Amendment and Aksu Gida Takeover

681. In early February the Banks were still trying to clear up loose ends arising out of the completion of the Facility, even though over €21 million had been drawn down under it by Baskan Gida by 8^th February. There were two particular outstanding matters of administrative concern. The first was that the Banks had yet to receive acknowledgments from the Ferrero Operating Defendants other than Ferrero France of the correction to the KBC account number notified to each of them by KBC's faxes sent on 21^st December. The second was that the Framework Contract signed by Ferrero Industrial on 15^th October 2001 had, at clause 6.2, omitted the Collection Account details for both the US dollar and Euro payments.
682. On 11^th February Mr Lewis of KBC chased Ismet Baskan in relation to these and other items, asking in particular for a signed copy of the page of the Framework Contract with the relevant account details completed. Mr Lewis appears to have overlooked the fact that Ferrero France had acknowledged receipt of the corrected account details on 10^th January.
683. There was little that the Baskans could do in relation to the request for acknowledgements from Ferrero of the corrected account details, since they had, of course, forged the Acknowledgments of Assignment from the Ferrero operating companies which KBC's 21^st December faxes had been designed to correct. The Baskans therefore let that sleeping dog lie. Nonetheless they sought to comfort the Banks in relation to the Framework Contract. On 12^th February Ismet Baskan telephoned Mr Casale about it, and on 13^th February sent him a completed version of clause 6.2, by fax and by email, describing it as part of the "framework contract (General Terms And Conditions For Sale Of Hazelnuts)" and continuing:

"As I told you on the phone, this is a missing document in bank files (actually they have it without bank accounts) so they want us to have it signed by you and us. I am also faxing the same document[s] to your office with our signature on it. It will be very helpful to us if we have this document signed by you and Mr Brunet within today."

Clause 6.2, as completed, reads as follows:

"Payment is made by the Ferrero Company, which has issued the Purchase Order. All payments made by Ferrero Brazil, Australia and USA to Baskan are in US dollars to KBC Bank NV, New York, in account KBC Bank NV London to account number 10775101 in favor of Akbank TAS as Collecting Agent for Baskan Gida Sanayi Ve Pazarlama AS account number 33052911 (or to such other bank account as may be advised by Baskan to Ferrero from time to time) quoting Baskan's invoice number and Ferrero's name.

All payments made by Ferrero Italy, Germany, Poland and France to Baskan are in Euros to KBC Bank NV Brussels in account KBC Bank NV London to account number 480-9039071-85 in favour of Akbank TAS as Collecting Agent for Baskan Gida Sanayi Ve Pazarlama AS account number 63052911 (or to such other bank account as may be advised by Baskan to Ferrero from time to time) quoting Baskan's invoice number and Ferrero's name."

684. Mr Casale was at the time on business in Plovdiv, Bulgaria. The email was sent to him direct, and the fax, with its enclosure signed on behalf of Baskan Gida, to Ferrero Industrial in Italy. Both reached their destinations on the morning of 13^th February. One of Mr Casale's subordinates at Ferrero Industrial faxed the enclosed page of the Framework Contract to Mr Casale in Plovdiv, and he signed and then returned it to Ferrero Industrial just before 3.30 in the afternoon. At some time during the next hour and a half Mr Rosa Brunet also signed it, and it was duly faxed back to Baskan Gida at 5 o'clock, and then forwarded to the Banks by Baskan Gida later that evening.
685. On the same day the Baskans implemented the next important stage in the Scheme, namely the request to Ferrero that it contract to purchase from Aksu Gida all hazelnuts the subject matter of outstanding contracts with Baskan Gida. The request took the form of faxed letters ("the Transfer Letters") on Baskan Gida headed paper addressed to each of Ferrero Italy, Ferrero France and Ferrero Germany, with copies in each case to Ferrero Industrial, containing the following identical text:

"Dear Sirs,

Due to new circumstances, we kindly request you to change our outstanding contracts as listed below to the following company who will takeover and respect all our commitments as per each contract in regard to quality, shipment period, price and everything.

We kindly request you to recheck and confirm the following quantities for which Messrs Aksu Gida will send its confirmation and contracts.

Best Regards

Baskan A.S.

[signed]

NAME OF COMPANY

AKSU GIDA SANAYI & TICARET LTD.STI.

Fatih Caddesi No.24, Ortaklar Ishani Kat.2/23

28100 Giresun/Turkey

VAT No Giresun VD 0380136672"

686. Each Baskan Gida Transfer Letter ended with a table identifying by order number and quantity the hazelnuts which Baskan Gida believed to be outstanding for delivery to that Ferrero operating company. Although this does not appear from the face of the Transfer Letters, Baskan Gida excluded from those tables amounts which were scheduled for delivery in the following few days, which continued to be delivered and invoiced by Baskan Gida. By 4.30 on the afternoon of 13^th February Mr Altare was in the process of faxing on to Ferrero France and Ferrero Germany copies of the Transfer Letters received from Baskan Gida.
687. The Banks claim that by signing the page of the Framework Contract containing the completed designated account details at a time when they knew, from Baskan Gida's almost simultaneous Transfer Letters, that there was to be no continuing contractual relationship between Baskan Gida and any Ferrero company, Mr Rosa Brunet and Mr Casale both made a fraudulent representation to the contrary, and took an important step in the furtherance of Baskan Gida's fraud. In summary, the Banks claim that Ferrero told them on the afternoon of 13^th February that it was business as usual with Baskan Gida, at precisely the moment when they knew that business with Baskan Gida was at an end.
688. Both Mr Rosa Brunet and Mr Casale maintained in their written and oral evidence that, at the time when they added their signatures to the relevant page of the Framework Contract, neither of them knew about the Transfer Letters whereby the Baskans had, in effect, pulled the trigger on the Scheme. The Scheme still remained in their minds just a possibility, as it had been since first being mentioned on 25^th January. Furthermore, they both maintained that even if they had been aware of Baskan Gida's Transfer Letters when they signed the Framework Contract, re-signing the latter still involved no misrepresentation, since there were outstanding deliveries and payments to be made under the Ferrero/Baskan Gida relationship during the period of the handover to Aksu Gida.
689. The Banks also alleged that by signing clause 6.2 of the Framework Contract Mr Casale and Mr Rosa Brunet also dishonestly represented that Ferrero still intended to make all payments for hazelnut deliveries into the Collection Accounts, whereas in fact they anticipated making all or most future payments to Aksu Gida, or to other Baskan Gida accounts. In response to cross examination to that effect, both Mr Rosa Brunet and Mr Casale pointed to the phrase in clause 6.2 of the Framework Contract which stated:

"or to such other bank account as may be advised by Baskan to Ferrero from time to time."

They also pointed to the fact that Baskan Gida did during the following few days send a number of invoices containing the prescribed endorsement for payment to the Collection Account for Euros including (but not limited to) the second group of split invoices. In fact, Ferrero paid more Baskan Gida invoices to the Collection Accounts after, than before, 13^th February.

690. I did not find Mr Rosa Brunet's account, in cross-examination, of the circumstances in which he signed the relevant page of the Framework Contract on 13^th February to be convincing. Having started by asserting that, when he signed the Framework Contract, he still regarded the transfer to Aksu as only a possibility, he then said that he could not recall whether he had been told about the Transfer Letters before he signed, but then sought to take refuge in a denial, when he signed the relevant page (which already bore Mr Casale's signature), that he made any connection between that page and the Framework Contract, describing it just as some formal document put in front of him for his signature as a matter of routine. This last piece of his oral evidence ran directly counter to his witness statement, where, at paragraph 312, he said this:

"I do remember that the fact that we were being asked to sign an amendment page of the General Terms and Conditions gave me the impression that the financing arrangement between the Banks and Baskan Gida had not been fully finalised."

691. In my judgment Mr Rosa Brunet probably did know when signing the relevant page of the Framework Contract that the Transfer Letters had been received, and did know that he was signing a page of the Framework Contract, although I think it unlikely that he took the trouble to remind himself of the detailed terms of the Framework Contract as a whole. Since the relevant page had already been signed by Mr Casale, and since no executive decision was being sought from Mr Rosa Brunet, I think it probable that he treated the addition of his signature as nothing more than an administrative formality, in relation to a clause which Mr Casale had already approved, forming part of a contract to which he had, on Mr Casale's advice, already decided to commit Ferrero four months previously. I think it unlikely that he gave any detailed consideration to the precise terms of clause 6.2, or, in particular, to the extract quoted above upon which both he and Mr Casale thought it useful to place emphasis in their evidence.
692. As to Mr Casale, I found his account, save in one respect, less unsatisfactory than that of Mr Rosa Brunet. He had been given a day's warning of the Baskans' request that he sign the relevant page of the Framework Contract, by telephone on 12^th February. He had also had the relevant page emailed to him direct by Baskan Gida before receiving the faxed page which he actually signed, so that he had time to consider, in his usual careful way, whether he should sign it on behalf of Ferrero Industrial. Since he was away on business unrelated to hazelnuts on both days, I consider it probable that he signed the relevant page before being told of the receipt in Italy of the Transfer Letters. I reach this conclusion, on balance, because the clear impression which I derive from the whole of the evidence (oral and written) about the Transfer Letters is that no-one at Ferrero Industrial appears either to have been surprised by their arrival, or to have thought that they called for any executive decision by Mr Casale or anyone else in authority. The decision to go along with the transfer to Aksu had already been made, albeit on a contingent basis, prior to 13^th February, so that all that was done by Ferrero Industrial when the copies of the Transfer Letters were received was to forward them on to the relevant operating companies, and this was done by Mr Altare.
693. The only unsatisfactory aspect of Mr Casale's evidence on this point was his attempt to suggest that Mr Altare would have dealt with the Transfer Letters without even informing anyone in authority at Ferrero Industrial of their receipt. In my judgment it is probable that Mr Altare informed Mr Rosa Brunet as soon as they arrived. The evidence was that their offices were near to each other, and that Mr Rosa Brunet's door was always open.
694. I reject the Banks' case that, whether viewed in isolation or against the background of the events which I have already described, the almost simultaneous receipt from Baskan Gida of the Transfer Letters and the request to sign the completed clause 6.2 of the Framework Contract were sufficient to bring home to anyone at Ferrero Industrial knowledge (in the relevant sense) that the Baskans were in the process of perpetrating a fraud on the Banks. For Mr Rosa Brunet the signature of the relevant page was simply too much of an administrative detail to ring the necessary alarm bells. Mr Casale was, in Bulgaria, too detached from the simultaneity of the two events.
695. Furthermore, I also reject the Banks' case that by countersigning the relevant page of the Framework Contract, Ferrero made any misrepresentation, let alone a fraudulent one. The Framework Contract purported to set out the trading relationship between Ferrero and Baskan Gida as at the date of its signature, 15^th October 2001. In one respect, namely the incomplete nature of clause 6.2, it did not do so. By February 2002 the Banks knew, first, that Ferrero had been asked but refused to promise not to vary that trading relationship without the Banks' consent and that, in November 2001, the 45 day period for payment had represented a major change in the nature of the relationship described in the Framework Contract. While therefore the Banks were to a limited extent entitled (as they did) to take some comfort from the Ferrero signature against clause 6.2 that Ferrero was still playing its part in the structure of the Facility, the signature of the relevant page went no further than that. Clause 6.2 expressly entitled Ferrero to pay amounts to such accounts as Baskan Gida should from time to time direct, and this is exactly what, during the following week, Ferrero continued to do.
696. The low key way in which Ferrero Industrial responded to receiving the Transfer Letters sheds real light on the extent to which the transfer to Aksu Gida was by 13^th February regarded within Ferrero as contingent. Although both Mr Rosa Brunet and Mr Casale described such a transfer as still only a "possibility" by that date, in my judgment they both regarded it as a probability, and their uncertainty about it at that stage mainly related to the likely date upon which it would be implemented. It also suggests that Mr Rosa Brunet and Mr Casale did indeed regard the transfer as amounting to no significant interruption with business as usual with the Baskans, as the persons who Ferrero thought controlled Aksu Gida, and the conduct of the hazelnut business following the transfer.
697. The evidence as to how Ferrero Germany responded to receiving the Transfer Letter firmly supports that conclusion. Mr Bolowich received the Transfer Letter addressed to Ferrero Germany on 13^th February, and on 14^th February his secretary Hanne Schlechter wrote to Melih Baskan at Aksu Gida attaching the first replacement contracts covering outstanding orders due from Baskan Gida and made a routine inquiry about whether part of the Aksu address could be omitted, due to what she described as "edp problems" (which I understand to be electronic document production problems). She expressed the letter as if coming from Mr Bolowich, and copied it to Mr Casale. Later, on 18^th February Mr Bolowich himself sent a manuscript fax message to Mr Melih Baskan headed "Aksu Gida – invoices", asking him to use the new order numbers and not the old Baskan numbers.
698. In his witness statement, before being shown his secretary's letter to Melih Baskan, Mr Bolowich stated that he had obtained oral confirmation from Mr Casale of the decision that new contracts should be placed with Aksu Gida, having not been informed about the transfer in advance, and that he did not recall giving any thought to whether there was any relationship between Baskan Gida and Aksu Gida, or what that relationship might be. In cross-examination, after being shown his secretary's letter and his own manuscript fax message, (by the first of which he expressed himself to be considerably surprised), he eventually agreed with Mr Wardell that Mr Casale must have instructed him to communicate with Melih Baskan about matters concerning Aksu Gida. In re-examination he said that he could not actually recall such a conversation, but this did not undermine the force of his assumption that he must have been told by Mr Casale to contact Melih Baskan.
699. In his own cross-examination, Mr Casale said that he had told Mr Bolowich that the person to contact in relation to Aksu Gida was Mr Franko, and his explanation for Mr Bolowich's and Ms Schlechter's communications with Melih Baskan was that, in effect, Mr Bolowich contacted the person he was used to dealing with. In that respect I prefer Mr Bolowich's evidence, although I accept that there later came a time when, at the Baskans' request, Mr Casale arranged for routine written communications concerning Aksu Gida to be made to Mr Franko, as part of the Baskans' scheme to appear to withdraw from the hazelnut business while, in reality, remaining in control of it. In the same way, Mrs Pay's evidence was that the Baskans no longer frequented the Giresun factory and office complex after the transfer to Aksu, but continued to act as her supervising managers thereafter, over the telephone.

The Split Invoices – Round 2

700. On 18^th February Baskan Gida sent a further batch of invoices to Ferrero for payment by Ferrero Italy, relating to deliveries made at Giresun between 15^th and 17^th February inclusive. The majority of this batch consisted of ten further pairs of split invoices, all relating to contract 5093, the effect of which was to split the payment of €65,736 in each case, so as to create an income stream flowing to the Collection Account at KBC having the appearance of payments under contracts 5096 and 5097. Mr Casale signed these invoices on the same day, and they were approved by treasury for payment, and paid on the following day.
701. Mr Casale returned from Bulgaria to Alba on Friday 15^th February, and in his witness statement said that he had learnt of the Transfer Letters on that day. He therefore signed the second batch of split invoices on his first full working day at Alba after learning of the transfer to Aksu Gida. Bearing in mind that Ferrero was in the process of cooperating with that transfer, it seems to me inconceivable that anything other than the most careful attention to incoming invoices, whether from Baskan Gida or Aksu Gida, was given by someone at Ferrero Industrial and/or Ferrero Italy, if for no other reason than to avoid payment being made to the wrong company. The need for care was reinforced by Ferrero Industrial's and Ferrero Italy's receipt of a fax from Baskan Gida on 14^th February correcting its earlier statement of the amounts outstanding under existing contracts with Ferrero Italy, and providing a schedule of deliveries which, notwithstanding the Transfer Letter, Baskan Gida intended still to make under two specified Ferrero Italy purchase orders, corresponding with Contracts 5093 and 5094. The fax was addressed specifically to Mr Rosa Brunet and Mr Casale.
702. It by no means follows that Mr Casale would have taken it upon himself to scrutinise incoming invoices from Baskan Gida, in order to ascertain whether they were properly payable. I would expect him to have required that task to be undertaken by one of his subordinates. Ferrero Industrial received a very large number of invoices from Baskan Gida and, no doubt, from other suppliers, and it cannot have been the job of Mr Casale, with his frequent absences from the office on business, to have conducted such checks.
703. Nonetheless, I would expect him to have asked the person who presented him with the bundle of Baskan Gida invoices which he signed on 18^th February whether they had been properly checked in the context of the handover to Aksu Gida, and I consider it probable that his staff would have told him first, that they were all in respect of deliveries which had been made, so that they were therefore payable, but secondly that ten pairs of them split the amount payable under ten specific deliveries so as to request payment to two different bank accounts, and I would expect Mr Casale to have ascertained, by inquiry or by checking himself, the 92% / 8% approximate split thereby occasioned. He would also probably have been told that this was not the first occasion on which Baskan Gida had submitted invoices split in this way, and that a smaller batch had been received and authorised for payment, while he was away, during the previous week.
704. The critical question is whether I should infer (as Mr Wardell submitted that I should) that Mr Casale appreciated that by splitting the invoices in this way, Baskan Gida were up to no good. Mr Wardell submitted that Mr Casale would have connected the amounts payable under the invoices directing payment to the designated account at KBC with the amount payable under Contracts 5096 and 5097, and thereby appreciated that the Baskans were seeking to create an income stream masquerading as a series of payments under those contracts, on the basis that information about the amount payable per delivery under each contract would be something which the diligent Mr Casale would carry round in his head. Alternatively he submitted that split invoices were so unusual, whether coming from Baskan Gida or any supplier, that Mr Casale would have carried out the relevant research of the office files necessary to make that connection, or alternatively studied the English small print in the prescribed endorsement on the invoices directing payment to KBC, so as to appreciate that Baskan Gida could not properly direct payment of only part of the purchase price to the Banks. Thus he would have discovered that the Baskans were engaged in an invoicing fraud against the Banks, having ascertained that invoices delivered to Ferrero Industrial under Contracts 5096 and 5097 had directed payment not to the Banks, but to a different Baskan Gida account. It followed, submitted Mr Wardell, that Mr Casale lied when professing an inability to recall this event in February 2002.
705. This analysis was well argued, and there is much to be said for it. In particular it draws upon Mr Casale's undoubted thoroughness and attention to detail, in a context where such attention was called for by the circumstances of the transfer to Aksu Gida then taking place. Nonetheless, I have not in the end been persuaded by it, for two main reasons. The first is that it is critically informed by hindsight, and by knowledge of the way in which, behind the scenes, the Baskans were in fact defrauding the Banks by the use of false copy invoices. The second is that if carried to its logical conclusion, any understanding by Mr Casale that the Baskans were in the process of perpetrating an invoicing fraud on the Banks, such that money which Ferrero should have been paying to the Collection Accounts was being mis-directed into the Baskans' pocket must have led Mr Casale to have appreciate that Ferrero itself was a potential victim of that fraud, by reason of its exposure to a liability to pay twice, if and so far as it wrongly paid the Baskans.
706. Mr Wardell did not dispute the second part of that analysis, but submitted that, in their desire to assist the Baskans in their fraud against the Banks, Mr Casale and Mr Rosa Brunet were prepared to expose Ferrero to risk, for reasons which he was driven to accept necessarily involved the obtaining of some private corrupt advantage, of which there was no evidence whatsoever.
707. In reaching this conclusion I have not ignored Mr Wardell's submission that, regardless whether the split invoices were, objectively, sufficient to put Mr Casale on notice that the Baskans were deceiving the Banks, I should nonetheless infer that Mr Casale appreciated this by reason of his having given a false account of the circumstances in which he signed off on the second group of split invoices. The gist of his evidence was that the first occasion upon which the split invoices came to his attention was when the Banks' claims that they had been underpaid by Ferrero came to be investigated internally within Ferrero, in late May or June 2002. He said that, in February, not only did he not suspect that the Baskans were using the split invoices to deceive the Banks, but that he did not notice, or have drawn to his attention, the fact that split invoices were being used at all. I have rejected the second part of that evidence. Mr Wardell submitted that I should conclude that the second part was a deliberate lie just as much as the first part, because the use of split invoices was so unusual that Mr Casale could not have forgotten having had them drawn to his attention when asked to sign them.
708. While Mr Casale was, to that extent, wrong in his evidence, I am not persuaded that he was lying. My conclusion is that, having appreciated or having had drawn to his attention the splitting of the invoices for ten, or all sixteen, of the deliveries under Contract 5093, he simply gave no sufficient thought to the reason why Baskan Gida might have done it to have alerted himself to possible wrongdoing on their part and, their reasoning being none of his business, gave no further thought to it, and forgot about it, until the invoices were investigated in the context of the Banks' claims later in 2002. Having thus forgotten about them, it is easy to understand why when giving evidence, the absence of any recollection of having noticed them at the time led him to the wrong assumption that the split invoices did not come to his attention at that time. Absent a serious suspicion that the Baskans were up to no good, the episode during which he signed those invoices cannot have taken more than a minute or two, during a busy day primarily directed to more important matters.

The Position of the Parties in Late February 2002

709. It is instructive to pause in this narrative of very detailed facts to consider the respective positions of the parties to this dispute during the second half of February 2002, after the transfer of the hazelnut business to Aksu Gida and the notification of that transfer to Ferrero. Beginning with the position on the ground in Turkey, it was for the most part business as usual. Baskan Gida had, prior to the transfer, been delivering to Ferrero at a rate of roughly 500mt a week from Giresun and its other factories. By the end of February, Aksu Gida deliveries amounted to 624mt out of total of just under 2,000mt delivered by Baskan Gida and Aksu Gida together, and in March the total of just over 2,000mt was entirely delivered by Aksu Gida. The workforce and middle management remained the same, and Aksu Gida even continued to use the name Baskan as part of its business logo. The Baskans continued behind the scenes to run the business albeit, as described by Mrs Pay, on the telephone rather than by attending at the premises. She said that, during the period when the hazelnut business was being conducted in Aksu Gida's name, her supervisor was Fuat Baskan.
710. The formal process of leasing the business premises to Aksu Gida and recording the transfer of hazelnuts and plant by invoices and stock records was largely completed by mid-February (assuming, without being able to decide that the dates on the documents are genuine, by contrast with the dates on the main transfer documentation).
711. There was however a real change in relation to Warehouse 2. Prior to 9^th February there had been a steady stream of deposits and withdrawals of hazelnuts into and out of Warehouse 2 maintaining aggregate stocks there between 1,000mt and 2,500mt. Deposits ceased after 9^th February, but withdrawals continued until 16^th February, reducing stocks from 1,972mt to 844mt, after which nothing went into or out of Warehouse 2.
712. There is no reason to suppose that the picture which I have described (other than in relation to Warehouse 2) was not apparent to Ferrero during late February. During late February it was Mr Altare's turn to be Ferrero's representative in Turkey. It is impossible to be sure how he divided his time between the Giresun factory and other plants in Turkey supplying hazelnuts to Ferrero, but likely that he spent between two and eight days in Giresun during late February.
713. At the same time, the Baskans succeeded in keeping the Banks completely in the dark about the transfer to Aksu Gida, and all other aspects of their fraud, until 1^st March. The Banks thought that it was truly business as usual, in all respects in accordance with the terms of the Facility. That is not to say that there were not what appeared to the Banks to be glitches, such as Baskan Gida's delay in obtaining Ferrero's signature to the completed clause 6.2 of the Framework Contract, and the failure of any Ferrero company other than Ferrero France to acknowledge receipt of the same correction to the Acknowledgements of Assignment.
714. More serious, at least at first sight, was a notification from Akbank that Ferrero had paid the first two payments which the Banks expected to receive under the Facility into a Baskan Gida account rather than to the Collection Account. The Banks had received purported copy invoices containing the prescribed endorsement requiring payment to the Collection Account by Ferrero, but of course Ferrero had received no such instruction in the invoices actually sent. Akbank's notification reached the Banks on 12^th February and, rather than embarrass their customer Baskan Gida by raising the matter directly with Ferrero, the Banks decided to seek an explanation from the Baskans first. By 15^th February the Baskans had, both directly and through PRI persuaded the Banks that the apparent misdirection of the payments was entirely the result of an administrative error by Ferrero, which would not occur again, and the payments were transferred at the Baskans' request to the Collection Account on 18^th February. Needless to say, the Baskans' story that they had sorted that matter out in discussion with Ferrero was entirely untrue.
715. Meanwhile, the Banks started to receive payments into the Collection Account from Ferrero largely, but not entirely, as a result of the sending of the split invoices. The effect was that the Banks received repayment of the whole of the first draw-down on the last contractually permissible day, namely 20^th February.
716. Another respect in which Baskan Gida were to the knowledge of the Banks not wholly in compliance with the terms of the Facility was in their failure to put in place in due time a new Capex facility with Yapi Kredi so as to remedy what the Banks perceived to be the unsatisfactory basis upon which, until then, Baskan Gida had funded its capital expenditure by short-term borrowing. Clause 27.1(a) of the Facility Agreement required this to be in place by the end of January 2002. But needless to say, the Baskans took no steps at all to establish such a facility with Yapi Kredi, with which they were by then in nearly terminal dispute, but managed to persuade the Banks throughout January and February that it was the subject matter of active negotiations, and would be put in place by the end of February, if the Banks were prepared (as they were) to provide the necessary extension of time.
717. Baskan Gida obtained its last draw-down under the Facility on the 26^th February, in the sum of €1,992,276, bringing the aggregate drawn down to just under €23 million. This last draw down was based upon Contract 5104, and supported by false copy invoices suggesting that Baskan Gida was to invoice Ferrero Germany and direct payment to the Collection Account. Contract 5104 was one of the three Early February forward contracts, in respect of which the Baskans had never intended that any deliveries would be made by Baskan Gida. On the contrary, they formed the bulk of Aksu Gida's intended order book for the residue of the 2001 season. By the time of the final draw-down on 26^th February, almost a fortnight had passed since the sending of the Transfer Letters, and the Ferrero Operating Defendants had all sent new orders for the entirety of the tonnage the subject matter of the early February forward contracts, to Aksu Gida.
718. In the meantime, Baskan Gida had made its last withdrawal of hazelnuts from pledge in Warehouse 2 on 16^th February. The aggregate tonnage released from Warehouse 2 during the week beginning 11^th February was 1,129mt, and the aggregate released from Warehouse 2 during the whole of the period of its operation purportedly pursuant to the Facility was 8,812mt. In respect of all the tonnage released, Baskan Gida was obliged to and did sign trust receipts promising to use the released hazelnuts for no purpose other than supply to Ferrero. While there are unfathomable difficulties in identifying precisely what the Baskans did with the released hazelnuts, Baskan Gida delivered only 408mt to Ferrero during its final week of deliveries beginning 11^th February. Bearing in mind that those nuts needed processing prior to delivery to Ferrero, it seems very unlikely that anything more than a very small proportion of the nuts released from Warehouse 2 during that week can have been delivered in accordance with the trust receipts, or that the Baskans had any intention that they should be.
719. That brief description illustrates the simultaneous perpetration by the Baskans of every element of their fraud on the Banks. The invoice fraud enabled them to direct payment to Baskan Gida of money which they pretended to the Banks they had directed to be paid by Ferrero to the Collection Account. The trust receipt fraud enabled the Baskans in effect to steal nuts pledged to the Banks by false promises to deliver them to Ferrero. The Baskans were fraudulently trading through Baskan Gida by borrowing money from the Banks which they had no intention that Baskan Gida should ever repay. At the same time the Baskans were putting the finishing touches to the Scheme for the transfer of Baskan Gida's assets, and in particular its stocks of hazelnuts, to the front company Aksu Gida. It will be recalled that the Protocol and Share Transfer Agreements with Indo-Med were not signed, at the earliest, until 19^th February, even though Ferrero had been sent the Transfer Letters six days previously.
720. Every aspect of those frauds was, as I have described, entirely concealed from the Banks. In my judgment, the Baskans also endeavoured to conceal as much of it as they could from Ferrero, Mr Dahod and, but to a lesser extent, from Mr Abidali.
721. For the purpose of the Banks' conspiracy claims, it is necessary to focus precisely upon what Ferrero (through Mr Casale and Mr Rosa Brunet) and Mr Abidali knew, believed or suspected in relation to the Baskans' frauds on the Banks, and when. For that purpose it is convenient to look at the period in late February 2002, just prior to the Banks' discovery of the fraud, and for me to make findings as to what Ferrero and Mr Abidali knew by then, in relation to each of the elements of the fraud which I have described.
722. Taking Mr Abidali first, he vigorously denied throughout the trial that he knew of or was party to any fraud on the part of the Baskans. His case was that through Indo-Med, he and Mr Dahod obtained payment of US$10 million of Baskan Gida's debt under the Dahod Agreement, by means of a lawful scheme, advised and approved by Turkish and English lawyers whereby Indo-Med obtained 80% of the shares of Aksu Gida, and Aksu obtained US$10 million worth of assets, mainly hazelnuts, from Baskan Gida by means of arm's length transactions at proper value, leaving Baskan Gida both able and free to pay its remaining debts to its other creditors out of assets not forming part of that sale.
723. I reject that case. The true position was that Mr Abidali had used Indo-Med and ensnared Mr Dahod in the creation of mainly backdated documents designed to give that impression, whereas the real transaction between them and the Baskans was that they were to be, so far as concerned their apparent ownership of Aksu Gida, nominees for the Baskans, while remaining investors in the Baskans' hazelnut business, conducted through Aksu Gida in the same way, and with the same profit share, as they had enjoyed vis-à-vis Baskan Gida under the Dahod Agreement. That this is so is unmistakably apparent from correspondence, in particular between Mr Abidali and Mr Dahod, and between Mr Abidali and Melih Baskan later in 2002, which I have yet to describe in detail, all of which was belatedly disclosed in these proceedings by Indo-Med's liquidator. The fact that the Baskans and Mr Abidali succeeded in deceiving the Turkish courts into believing that the Scheme had the reality portrayed by the backdated documents in no way dissuades me from that conclusion. It merely demonstrates the dedication with which the Baskans and Mr Abidali participated as conspirators, with partial success, in the fraudulent use of Aksu Gida as a front company.
724. It is I consider equally clear that Mr Abidali well understood that the implementation of the Aksu Gida Scheme would injure all Baskan Gida's creditors who (or which) were owed more than Indo-Med. On his own admission, the Scheme was introduced to Mr Abidali by the Baskans as one designed to achieve payment of Baskan Gida's debts to its smaller creditors, of which Indo-Med was described as the largest. Mr Abidali knew from the beginning of 2002 that Baskan Gida had obtained a large trade finance facility from the Banks and, although there is no evidence that he knew the amount of the indebtedness of Baskan Gida to the Banks thereby created from time to time, I am satisfied that he believed that the Banks were, like Yapi Kredi, among those larger creditors of Baskan Gida who would suffer by the implementation of the Scheme. I reject as a lie Mr Abidali's evidence that he believed that Baskan Gida retained sufficient assets to pay its larger creditors after the asset transfer to Aksu Gida. On the contrary, I am satisfied that Mr Abidali well understood that, save for the land on which its various factories and warehouses were situated (and which he knew was charged to Yapi Kredi or other Turkish banks) Baskan Gida retained no significant assets with which to pay back its borrowings from the Banks.
725. Regardless whether the real transaction between the Baskans, Mr Abidali and Mr Dahod was that pretended by the backdated documents, or that which I have described, its effect was to repay in kind (i.e. in the assets transferred to Aksu Gida) all or nearly all of Baskan Gida's then indebtedness to Indo-Med under the Dahod Agreement, with a corresponding diminution in the assets available to pay Baskan Gida's other creditors, including the Banks. Mr Abidali can have been in no doubt from the way in which he described the transaction as having been introduced to him by the Baskans that Baskan Gida was insolvent, so that the transaction was, on any view, what used in England to be called a fraudulent preference (now a wrongful preference). Whether that preference of itself involved unlawful means is a matter to which I shall return, but Mr Werbicki sensibly accepted that if the sham documents were designed to give a false appearance to the Scheme which (if believed by third parties) would disguise the preference and thereby shelter the assets transferred from recovery in the event of Baskan Gida's liquidation, that would involve unlawful means. The false appearance of creditor pressure created by the combination of the Protocol Agreement, Promissory Notes and the bankruptcy proceedings can have had no purpose other than the sheltering of the assets transferred from recovery under Turkish insolvency law as, in substance although not in form, a preference.
726. Mr Abidali plainly knew of the falsity of the Protocol Agreement and the Promissory Notes, but Mr Werbicki tried hard to persuade me that Mr Abidali had no knowledge of the bankruptcy proceedings, his involvement being limited to giving a litigation power of attorney on behalf of Indo-Med to the Turkish lawyer Mr Zenginpeduk, who thereafter acted purely on the instructions of the Baskans in bringing the bankruptcy proceedings. While I accept that Mr Zenginpeduk probably did take his detailed instructions from Mr Alpay or some other representative of the Baskan family, I reject the substance of that submission. The Second Protocol Agreement, which Mr Abidali signed (after Mr Trotter's anglicisation in September 2002) makes repeated express references to Indo-Med having instituted bankruptcy proceedings against Baskan Gida, and purports to be a settlement of that claim. Mr Abidali therefore knew of every element in, and the falsity of, the appearance of creditor pressure which lay at the heart of the Scheme. I have no reason to suppose that the legal reason for the creation of that false impression was explained to him. He knew only that the falsity involved was part of a structure designed to protect the assets transferred from the claims of Baskan Gida's major creditors.
727. At the other end of the scale, there is no evidence that demonstrates that Mr Abidali knew, believed or even suspected that the Baskans were in early 2002 engaged on the invoice fraud against the Banks. The evidence does not show that Mr Abidali knew anything about the detail of the Facility Agreement, or the mechanics of its operation, and I can think of no reason why the Baskans should have explained to him the way in which, by the use of fictitious copy invoices, the Baskans diverted to Baskan Gida an income stream from Ferrero which should have been paid to the Banks.
728. The position is more difficult in relation to the trust receipt fraud and what I have labelled Baskan Gida's fraudulent trading. Mr Abidali knew of course from early January that an essential part of the Scheme proposed to him was that Baskan Gida would transfer a large and valuable stock of hazelnuts to Aksu Gida, as the seed corn (as Mr Wardell put it) or more conventionally the working capital for its ongoing business, a large part of which was to consist of trading with Ferrero. Aksu Gida had no financial resources of which Mr Abidali was aware with which to purchase hazelnuts, and the evidence discloses no discussion between him, the Baskans and Mr Dahod about the obtaining of trade finance or other borrowings. On the contrary, the whole of the Dahods' contribution to Aksu Gida was to take the form of hazelnuts and a small amount of other assets.
729. Mr Abidali's case was that, for all he knew, the hazelnuts transferred had originally been acquired by Baskan Gida with the use of the Dahods' investment, which had reached some US$13 million by November 2001, and remained in continuing use for the purposes of Baskan Gida's business with Ferrero thereafter, in accordance with the Dahod Agreement. He said that he had no knowledge of the provisions in the Banks' Facility relating to Warehouse 2, the lien or the trust receipt in relation to the hazelnuts to be purchased with the Banks' money, and learned nothing to put him even on suspicion that the Baskans were planning dishonestly to use the Banks' money and the Banks' hazelnuts as the means of providing Aksu Gida with the wherewithal to start, and continue, its business. Mr Abidali also maintained a vigorous case, albeit without recourse to forensic expert witnesses, which he said were beyond his financial resources, to the effect that the nuts actually transferred by Baskan Gida to Aksu Gida were not of a quality sufficient for use in the Ferrero business, and had not in fact come from Warehouse 2.
730. In my judgment Mr Abidali did believe that, by procuring that Baskan Gida continued to trade from mid January until late February 2002 while putting in place the Scheme to transfer its business and most of its assets to Aksu Gida, the Baskans were defrauding Baskan Gida's major creditors, by continuing to incur debts which they had no intention that it should repay in full. My reasons are as follows. First, Mr Abidali knew by the beginning of January that Baskan Gida had obtained a major trade finance facility from the Banks. Secondly, although I accept that he may not have known whether the Facility was to be used to fund Ferrero business, or alternatively Baskan Gida's other business, and may have been unaware of the amounts and dates of relevant draw-downs, he knew that Baskan Gida was to continue to trade actively until the moment of transfer of the hazelnut business to Aksu Gida. He knew more than enough about the realities of trade finance to appreciate that such trading involved the making of periodic draw-downs of working capital from any banks providing trade finance.
731. Thirdly, Mr Abidali knew (as is apparent from his 8^th February email to Mr Dahod) that the transfer to Aksu Gida was being planned and, to the extent possible, implemented in secret, at least from Yapi Kredi, and, probably, from all Baskan Gida's creditors other than Indo-Med. The reference in his 8^th February email to Mr Dahod to the need to avoid the risk that:

"wind of their ideas may be getting out"

makes it clear that Mr Abidali well understood that the success of the Scheme depended in part upon those creditors who were to be left behind having no advance knowledge of the Scheme before the transfer had taken place, and therefore, no opportunity to attempt to prevent it.

732. Fourthly, Mr Abidali believed, despite his denial, that the effect of the Scheme would be to leave Baskan Gida's major creditors, including the Banks unpaid, at least in substantial part. Fifthly, and by contrast with the way in which the Scheme was presented to Ferrero on 25^th January, nothing in the manner in its presentation to Mr Abidali involved the suggestion that it was merely a contingency plan, against the possibility that negotiations with Yapi Kredi might not work out, nor was the debt being claimed by Yapi Kredi explained to Mr Abidali, according to his evidence, as one owed by the family rather than by Baskan Gida. It follows that the presentation of the scheme to Mr Abidali necessarily involved a mutual understanding between him and the Baskans that Baskan Gida was by early January 2002 already insolvent.
733. Mr Abidali's evidence was that, following discussion with Mr Dahod, he approved the scheme in principle by 19^th January, even though the documents took until much later to complete and backdate. It follows that, far from there having been any perception of light at the end of the tunnel, following which Baskan Gida might have paid its debts to its main creditors in full, Mr Abidali knew that the Baskans were, from mid-January at the latest, continuing to incur credit for the ongoing business of Baskan Gida with no honest intention that the creditors with whom such credit was being incurred would be paid in full, and this must have remained his understanding until the completion of the Scheme actually led to Baskan Gida ceasing to trade, in late February.
734. There remains the question whether Mr Abidali suspected, believed or knew that, during the same period, the Baskans were arranging for the assembly of the stock of hazelnuts to be transferred to Aksu Gida by the misuse either of nuts belonging to the Banks, or of finance fraudulently obtained from the Banks. No case that he knew that the nuts transferred actually belonged beneficially to the Banks was even put squarely to Mr Abidali in cross- examination, although it was put that he helped the Baskans with the timing of the draw-downs and transfer of nuts to Aksu Gida. That was in my judgment a sufficient putting of a case that the nuts transferred had been acquired by Baskan Gida with the use of the Banks' money.
735. The strongest argument in favour of Mr Abidali was that, having with Mr Dahod contributed in excess of US$13 million to enable Baskan Gida to buy nuts for its Ferrero business, he may reasonably have assumed that the nuts being transferred to Aksu Gida had been purchased with the use of Mr Dahod's revolving loan.
736. There are however a number of difficulties with that argument. The first is that, as Mr Abidali admitted in cross-examination that he knew at the time, the Dahod Investment monies were paid by Indo-Med into Baskan Gida's general account, rather than being segregated for Ferrero business. I was not persuaded by Mr Abidali's assertion in cross-examination that he nonetheless trusted and believed that the Baskans were restricting the use of the money for that purpose.
737. Secondly, Mr Abidali stoutly maintained, albeit with the benefit of hindsight rather than on the basis of information known to him at the time, that the nuts transferred to Aksu Gida were not suitable, in terms of quality and specification, for Ferrero business, basing himself on the inspection and testing certificates signed by the Giresun quality control staff at or about the time of the transfer. I consider that in fact a substantial part of the nuts transferred to Aksu Gida were then processed and sold to Ferrero, but it hardly assists Mr Abidali's case that he assumed that the nuts transferred had been purchased with Mr Dahod's money for him to assert that they were unsuitable for Ferrero business.
738. In my judgment Mr Abidali must have believed that a significant part of the nuts transferred to Aksu Gida had been purchased with the Banks' money, although he neither inquired nor knew precisely what proportion of the nuts were so attributable. My reasons are as follows. First, during the latter part of 2001, Mr Abidali had been told by the Baskans that the whole of the Dahod Investment was urgently needed to enable Baskan Gida to satisfy Ferrero's requirements, therefore by implication leaving nothing spare for the purpose of building up any stockpile. Secondly, Mr Abidali must have appreciated that a company such as Baskan Gida with serious working capital difficulties would be unlikely to have the capital resources with which to build up such a stockpile, beyond the immediate requirements needed for processing nuts to meet Ferrero orders imminently due for delivery. In short, Baskan Gida would be operating hand to mouth, rather than on the basis of comfortable hazelnut reserves. In this respect I reject Mr Abidali's evidence that he thought that Baskan Gida maintained large hazelnut stocks at all times. Thirdly, Mr Abidali knew, from the beginning of 2002, that Baskan Gida's working capital requirements were being met from a mixture of the Dahod Investment and the new facility granted by the Banks. If he made no inquiries as to how, in detail, those alternative sources of working capital were being applied the inevitable inference is that he must have believed that the stockpile transferred to Aksu Gida by the end of February had been acquired from a mixture of those two sources. In that respect, I think it most unlikely that the Baskans would have had any wish to provide him with any more specific explanation, and that he would have had no wish to investigate whether his belief to that effect was true, still less as to the precise attribution of the monies used to build that stockpile.
739. I turn to consider the extent of Ferrero's knowledge immediately prior to the discovery by the Banks of the Baskans' fraud. I have dealt with this in sufficient detail on an event by event basis to make it unnecessary to provide more than a summary at this stage.
740. Notwithstanding receipt of the split invoices, I do not consider that anyone at Ferrero knew or even suspected the invoice fraud. Apart from the split invoices, nothing in the pattern of dealings between Baskan Gida and Ferrero following the grant of the Facility was sufficient to put Ferrero on notice, let alone knowledge, of that fraud. For the reasons which I have given, the split invoices were not either. True it is that Ferrero received numerous invoices from Baskan Gida after the grant of the Facility requesting payments to be made otherwise than to the Collection Account, in circumstances where a minute study of the terms of the Notices and Acknowledgements of Assignment might have suggested that payment should have been made to the Banks. There was however no evidence that anyone at Ferrero knew precisely which Baskan Gida contracts had been submitted to the Banks for the purposes of obtaining finance, or knew anything of the detailed terms of the Facility Agreement and associated documentation other than the assignment documents which I have described in detail.
741. Whether rightly or wrongly as a matter of law, (which is for the Italian courts to decide), I am satisfied that Mr Casale assumed that in respect of all invoices to Ferrero which Baskan Gida was required under its agreement with the Banks to direct payment to the Collection Account, a statement to that effect would appear on each invoice, as prescribed by the Acknowledgements of Assignment, and I am satisfied that at no time did Mr Casale or anyone else at Ferrero suspect prior to the events at the end of February which I am about to describe, that Baskan Gida had for dishonest purposes sent Ferrero invoices directing payment to the wrong account, as part of a fraud on the Banks.
742. I am equally satisfied that no-one at Ferrero knew or suspected before the end of February 2002 that Baskan Gida was perpetrating the trust receipt fraud, that is making dishonest use of hazelnuts belonging to the Banks by transferring them to Aksu Gida, rather than processing and selling them to Ferrero. There is no evidence that Ferrero knew anything about the security provisions in the Facility Agreement. All they had was an un-formed understanding that some hazelnuts at the Giresun factory were held on security terms that required third party consent to their release. Without knowledge of the terms of the lien, or of the trust receipt, it is difficult to see how Ferrero could have known or suspected a breach of those terms. Furthermore, the evidence does not suggest that, by contrast with what Mr Abidali was told, Ferrero was given any explanation by the Baskans as to how Aksu Gida was going to obtain the hazelnuts which it needed to continue supplies to Ferrero or, more generally, how Aksu Gida's working capital requirements were going to be met.
743. As for Baskan Gida's fraudulent trading, I have concluded that Ferrero in the persons of Mr Casale and Mr Rosa Brunet probably did come to suspect, during the period between 25^th January and the end of February, that the Banks had not been informed of the Baskans' plan to transfer the ongoing hazelnut business to Aksu Gida, and their knowledge of the likelihood that Baskan Gida/Ferrero contracts would be used for the purpose of raising finance with the Banks must also have given them reason to suspect that the implementation of the Scheme would lead to the Banks not being repaid in full in relation to borrowings obtained during the period when the Scheme was being planned and, later, implemented. But I do not consider that, at any time before the end of February 2002, the suspicion had matured in the minds of anyone at Ferrero into a belief that fraudulent trading of that kind was probably taking place, or that, by making no inquiries either of Baskan Gida or of the Banks in that regard, Mr Casale or Mr Rosa Brunet were deliberately avoiding learning for sure that which they believed was probable. In summary, there was nothing sufficient to demonstrate to Ferrero that Baskan Gida was borrowing against contracts which it did not intend to perform, in circumstances where performance by Aksu Gida would (rather than merely might) lead to the Banks not being paid.
744. I have reached this conclusion on balance by considering, both separately and collectively, the events likely to have given rise to that suspicion, namely the 25^th January meeting, the December and January countersigned contracts, the Early February Contracts, the receipt of the Transfer Notices (coupled with the Framework Contract amendment) and finally the receipt of the split invoices. I have reviewed each of those events separately, and their effect upon Mr Casale and Mr Rosa Brunet's knowledge and suspicions, above. Taken collectively, I consider it likely that their effect would have been to trigger a suspicion that the Baskans might not have been dealing openly with the Banks in relation to the proposed transfer to Aksu Gida, but not to a level which can be properly be described as a belief that some form of fraudulent trading was likely or probable. To the specific reasons for that conclusion already given, I would add the following general observations.
745. First, perhaps the most important events which should have focused minds at Ferrero upon the honesty of the Baskans' dealing with the Banks were the making of the late January and early February contracts. But the evidence is that the late January contracts were substantially performed by Baskan Gida, and that the early February contracts were only to a modest extent used for the raising of further finance. Secondly, both those sets of contracts were made (or remade) at a time when the Scheme was described to Ferrero as merely contingent, the impression being given that the prospect of a satisfactory outcome of the negotiations with Yapi Kredi was still on the cards, such that there was light at the end of the tunnel for Baskan Gida itself. Thirdly, the evidence does not suggest that there was any way in which Ferrero could know the extent to which and the terms upon which the Baskans might use those contracts for the raising of further finance with the Banks and, in the absence of any knowledge of the terms of the Facility, Ferrero had no way of knowing without detailed inquiry whether in practice the Baskans could obtain funding from the Banks on terms which were otherwise than fully secured, such that a transfer of the business to Aksu Gida would have necessitated a negotiation between the Baskans and the Banks sufficient, for example, to transfer their security and rights to repayment to the assets and business of the new company. My finding that the Baskans were doing no such thing derives from an understanding of the invoicing fraud, the trust receipt fraud and the detailed terms of the proposed Scheme, of all of which Ferrero were ignorant, and about none of which (being neither a shareholder, investor in or creditor of Baskan Gida) Ferrero had any duty or commercial incentive to inquire.
746. Finally, in relation to the transfer scheme itself, Ferrero knew nothing about the detailed terms of the sham documents, or about the joint venture involving Indo-Med and the Dahod Investors, or their participation in the Scheme. All Ferrero knew was that the Scheme, as explained to them on 25^th January, involved an element of subterfuge and deception of Yapi Kredi, since the new company was, at one and the same time, to appear to be independent from the Baskans while in reality continuing to be controlled by them. Again, I consider that Mr Casale and Mr Rosa Brunet probably suspected that the Baskans may have been keeping the Banks similarly in the dark about the difference between the outward appearance and the inward reality, but not to have reached a stage where they knew for sure or believed that this was so, or that the Baskans' reasons (whatever they were) for keeping the reality from Yapi Kredi also necessarily applied to the Banks.
747. Finally, although Ferrero had a substantial commercial motive for wishing to cooperate with any plan by the Baskans which would preserve their business and in particular their supply chain from potential destruction at the hands of Yapi Kredi, I am not persuaded that Ferrero as an institution, still less Mr Casale and Mr Rosa Brunet as individuals, had any motive for wishing to associate Ferrero with a scheme which they knew or believed would involve a fraud on the Banks. In my judgment Ferrero's commercial objective to preserve its supply chain and to avoid becoming beholden to a single supplier was insufficient to justify participation in what its senior management believed was a fraud, and no evidence of any kind has emerged to suggest that either Mr Casale or Mr Rosa Brunet had any personal corrupt or other motive consciously to expose Ferrero to the risks attendant upon participation in a fraud. On the contrary, as I have already stated, my impression of both of them, supported by all the evidence which I have seen, suggests that their loyalty to Ferrero and their determination to further its best interests was beyond question.
748. In reaching those conclusions about what Ferrero through Mr Casale and Mr Rosa Brunet did or did not suspect, believe or know at the material time, I have taken full account of the fact that both of them have been untruthful in respect of significant parts of their evidence, and in some instances, deliberately so. I have also considered whether aspects of Ferrero's conduct after the end of February 2002, during the period following the Banks' discovery of the fraud, justifies an inference that Ferrero knew about it, or about parts of it, before the end of February. Finally, I have taken full account of the Banks' submissions that in relation to important factual issues, Ferrero had decided not to call certain available witnesses. I have concluded that none of those matters, taken separately or collectively with all the other evidence, justifies a conclusion that Ferrero's knowledge or suspicion of the Baskans' fraud was greater than that which I have thus far described. I shall set out my reasons for that conclusion at a later stage in this judgment, after reviewing the whole of the facts chronologically.

The Aftermath

The Banks Discover the Baskans' Fraud

749. The second draw down by Baskan Gida was due for repayment to the Banks on Friday 22^nd February and, in the light of Baskan Gida's delay in repaying the first draw-down, the Banks assumed that nothing would be received before the beginning of the following week, (i.e. 25^th February) by when the third draw-down would also have become payable. Nothing was received on 25^th or 26^th February, and promissory notes given by the Baskan family were due to expire on the 27^th and 28^th February in respect of those loans. If they were to be relied upon, they would have to be protested within two days after expiry. The Banks therefore sought from the Baskans an explanation for non-payment and an extension of the promissory notes. Since the Banks expected payment to be received from Ferrero (in accordance with false copy invoices received from the Baskans) Mr Byles and Mr van Broekhoven wrote by email to Mr Casale on 27^th February referring to the Facility and identifying by invoice numbers two sets of invoices for €1,641,792 and €1,380,456 due for payment on 22^nd and 25^th February by Ferrero Italy, continuing:

"As you will be aware, Baskan's rights from Ferrero have been assigned to the above banks. Please therefore arrange for immediate payment or advise us of any reasons for delaying the payment."

750. No satisfactory explanations were received for the delayed payment either from Ferrero, the Baskans or from PRI. All the Banks got was an extension of the promissory notes coupled with an explanation that the Baskans were due to have a family meeting to sort matters out at the weekend. Accordingly, on Friday 1^st March the Banks called an event of default. Later on the same day they learned for the first time, via SGS, of the "hiring" of the Baskan Gida business to Aksu Gida and, on further inquiry, that Yapi Kredi had sought to attach hazelnuts at the Giresun factory, and was alleging that Baskan Gida had been made fraudulently bankrupt and that Yapi Kredi would be suing Baskan Gida for "swindling". By the end of Friday the Banks learned (as was the fact) that a judge and appointed experts from the local Turkish court had attended at the Giresun premises together with police, and that Latif Aksu was claiming that the business had been leased to Aksu Gida a month previously. The Banks' immediate concern was to ensure that hazelnuts still in pledge at Warehouse 2 were protected from Yapi Kredi's attachment. In that respect the Banks succeeded, but only in relation to the modest quantity of poor quality hazelnuts still located in Warehouse 2 by that date, which the Banks eventually sold to Indo-Med for US$720,000.

March 2002 – Mr Byles speaks to Mr Casale

751. The Banks received no immediate response from Ferrero to Mr Byles' and Mr van Broekhoven's email of 27^th February. On that day Mr Casale had been visiting Ferrero Germany, and he did not return to his office in Alba until the week beginning Monday 4^th March. After some inconclusive brief conversations between Mr Byles and Mr Scavino, the first meaningful contact between the Banks and Ferrero consisted of a short telephone conversation initiated by Mr Byles directly with Mr Casale on either 1^st or 4^th March. It was probably on 4^th March. By this time Mr Casale had still not returned to the office and read the Banks' email, but had been informed by Mr Scavino or another of his colleagues (as was the case) that Ferrero Italy had paid all Baskan Gida's invoices in full.
752. There is no dispute about what occurred during the telephone conversation, since Mr Byles broadly accepted, in cross-examination, Mr Casale's written account of it in his witness statement. It was a short conversation. Mr Byles told Mr Casale that there were outstanding payments due to the Banks. Mr Casale responded by saying that he had checked with Ferrero Italy and that none of the invoices from Baskan Gida were outstanding. Mr Byles did not ask into what bank account payment had been made, but told Mr Casale that Baskan Gida was no longer operating. Mr Casale's reply was that he knew that, and he volunteered that Ferrero was in contact with Aksu Gida, which was supplying Ferrero with hazelnuts. Mr Byles did not inform Mr Casale of any security interest of the Banks in hazelnuts stored at the Giresun factory, nor of the rumours and suspicions then reaching the Banks that the Baskans had been involved in a fraud.
753. Each side blames the other for having been uncommunicative during that brief conversation. For the Banks, it was put to Mr Casale that a reasonably careful study of the Banks' 27^th February email would have revealed that the Baskans had been practising an invoicing fraud, since no invoices corresponding to the numbers specified in the email had actually been received by Ferrero Italy, let alone with a direction for payment to the Banks. Mr Casale acknowledged this in cross-examination, but said that he had not at the time of his telephone conversation on 4^th March yet read the email. For their part, the Ferrero Defendants criticised Mr Byles for not notifying Mr Casale of the Banks' continuing security interest in nuts stored at Giresun. Mr Byles sought to explain that by saying that, in his view, the Banks' nuts were either still secure in Warehouse 2, or subject to Yapi Kredi's attachment, so that there was no likelihood of further sales by Aksu Gida of the Banks' nuts to Ferrero, against which any warning needed to be given. For his part, Mr Byles criticised Mr Casale's tone as not being forthcoming, and as sounding rather annoyed at being telephoned. Mr Casale's response was that the call had reached him on his mobile telephone while taking his son to the local swimming pool. Mr Byles acknowledged that, by the time of his telephone conversation with Mr Casale, the Banks suspected the Baskans of a fraud, but chose not to reveal their suspicions at that stage to Ferrero.

Investigations by the Banks

754. In the meantime the Banks sent Mr van Broekhoven and Mr Lewis to Istanbul, where they met Mr Politis and Mr Sason of PRI on 5th March, who had in response to earlier inquiries from the Banks been seeking to find out what was going on. PRI had by this stage abandoned its former loyalty to the Baskans as clients (perhaps from a belated recognition of the need to preserve their reputation with the banking community) and lost no time in informing the Banks that, in their view, Aksu Gida was a front or "shell" company still being run by the Baskans notwithstanding their protestations to the contrary, and that Ferrero was cooperating with the Baskans' scheme to protect their assets from angry creditors.
755. Mr van Broekhoven and Mr Byles had a face to face meeting with Ahmet, Resat and Ismet Baskan on 7^th and 8^th March, from which (according to their notes made at the time) they derived the following information. First, the Baskans told them about the US$13 million loan obtained from the Dahod Investors, describing them (without naming them) as a religious group based in the USA. The US$13 million loan was described (in accordance with the sham documents, but contrary to the truth) as having been repayable in December 2001. Secondly, the Aksu Gida scheme was described as a "rental scheme" frequently used in Turkey, carried out in close cooperation with the religious group and brokered by what they described as the Baskans' London agent (a reference to Mr Abidali although he was not named either). The Baskans denied being in control of Aksu Gida, and said that the religious group were in control.
756. Thirdly, and most importantly, the Baskans admitted that hazelnuts released from pledge in Warehouse 2 had been transferred to Aksu Gida, under what the Baskans described as "extreme pressure from relationship creditors" (by which the Baskans meant the religious group), and that at least some of those nuts had been on-sold to Ferrero, for direct payment by Ferrero to Aksu Gida. Thus the Banks discovered from the Baskans' frank admission that nuts which had been financed under the Facility were being on-sold to Ferrero via Aksu Gida by means which wholly circumvented both the Banks' security and structured mechanism for repayment by the assignment of Baskan Gida's receivables from Ferrero. The Banks drew the inference that Ferrero was "part of the scheme", and that, (correctly in my judgment) Ferrero would not have participated in the Scheme by buying large quantities of nuts from Aksu Gida if it did not itself believe that the Baskans remained in effective control of that company.
757. During the same visit to Istanbul the Banks received advice from their Turkish legal advisers Pekin & Pekin (for which privilege was not claimed) to the effect that the Banks' pledge over nuts in Warehouse 2 would under Turkish law be treated as released once those nuts left the warehouse. After taking Turkish legal advice, the Banks appear to have formed a clear view that pursuing bankruptcy proceedings against Baskan Gida would produce a minimal return, not least because its factory premises appeared to have been mortgaged to Yapi Kredi.
758. Thus by the end of the first week in March the Banks had been told by the Baskans themselves that nuts pledged to them had been dishonestly misappropriated and transferred to Aksu Gida, by Ferrero and the Baskans that Ferrero was buying nuts from Aksu Gida in place of Baskan Gida, and by PRI that, notwithstanding the Baskans' assertion that the "religious group" (i.e. the Dahod Investors) were in charge of Aksu Gida, there was reason to think that in reality Aksu Gida was a front for the Baskans. Mr Byles also acknowledged in cross-examination that he had already concluded by that stage, in part from the fact that Ferrero claimed to have paid all invoices from Baskan Gida, that some form of invoicing fraud had been perpetrated by the Baskans.
759. The Banks' principal witnesses (Mr Lewis and Mr Byles) were taken to task in cross-examination with the undoubted fact that, in the context of the Banks' pursuit of a proprietary claim in relation to hazelnuts sold by Aksu Gida to Ferrero, the Banks shared none of the information which I have described with Ferrero. The Banks did not communicate further at all with Ferrero after Mr Byles' short conversation with Mr Casale on 4^th March, until 26^th April, and at no time informed Ferrero of their suspicions that there had been an invoicing fraud, and that in buying nuts from Aksu Gida, Ferrero risked obtaining property which had been dishonestly misappropriated from the Banks' security. The Banks did not inform Ferrero even of the existence of the security, let alone of the fraudulent abuse of it which the Baskans had actually admitted.
760. The Banks' witnesses' response was (from Mr Lewis) that to inform Ferrero of suspected fraud which the Banks could not then prove, could have serious consequences, and (from Mr Byles) that his request to senior management for an open face to face discussion with senior management from Ferrero was turned down by his Japanese superiors at BTM due to the perception that it would involve an unacceptable loss of face, in inquiring about matters of which BTM would appear to be ill informed. Both witnesses suggested that Ferrero itself was, in any event, being uncooperative by that stage, and both regarded it as unlikely that any hazelnuts originally released from Warehouse 2 were by early March still in the hands of Aksu Gida awaiting delivery to Ferrero, so that there was no practicable prospect of protecting the Banks' proprietary rights by giving Ferrero notice of them.

Ferrero's reaction

761. Looking at the matter from Ferrero's perspective, the Banks' 27^th February email, suggesting that more than €3 million were outstanding on invoices from Baskan Gida to Ferrero which had been assigned to the Banks, must have come as a considerable shock if (as I conclude was the case) Ferrero was until then unaware that the Baskans had been perpetrating an invoice fraud.
762. As Mr Casale acknowledged in cross-examination, a relatively painless and swift check of Ferrero's internal records would quickly have revealed that the invoices specifically numbered and thereby identified as outstanding in the Banks' email did not match invoices actually received by Ferrero from Baskan Gida. It is worth quoting the exchange (at day 37 page 30):

"Q: … anyone receiving this [email] who checked the system would immediately know that a fraud had been perpetrated by the Baskans.

A: Yes. Or that there was false document."

In re-examination Mr Casale explained that by "false" he meant "wrong", meaning that checking the Banks' email against Ferrero's records would have revealed that there had been a serious mistake, but not necessarily a fraud (day 38 page 145).

763. Mr Casale's evidence (both in chief and in cross-examination) was that he had not instructed his subordinates specifically to check the invoice numbers in the Banks' email with Ferrero's records, but simply to check whether all Baskan Gida's invoices to Ferrero Italy had been paid, which revealed the (correct) response that they had. He said that after his discussion with Mr Byles he had telephoned Melih Baskan, to be told that there had been a big misunderstanding which the Baskans would sort out, and which was not Ferrero's problem. He said that shortly thereafter Melih Baskan told him that the problem had been sorted out and that, in the absence of any further inquiry from the Banks, no further investigation was at that stage carried out internally within Ferrero.
764. In his evidence, Mr Rosa Brunet said that he could not recall Ferrero's receipt of the Banks' 27^th February email, nor how it was dealt with. In his witness statement, Mr Do said that the matter was not reported to him at all at the time. In oral evidence in chief he corrected that by saying that Mr Casale had told him about his conversation with Mr Byles and Melih Baskan, at a meeting with Mr Rosa Brunet in March. He said that the impression that he had been given by Mr Casale was that all invoices had been paid, and that Melih Baskan had told him that the misunderstanding which had led to the Banks' email was not a problem for Ferrero.
765. Ferrero's case, and its witnesses' evidence, was that no suspicion of a fraud by the Baskans arose within Ferrero until late July 2002. Until then, Ferrero thought that the Banks' claims to have been unpaid in large amounts by Ferrero were as likely to have been the result of a mistake within the Banks, as within the Baskans' enterprise, and merely showed that something had gone wrong with the working of the Facility.
766. By contrast, the Banks' case is (of course) that Mr Casale and Mr Rosa Brunet knew rather than merely suspected that there had been a fraud, had participated in it throughout, and that what they categorise as Ferrero's obstructive and uncommunicative response to the Banks' inquiries is evidence justifying an inference that Mr Casale and Mr Rosa Brunet, who were in charge of Ferrero's response, were guiltily keeping their heads down, and doing their best to prevent the Banks discovering the truth.
767. The parties directed considerable forensic effort to that issue, such that the history of the communications (or lack of them) between the Banks and Ferrero following the Banks' discovery of the fraud and throughout the rest of 2002 was examined in minute detail, even though those facts formed no part of the essential structure either of the Banks' claim or of Ferrero's defence. It is convenient first to provide a factual account of those exchanges (and of such internal goings-on in Ferrero as have been revealed by the evidence) before expressing my conclusions.

Contacts between Ferrero and the Baskans

768. The other main issue arising from the detailed examination of events from March onwards during 2002 is the question whether in fact the Baskans were controlling Aksu Gida and its successor Baskan Yuksel and, more contentiously, whether Ferrero and Mr Abidali knew that. In that regard, I have already expressed my conclusions that the Baskans were in control of both companies throughout, and that both Ferrero and Mr Abidali knew that, but my description of events from March onwards in 2002 provides much (although not all) of the material which has led me to those conclusions.
769. Although Mr Casale had no direct contact with the Banks in March or April 2002, he did have further contact with Melih Baskan. On 25^th March Melih Baskan emailed Mr Casale suggesting that, for the future, they communicate from computer to computer using hotmail addresses and the MSN messenger service. On the following day Mr Casale emailed asking Melih Baskan to ring him and, presumably in the absence of a response, then sent him the following message:

"Good morning

I just learnt that no merchandise came to Cumayeri therefore there will be no production for us today and the people there does not know the programmes for the nexts days.

Please call me back.

Thank you

Casale"

Cumayeri was one of the former Baskan Gida plants by then being operated by Aksu Gida. The email clearly suggests that, having failed by contact with local management at Cumayeri to establish an agreed short-term production and delivery programme, Mr Casale was looking to Melih Baskan to sort the problem out. The natural inference is that Mr Casale regarded Melih Baskan (or other members of the Baskan family for whom Melih Baskan was a convenient point of contact) as remaining in ultimate managerial control of Aksu Gida's operations.

770. Mr Strauss tried long and hard to persuade me to the opposite view. He submitted (correctly) that the surviving documents showed regular contact between Ferrero operating company employees and Mr Franko, and that Mr Franko was not (as he had been previously) simply based in Istanbul, but attended the Giresun plant. He relied upon Mrs Pay's evidence that, from March 2002 onwards, the Baskans absented themselves from the Giresun plant and continued a supervisory role, invisibly as it were, over the telephone. Above all, he submitted that, leaving aside Mr Bolowich's initial communications with Melih Baskan, the exchange between Mr Casale and Melih Baskan which I have just described stands as an isolated incident among the surviving documents, suggesting that, in sharp contrast to the much more frequent surviving exchanges between Mr Abidali and Melih Baskan, the Baskans took the trouble to ensure that their continued supervision of Aksu Gida was as invisible to Ferrero as it was designed to be to the rest of the outside world. He concluded that, since all Ferrero's orders for performance by Aksu Gida for the rest of the 2001 season had been placed with Baskan Gida by the time of the Aksu take over, that there was in any case little need for contact between Mr Casale and Melih Baskan about operational matters affecting Aksu Gida thereafter.
771. I accept that the arrangements made for the day to day administration of the Ferrero/Aksu Gida trading relationship from March 2002 onwards was set up in such a way as to minimise, as far as possible, any contact, visible to the outside world, between Ferrero and the Baskan family, so that for day to day matters, Mr Franko became Ferrero's routine point of contact at Aksu Gida. That much was consistent with Melih Baskan's explanation at the 25^th January meeting that the Baskan family would have to appear to disengage from the hazelnut business, without doing so in fact.
772. Otherwise however, I reject Mr Strauss's submissions on this point, for the following reasons. First, as already explained, I consider it inconceivable that Ferrero would have agreed to transfer in bulk its existing order book from Baskan Gida to Aksu Gida if it had not been assured that the Baskans were to remain in overall control. Secondly, there is nothing in Mr Strauss's point that the email exchange in March which I have described was an isolated occurrence. Ferrero routinely destroyed its operational emails after a relatively short period, such that the Ferrero Defendants (and Ferrero Industrial in particular) have not in fact disclosed what must have been a substantial email traffic with Aksu Gida for the relevant period. The exchange in March 2002 between Mr Casale and Melih Baskan was disclosed as part of the contents of Mr Casale's personal lap top computer. Furthermore, the content of the March 2002 emails which I have described demonstrates that Mr Casale sought to communicate with Melih Baskan primarily by telephone, and only resorted to a short email when Melih Baskan did not respond to an invitation to call him back. Telephone conversations have left no records capable of forensic analysis. Nor was it an isolated survivor in fact. There was another contact between Mr Casale and Melih Baskan about the affairs of Aksu Gida on or about 1^st May 2002, to which I shall shortly refer, of which a surviving fax provides evidence. Finally, even if it had been an isolated survivor, Mr Casale's 26^th March email tellingly reveals an inbuilt assumption that Melih Baskan was the person to speak to if routine communications at a more junior managerial level failed to produce the desired result. It is that underlying assumption which gives the email such weight as probative of Ferrero's knowledge that the Baskan family were to remain in control of Aksu Gida, notwithstanding their protestations to the contrary to the outside world, as reflected for example in what they told the Banks' representatives at their meeting earlier in March.
773. The 1^st May communication was a telephone call, referred to in an Aksu Gida fax to Mr Casale of that date, during which it was agreed to shift 240mt of the hazelnuts outstanding under what had been Contract 5104 from Aksu Gida to Azer Gida, a Baskan company in Azerbaijan which had no been caught up in the re-organisation in Turkey, and which remained therefore clearly under Baskan family control. Understandably Mr Casale could not remember this routine conversation when cross-examined about it. For this novation to have resulted from a single telephone conversation suggests that a Baskan family member must have been on the other end of the line. I find that it was probably Melih Baskan, speaking at the same time for both Aksu Gida and Azer Gida.

The Freezing injunction on Aksu Gida

774. On 15^th April Yapi Kredi obtained a form of freezing injunction from the Turkish courts, in connection with its claim to set aside the asset transfer to Aksu Gida as a fraud on Baskan Gida's creditors. Until then, trading between Aksu Gida and the Ferrero Defendants had been proceeding smoothly, but the injunction had the effect of freezing the bank account into which the Ferrero Defendants had been invited to make payments for deliveries received. This led to Aksu Gida making a series of requests to the Ferrero operating companies that, in future, payments be made not to Aksu Gida's bank, but to a private individual named Onal Ozel at a different Turkish bank. The request began with a telephone inquiry by Mr Franko of Mr Bolowich whether Ferrero Germany could stop a payment then being made, which Mr Bolowich told him was no longer possible, and a request to make no further payments to Aksu's bank due to what Mr Franko described as "problems with his bank". This was followed by an email on 25^th April requiring payment of all future invoices to be made to Mr Ozel, with which at least some of the Ferrero Operating Defendants complied, until it was reported to Mr Do, who ordered it to be stopped immediately, on the basis that payment of debts due to companies ought not to be made to private individuals. Mr Casale's evidence, which I accept, was that Aksu Gida had explained that the reason for the request was to prevent Yapi Kredi interfering with its business, but that Ferrero were not told of the injunction.

Late April – The Banks enquire further of Ferrero

775. Aksu Gida's attempts to evade the consequences of the Turkish court injunction occurred at the same time as the Banks' second request to Ferrero to explain what had become of the payments due in respect of assigned invoices. This took the form of faxes in substantially identical form to each of the Ferrero Operating Defendants, referring to, and identifying by way of contract number, purchase order number, invoice number and amounts due the precise amounts owing to the Banks under Baskan Gida's assignment of the proceeds of sale of hazelnuts, seeking immediate payment, or, in the alternative, an explanation of the reason for non-payment, and looking for a prompt response. The aggregate amounts claimed exceeded €8.213 million from Ferrero Italy, €5.443 million from Ferrero France and €10.293 million from Ferrero Germany, an aggregate claimed from Ferrero as a whole in excess of €23.949 million.
776. Each of the three Ferrero Operating Defendants reacted to the Banks' faxed demands in a different way. At Ferrero Germany Mr Schulz sought immediate clarification by telephone from KBC, and received a spreadsheet giving full details of the amounts claimed, on the following day. After a further telephone call with Mr Lewis, Mr Schulz replied by fax on 29^th April, to the effect that none of the numerous invoices in respect of which the Banks were claiming payment had been registered in Ferrero Germany's accounts, and that deliveries by Baskan Gida had ceased in February/March 2002.
777. In response, KBC asked by fax sent on 2^nd May whether Baskan Gida had made any deliveries under five numbered purchase orders (in fact the numbers referred to contracts rather than purchase orders). To this inquiry, despite a further chasing fax and several telephone calls, KBC received no reply. It is evident from an internal fax between Mr Bolowich and Mr Casale on 10^th May that Mr Schulz had felt obliged to reply, but had been told not to by Ferrero Industrial. In cross-examination Mr Casale said that Mr Do had been shown, after the event, a copy of Mr Schulz's reply to KBC on 29^th April and had disapproved the inclusion in that reply of what he considered to be confidential information about Ferrero's turnover with Baskan Gida, the inference being that the instruction not to reply any further came from him. Mr Do's evidence was that he only saw Mr Schulz's reply at a meeting in late May, but I consider that unlikely.
778. It is evident from notes made in manuscript by Mr Bolowich on the relevant correspondence that he at least at Ferrero Germany took a careful look at the material supplied by KBC, assisted Mr Schulz with his initial reply, and appreciated that Ferrero Germany had received substantial supplies from Baskan Gida under the purchase orders identified. He also took the trouble to copy Mr Casale in on the whole of the exchange taking place between Ferrero Germany and KBC.
779. Turning to Ferrero France, Mr Caroly reported on KBC's inquiry to Mr Casale by email on 30^th April, stating that Ferrero France had no outstanding Baskan Gida invoices, and suggesting that Mr Casale check with Baskan Gida. He was instructed by Mr Casale to tell KBC that all Baskan Gida invoices to Ferrero France had been settled, which he attempted to do on 2^nd May, although it appears that his response may have been lost in transmission, or internally at KBC, before reaching Mr Lewis.
780. As for Ferrero Italy, it simply made no reply to KBC's inquiry at all. Mr Casale sought, unconvincingly to my mind, to explain this away by reference to the fact first that the inquiry had been made to Ferrero Italy's treasury department, and not passed on to him, and that in any event both the original inquiry and KBC's chasing letter were both received on Italian public holidays.
781. The upshot was therefore that the Banks received nothing approaching a constructive response to their detailed inquiry as to what had become of payments which, according to the false copy invoices which they had received, should have been made to them by Ferrero. Save for the loss in transit of Ferrero France's curt response, the Banks had received in reply precisely what Ferrero wished them to receive. For its part, Ferrero had by the end of April received the unambiguous warning that something had gone so gravely wrong in the operation of the Banks' Facility that the Banks were persisting in a claim to be entitled to payment by Ferrero of approximately €24 million by reference to invoices of which the Banks had given full particulars, but of which Ferrero had no records, and by reference to contract and purchase order numbers in relation to which Ferrero had received very substantial deliveries for which it had paid Baskan Gida in full. Plainly, Melih Baskan's assurance that it was not Ferrero's problem, and had been sorted out with the Banks was, by the end of April, demonstrably incorrect, and Ferrero faced at least the risk that, if the underlying receivables had indeed been duly assigned, and proper notice given to Ferrero, that it might find itself obliged to pay twice for the same goods.
782. The Banks' response to the impasse was to leave any further pursuit of inquiries of Ferrero to their lawyers. Ferrero's response was to pursue an internal inquiry, and to raise the matter further with the Baskans, a process which culminated at an internal meeting in Frankfurt on the 21^st to 22^nd May, in relation to which, despite there being no claim for privilege, no documentary record has been disclosed.

Baskan Yuksel

783. Ferrero's inquiry was interrupted by a request purporting to come from Aksu Gida to transfer all outstanding orders to a new company, Baskan Yuksel. This change was prompted by the increasing difficulties encountered by Aksu Gida in continuing its hazelnut business in the face of Yapi Kredi's hostile proceedings, and, in relation to Ferrero, was triggered by Ferrero's decision on Mr Do's instructions, not to accede to Aksu Gida's request to make payments to Mr Ozel.
784. Mr Casale said that the first intimation of this change to Ferrero came in the form of a telephone conversation between him and Mr Franko in early May. Whether or not it was really Mr Franko or rather Melih Baskan, (which I consider more probable in relation to an issue of this importance), formal requests were sent to the Ferrero Operating Defendants and Ferrero Industrial on 7^th May. Save that the outstanding order numbers and quantities were different, and for the name change necessary to convert them into transfer letters from Aksu Gida rather than from Baskan Gida, each of the documents was in precisely the same form as the February Transfer Letters, and the Ferrero Defendants complied with the request in exactly the same way. The Ferrero operating companies sent Baskan Yuksel fresh purchase orders for the balance of the amounts outstanding under the existing Aksu Gida orders, and deliveries continued to be made by Baskan Yuksel, from the same premises, in exactly the same way as they had previously been made by Aksu Gida. It was, incidentally, at this stage that Ferrero Italy rescheduled delivery under contract 5105 for the following season, the time having by then passed for deliveries which would satisfy Ferrero Italy's quality control in relation to the 2001 crop.
785. Baskan Yuksel had in earlier years been a partnership between the Baskan family and a Mr Yuksel. Its legal ownership by May 2002 is not clear but, if it was owned otherwise than by the Baskan family, it was just as much a front for the continued disguised conduct of their hazelnut business by the Baskans as had been Aksu Gida. Mrs Pay continued to be employed in the business, and in response to my question said that during the Baskan Yuksel period, her supervisor was Melih Baskan. No attempt was made to re-construct Baskan Yuksel as a company owned by Indo-Med and Latif Aksu, and the transfer to Baskan Yuksel appears not to have caused a ripple on the surface of the commercial understanding between the Dahods, Mr Abidali and the Baskans. They continued to be engaged in a three cornered joint venture or partnership in the conduct of the hazelnut business, with the Dahod Investors and Indo-Med continuing to enjoy the same profit share as previously.
786. For local consumption however, a sham processing contract was constructed between Baskan Yuksel and Aksu Gida, pursuant to which hazelnuts purportedly purchased by Baskan Yuksel for re-sale (including to Ferrero) were to be processed for a fee by Aksu Gida at the Giresun factory. In practice, the entire operation of purchase, processing and re-export, together with the administration of the necessary documentation, continued to be carried out at the Giresun premises. Nonetheless the cash flow generated by the sale and purchase of hazelnuts was routed through a bank account of Baskan Yuksel, thereby conveniently avoiding the effect of the Turkish freezing injunction on Aksu Gida's account.
787. If, which I consider unlikely, there had by then been any lingering doubt in the minds of the responsible employees at Ferrero as to the ultimate control of Aksu Gida by the Baskan family, the circumstances in which the Aksu Gida transfer to Baskan Yuksel was effected and notified to Ferrero were such as to remove it, notwithstanding the evidence of all Ferrero's main witnesses to the contrary. Ferrero's evidence, which I do accept, was that the motivation for agreeing to transfer outstanding orders from Aksu Gida to Baskan Yuksel was, again, to preserve if possible their supply chains for what continued to be high quality hazelnuts. By the time the Ferrero Operating Defendants placed their fresh orders with Baskan Yuksel, the hazelnut market had fallen substantially in price, such that the prices originally agreed in and before February with Baskan Gida, and transferred unaltered to Aksu Gida, were significantly greater than the lowest prices obtainable in the market by early May. Nonetheless the new orders were placed at the same prices as had been negotiated in and before February, Mr Casale's evidence being that Ferrero did not think it right to renegotiate them. If Baskan Yuksel had been, genuinely, a separate concern from Aksu Gida, it is difficult to conceive of any reason why Ferrero should have agreed to reorder nuts outstanding from Aksu Gida at what had become the unnecessarily high prices originally negotiated with the Baskans. By contrast, if (as I find) Ferrero understood that Baskan Yuksel was just another front for substantially the same family business concern, it is entirely understandable that the prices were not changed.

Internal enquiry within Ferrero

788. Besides beginning an internal inquiry into the genesis of the issue reflected in the Banks' repeated claims for outstanding payments, Ferrero also obtained from Melih Baskan confirmation on Baskan Gida headed paper that each of the Ferrero Operating Defendants had duly paid all sums owing on the orders in respect of which the Banks were making claims. These documents were faxed to Ferrero on 15^th May, but backdated to different dates in February.
789. Mr Casale said that he requested this confirmation from Melih Baskan at a meeting which he and Mr Rosa Brunet had with him purely by chance while they were stopping over at a hotel in Samsun in the course of a journey to Turkey about cherries. He said that he took the opportunity of that chance meeting to inquire of Melih Baskan about the reason for the Banks' continuing claims, that he was reassured that Melih Baskan would sort it out directly with KBC and, then or on some other occasion, promised to provide the written confirmations given. Mr Casale denied in cross-examination that he had asked for the confirmations to be backdated and, on balance, I believe him. I am by no means persuaded by his evidence that he failed to notice that the documents, once received, had been backdated. In my judgment he had no particular motive for asking for backdating, and probably regarded the dates on the documents, once received, as not being worth making an issue about. I have had difficulty suspending my disbelief that the meeting between Mr Casale, Mr Rosa Brunet and Mr Baskan in Samsun was a pure coincidence, but since Mr Casale had every opportunity to discuss matters with Melih Baskan on the telephone whenever he wished to do so, and in all probability did so on numerous occasions, the question whether their meeting was or was not pre-arranged does not greatly matter.
790. The internal Ferrero meeting at Frankfurt did include a (probably pre-arranged) session dealing with the Banks' continuing claims. It was attended by Mr Rosa Brunet, Mr Do, Mr Bolowich and Mr Casale, and it is extraordinary that no documentary record of any kind has survived in relation to it. Nonetheless, the evidence of the Ferrero witnesses was that it led to a decision that an internal investigation into, and any response to, the Banks' claim should be conducted centrally by Ferrero Industrial, rather than separately by each operating company, and that Ferrero should instruct a Turkish lawyer, in the event a Mr Gun, to assist them with their inquiries.
791. During June and July Ms Johns of BTM's legal department made repeated contacts with in-house lawyers at Ferrero for the purpose of arranging a meeting with senior Ferrero staff, but without success. Mr Do sought in cross-examination to explain Ferrero's refusal to meet the Banks as based on their failure to provide an adequate agenda for any meeting. I reject that. There is no evidence in the party and party correspondence that such an agenda was requested. In my judgment, just as the Banks had been reluctant in April to meet Ferrero while possessed of inadequate information, so Ferrero were reluctant to meet the Banks in June and July, in advance of concluding their own internal investigation. Eventually, the Banks were told to meet Mr Gun, and this eventually took place on 25^th September.
792. By this time Ferrero had made a thorough study of the orders, contracts, invoices and deliveries involving Baskan Gida and Aksu Gida, part of which survive in the form of tables prepared by Mr Altare and sent to Mr Casale on the 23^rd July. Ferrero had also obtained from the Banks copies of the assignment documentation relied upon by the Banks for their assertion that they were owed over €22 million by Ferrero, including the Acknowledgements purportedly signed by the Ferrero Operating Defendants, but which had in fact been forged by the Baskans. These were sent by Ms Johns on 30^th July, by which time there had also been brief and inconclusive meetings between Mr Casale and Melih Baskan on 25^th July and between Mr Casale and Mr Byles on the following day.
793. In their evidence, the Ferrero witnesses dated Ferrero's perception that there had been a fraud from their study of the forged copy Acknowledgements. It appears from the Ferrero witnesses' evidence that, from the moment when Ferrero appreciated (on expert advice) that the Acknowledgements were forgeries, they assumed that the Baskans had been the perpetrators. They describe having put this allegation to the Baskans at a meeting in Mr Gun's office in Istanbul in August, and it is evident that Mr Do and his colleagues were not persuaded by the Baskans' denial.
794. This did not however stop Ferrero continuing to deal with the Baskans, under the guise of Baskan Yuksel, in connection with the hazelnut purchasing programme for the new season. In this respect I reject the Ferrero witnesses' repeated assertion that they did not even at this late stage understand that Baskan Yuksel was a front for the Baskans. I have been assisted in reaching that conclusion by two documents, neither of which is on its own conclusive, but which together reinforce the wealth of other evidence both that Aksu Gida and Baskan Yuksel were fronts for the Baskans, and that Ferrero knew it. The first is an attendance note made by Mr Trotter of a conversation with Mr Abidali on the 13^th August, the material part of which reads as follows:

"Baskan has had a very constructive meeting with Ferrero's in Turkey. They have declared themselves happy with the Baskan supply and connection and wish to continue uninterrupted. With regard to the Bank of Tokyo issue, it has been agreed that if anything arises for liability of Ferrero then Baskan will cover it entirely. On that basis Ferrero are simply pressing ahead."

Since Mr Abidali did not claim to have attended the relevant meeting, Mr Trotter's attendance note stands as third hand hearsay of what Melih Baskan must have told Mr Abidali. Furthermore Melih Baskan may have had good reason to gild the lily when dealing with Mr Abidali as a representative of the Dahod Investors.

795. The second document is a letter of advice from Mr Gun to Mr Casale by email on 26^th September, the day after Mr Gun's meeting with representatives of the Banks at which, according to the Banks' contemporaneous note, Mr Gun had expressed Ferrero's refusal to cooperate further with the Banks until it received a full explanation of the forged Acknowledgments, having already received the Baskans' denial of any participation in the forgery. Mr Gun's advice is redacted on the grounds of privilege, but the letter contains the following description of a conversation between him and Cevat Baskan, following the meeting with the Banks:

"I have spoken to Cevat Baskan yesterday and today and informed him that I met KBC bank representatives and told them that the notices and acknowledgements were fake, but, they insisted they were valid. I told him that my impression was that KBC bank had no chance of recovery from Baskan and therefore tried to sue Ferrero as a last hope, based on the notices and acknowledgements. I said the issue whether or not KBC would be successful was a separate matter but it would be risky for Ferrero to continue business with Baskan or any other companies associated with Baskan as KBC would interpret this co-operational collusion between Baskan and Ferrero. I told him that in these circumstances I could not recommend Ferrero to continue business with either Aksu, Baskan Yuksel or any other company associated with Baskan as it would increase the likelihood of KBC bringing proceedings versus Ferrero.…"

Although Mr Gun's letter speaks of Aksu Gida and Baskan Yuksel being "associated with" rather than controlled by the Baskans, the sub-text is clear. Furthermore, as these proceedings demonstrate, Mr Gun's warning of the risks for Ferrero in continuing to deal with such Baskan related entities was prophetic. In fact, Ferrero ignored Mr Gun's warning and continued to deal with the Baskans through Baskan Yuksel until the end of 2002, after which this and other litigation with the Banks ensued.

Litigation Begins

796. The trigger for litigation was the Banks' letters before action to each of the Ferrero Defendants dated 20^th December 2002, claiming as assignee of Baskan Gida's Ferrero receivables, and based upon the Framework Contract, the Master Assignment, the Notice of Assignment of the Framework Contract to Ferrero Industrial, and Ferrero Industrial's Acknowledgement dated 17^th December. The letter to each operating company continued:

"We believe that a notice in similar terms to the one sent to [Ferrero Industrial] may have been sent to [the name of the operating company] on 19^th December 2001 and acknowledged shortly afterwards."

The sentence quoted above is the only reference in the letters before action to the forged Acknowledgments.

797. Ferrero's response took the form of a criminal complaint lodged with the Public Prosecutor in Alba by Ferrero Industrial and Ferrero Italy (signed by Mr Rosa Brunet on behalf of Ferrero Industrial), alleging that the Banks had made use of forged documents to pursue a false claim. When in July 2004 the Public Prosecutor invited the preliminary investigating magistrate at the Alba court to dismiss the charge, Ferrero Italy and Ferrero Industrial (the latter acting by Mr Do) objected to the dismissal, alleging in particular that Ms Johns and Mr Head, the partner at Forsters with the conduct of the Banks' litigation against Ferrero, had personally incurred criminal responsibility by reason of their use of the forged Acknowledgments.
798. In November 2004 the first instance court dismissed the complaint. Ferrero nonetheless appealed to the Court of Cassation, and the criminal complaint was only finally dismissed by that court in May 2005.

Ferrero's Conduct from March 2002

799. I now return to express my conclusions on the issue whether Ferrero's conduct following the Banks' discovery of the fraud was that of a guilty participant in it, seeking to avoid being found out. In my judgment Ferrero's case, and its witnesses in their evidence, have consistently and deliberately played down the extent of their appreciation, from March 2002 onwards, that the Banks were the victims of a fraud by the Baskans. I consider that what had before the end of February been a mere suspicion that the Baskans might not have been dealing frankly with the Banks matured in early March into a belief at least in Mr Casale's mind that they probably had defrauded the Banks. Whenever it was in early March that he studied the Banks' 27^th February email, he must have appreciated from the fact that Ferrero Italy had received none of those invoices which the Banks itemised that the probable explanation was an invoicing fraud. Once he had been copied in by Mr Bolowich and informed by Mr Caroly of the similar requests sent in late April to Ferrero Germany and Ferrero France, also relying upon copy invoices none of which appeared in the operating companies' records, the conclusion would be hard to avoid.
800. Mr Casale's evidence was that it was not until he read a letter from Ms Johns dated 3^rd July that he realised that the amount being claimed by the Banks from Ferrero was as high as €22 million. In my judgment he understood that perfectly well by the end of April, once informed about the amounts being claimed against Ferrero France and Ferrero Germany, to which he must have added the amount which he knew was being claimed from Ferrero Italy.
801. Since I do not doubt Mr Casale's complete loyalty to his employer, I infer that, faced with claims from the Banks which he knew by late April to amount in aggregate to more than €20 million, he both reported that fact and his belief as to the probable reason for it to Mr Rosa Brunet and to Mr Do.
802. The Banks may therefore fairly ask rhetorically why should Ferrero, if believing that the Banks were victims of a fraud by the Baskans, have been so uncooperative in response to the Banks' inquiries, and why during this trial should Ferrero and its witnesses have played down the extent of their belief to that effect, to the point of lying about it, and for that matter lying about their knowledge that Aksu Gida and Baskan Yuksel were front companies for the Baskans, if not for the purpose of seeking to protect themselves from identification as participants in that fraud? Such conduct is undoubtedly consistent with an inference of guilty participation. The question is whether it is probative of it.
803. My conclusion is that Ferrero's conduct in and after March 2002 falls a little short of justifying that inference, and is to be explained on other grounds, as follows. By the beginning of March Ferrero had committed itself to obtaining the bulk of its hazelnut requirements from the 2001 crop from the Baskan front company Aksu Gida, having done so in circumstances in which the proposed transfer of Baskan Gida's business was not perceived to give rise to any risk that Ferrero might be liable to make double payment in respect of the same hazelnuts. The Banks' email of 27^th February suggested to Mr Casale for the first time that this assumption might not have been correct. By the end of April Ferrero was facing claims from the Banks in excess of €20 million, entirely in respect of hazelnut deliveries for which it had already made payment in full, apparently as the result of the conduct of the Baskans, upon whom Ferrero was continuing to rely for future high quality supplies.
804. In my judgment the acutely worrying uncertainty in the minds of Ferrero's responsible managers (Messrs Casale, Rosa Brunet and Do) during the whole of the period from March 2002 onwards was not whether there had been a fraud by the Baskans but whether there was a real risk that Ferrero rather than the Banks were potential victims of it.
805. It was I consider obvious to Ferrero from an early stage that, if there was a real risk of liability to make double payment pursuant to the Banks' claims, there was little or no prospect that the Baskans, as authors of Ferrero's predicament, would have the resources to satisfy the Banks' claims, at least in the short or medium term, despite the Baskans repeatedly proffering to Ferrero their desire to do so. The result is that Ferrero accurately appreciated that the likely outcome of the Baskans' fraud would be a very serious dispute between Ferrero and the Banks as its joint or alternative victims, and one in which, rightly or wrongly, Ferrero did not envisage that its best interests would be served by cooperation with, or the provision of assistance to, the Banks in their inquiries. Like the Banks in March, Ferrero lay low until, at the end of July, the emergence of the forged Acknowledgements as an important part of the Banks' case appeared to put a defensive weapon into Ferrero's hands, thereafter used as the basis for refusing further cooperation and, after the Banks' letters before action, for the aggressive defence constituted by the criminal complaint.
806. Whilst that analysis sufficiently explains Ferrero's conduct towards the Banks from March to the end of 2002, it leaves outstanding, as calling for further explanation, the question why Ferrero should have chosen to continue to do business with the Baskans through Aksu Gida and Baskan Yuksel for the rest of 2002, to have decided to put forward a false case that they did not know that they were front companies for the Baskans, and to support that case by lying about it in evidence.
807. In my judgment Ferrero continued to do business with the Baskans largely because it had committed itself to that course shortly after the 25^th January meeting with Melih Baskan, so that at no time thereafter did a cold calculation of Ferrero's commercial interests favour an about-turn. The basic motivation remained, namely to preserve, if possible, a supply chain which had served Ferrero very well in terms of quality and quantity over many years. It may be that Ferrero calculated (wrongly in the event) that it would be impossible for any third party such as the Banks subsequently to prove that Ferrero knew in 2002 that both Aksu Gida and Baskan Yuksel were indeed front companies for the Baskans.
808. As for the false case presented to the contrary in this litigation, I consider that the motivation lay not, as Mr Wardell submitted, in the desire of Mr Casale and Mr Rosa Brunet to conceal their disloyal shortcomings from their employers, but in Ferrero's desire, shared by them and by Mr Do, to avoid painting Ferrero into precisely the corner about which Mr Gun warned in the letter from which I have quoted, namely that of becoming tainted by the Baskans' fraud by virtue of a continuing association with their companies. It is perhaps an irony that the combined effect of initial obstructiveness, continued dealings with the Baskans' companies and subsequent untruths about their state of knowledge merely increased that suspicion of complicity which led the Banks to make the conspiracy allegations which they have against Ferrero in this litigation.

Dealings between Mr Abidali, Mr Dahod and the Baskans from March 2002 onwards

809. I have left until last an account of the continuing relationship between the Baskans, Indo-Med, Mr Abadali and the Dahods, from March until the end of 2002. It was, in short, business as usual, in that the profit sharing joint venture was transferred seamlessly from Baskan Gida to Aksu Gida, and from Aksu Gida to Baskan Yuksel, without any apparent change in Mr Dahod's understanding of the substance of the business relationship between the parties. It is convenient to analyse the evidence about this relationship in three strands. The first concerns Mr Abidali's continued efforts to assist the Baskans in the promulgation in Turkey of the sham transactions which I have described, mainly for the purpose of fending off hostile claims by Yapi Kredi. The second consists of routine communications about day to day trading matters, arising from the fact that, in addition to being the nominee holder of the Dahod Investors' shareholding in Aksu Gida, Indo-Med continued to be a major trading partner with the Baskan business, in both its Aksu Gida and Baskan Yuksel stages. The third strand consists of the correspondence bearing directly upon the true nature of the parties' mutual interest in that business.
810. With the assistance of the Baskans' legal team (headed by Mr Alpay) Indo-Med took the appropriate steps to obtain registration in Turkey as 80% shareholder of Aksu Gida. More importantly Mr Abidali contributed an affidavit to Aksu Gida's defence of Yapi Kredi's claim, in June 2002. The affidavit contained three connected lies. The first was that the Indo-Med funding of US$12.55 million (a reference, without naming them, to the Dahod Investors' contribution) had been for the pre-finance of raw material purchases by Baskan Gida for supply to Indo-Med (whereas in truth it had been for the pre-finance of supplies to Ferrero). The second was that the Protocol Agreement and promissory notes had indeed been made in November 2001, so that the debt had become due but unpaid in January 2002. The third was that the genesis of the transfer to Aksu Gida was a providential meeting between Indo-Med and Mr Aksu which enabled Indo-Med, by enforcement of its debt, to exploit an opportunity to enable Aksu Gida to continue Indo-Med's supply of hazelnuts. Thus the Baskan Gida/Aksu Gida business transfer was explained as the result of commercial pressure by Aksu Gida's new owners to enforce an overdue debt, a picture apparently verified by the sham and backdated documents, and which Mr Abidali well knew to be false when making his affidavit.
811. Although the interim relief which Yapi Kredi obtained, for example freezing Aksu Gida's bank account, seriously damaged the future of the hazelnut export business such that, in the end, it did not survive, Yapi Kredi's claim to set aside the asset transfer from Baskan Gida to Aksu Gida as a transaction at an undervalue ultimately failed before the Turkish courts, which appear to have accepted at face value the transactions recorded in what I have concluded are the sham documents, as verified by Mr Abidali's affidavit.
812. There survive a large number of communications between Mr Abidali and the Baskans dealing with the ongoing routine aspects of the Indo-Med/Baskan trading relationship during 2002. Mr Abidali was taken to them in detail in cross-examination by Mr Wardell, and Mr Werbicki provided a helpful explanatory table of them in a schedule to his written closing submissions.
813. It is unnecessary for me to go through these documents in detail, notwithstanding Mr Werbicki's submission that, whether taken singly or together, they fall well short of showing that Mr Abidali regarded the Baskans as remaining in managerial control of the business conducted by Aksu Gida and Baskan Yuksel throughout. In my judgment they clearly justify an inference precisely to that effect. They show Mr Abidali and other employees at Indo-Med consulting the Baskans (mainly Melih Baskan) in relation to a range of routine matters relevant to the Aksu Gida/Baskan Gida hazelnut business, usually in relation to issues arising from the on-sale to its own customers by Indo-Med of nuts supplied by Aksu Gida or Baskan Yuksel.
814. I shall content myself with one example, namely a series of emails between Mr Abidali, his brother Saifuddin (based in Cairo) and Melih Baskan, copied to Mrs Pay and Mr Franko, relating to a consignment of nuts ordered by Indo-Med from Aksu Gida for delivery to a customer in Egypt. Due to incorrect labelling the Egyptian customer was forced to pay excess import duty and Saifuddin Abidali complained to his brother Mr Abidali about it. The gist of the exchange was that the labelling mistake was treated as Melih Baskan's responsibility. He personally apologised to Saifuddin Abidali for the difficulty and agreed to bear the financial consequences. When an appropriate deduction was agreed, the balance of the payment for the consignment was treated as a "balance due to Baskan" for a supply by the front company. In cross-examination, Mr Abidali was forced to accept that if the Baskans were not in substance in control of Aksu Gida, then he had in substance been lying to his own brother in private emails.
815. The final element of this evidence consists of communications relating directly to the nature of the relationship between the Dahod Investors, Indo-Med, Mr Abidali and the Baskans. I broke off my description of that chain of communications after referring to the items which, after meeting Mr Dahod in Boston in January 2002, Melih Baskan had promised to deal with once "back on station" in Turkey.
816. The outstanding points appear to have been the subject of a face to face discussion in New York in April, and profit share payments at the rate of US$100,000 a time appear to have been planned, as payments on account of the profit shares due under the Dahod Agreement. At the end of May Mr Dahod was chasing Mr Abidali for a "status on Baskan" and expecting a remittance of profit share payments. At the end of August Mr Abidali sent Mr Dahod a summary account of tonnage delivered to Ferrero from the 2001 crop together with the Dahod Investors' profit share, amounting to US$3.8 million, of which Mr Abidali said that US$1.2 million had already been paid on account from money received in instalments from Turkey. He referred to part of the 2001 crop as having been carried forward to the new crop (no doubt an oblique reference to the postponement of delivery date agreed by Ferrero Italy for the bulk of its last order). Mr Abidali concluded by stating that "the total capital employed remains US$13.15 million".
817. On 14^th October Mr Abidali emailed Mr Dahod to pass on the Baskans' detailed proposals for a continuation of the profit sharing joint venture into the 2002 crop season. The email demonstrates to my mind more clearly than any other single document that Mr Abidali regarded Aksu Gida as a Baskan business, funded largely by working capital provided by the Dahod Investors under the Dahod Agreement, but by October 2002 being used for the whole of the Baskans' business and not just its exports to Ferrero. Furthermore, the sham transfer by Aksu Gida to Baskan Yuksel is not even mentioned, although by then it had purportedly occurred several months earlier. It is worth quoting in full:

"Dear Murtaza bhai

Melih apologises for the delay in putting together the figures he had promised.

Whilst the following are not exact figures they are pretty close rounded ones.

Currently the total working capital available to Aksu is about USD 15 million of which about USD 4 million is blocked in slow moving carry forward stock (small sizes for blending and reprocessing purposes)

The balance cash is being used for the whole business, i.e. in addition to the Ferrero biz as it is necessary to run parallel with other customers to enable the movement of all sizes and grades. This is essential as the old crop goods cannot be blended into Ferrero merchandise.

This limits the volume of business they can commit to Ferrero. Their estimate is that at the present pace of turning money round they may not be able to maintain the tonnage that they must move to liquidate the old crop effectively. On the other hand they must do the Ferrero business also and they are in a bind.

They are in the peak of the season and this is the time that they must make the commitments or decide their limitations.

On the current financial resource, they estimate that they may not be able to do more than about 20,000 MT in all. Of course as they liquidate more of the carry forward stock this will ease the situation.

They estimate that they must have a minimum of liquid USD 20 million to enable them to export at least 40,000 MT of which 50% would be Ferrero.

They accept and realise they are currently using your investment funds for business other than the Ferrero business only. There has to therefore be a sharing of the benefit on this side also. However there are some aspects that we have to consider.

1) There can also be a loss factor on this and our principals require a sharing of loss or profit (otherwise it is interest)

2) The profit generation on this side can and generally is a higher percentage as there is a speculative aspect to it as well, unlike the Ferrero business which is a cost plus basis.

3) Baskan also need to rebuild their capital and cannot afford to give away all of the profit element.

They are therefore also confused as to what precise proposal to make to you/us.

The present cash resource will definitely not allow them to maintain last seasons volume of Ferrero biz.

On the other hand they are asking for more working capital.

They are embarrassed.

In year 2001 crop they shipped a total of 57,000 MT.

On the tonnage besides Ferrero they targeted to earn a nett profit of at least 5%. Actually they earn much more because of production efficiencies and product mix. However last year because of all the other factors that came into play they did not achieve this. They ended the year with a loss of about USD 2.5 million instead. In previous years, including paying a heavy rate of interest to the banks they have easily achieved a much higher rate of nett income.

In my discussion with Melih, and from his discussions with his father and other family members, they are asking for what you would consider a fair proposal keeping in mind all these factors.

What is clear, is that this trading year appears to be a perfect year for making a substantial profit and very much more also very important with the whole client base for establishing the future continuity and reliability of the 'Baskan' family.

They would rather that you were able to actually see what they are all about and for this purpose still await your visit earnestly. It would make them feel better in asking for a larger working capital and participation in the overall business.

The decision on then Aksu case is expected to be determined within this month.

Please let me know your thoughts.

Salaams

Shabbir"

818. Despite Mr Abidali's best efforts in cross-examination, and Mr Werbicki's in closing submissions, it is unmistakeable that by the repeated reference to "they" Mr Abidali was referring to the Baskans and not merely to Aksu Gida. Mr Werbicki submitted that this was merely an offer by which the Baskans were seeking to re-enter a business by the owned and controlled by the Dahod Investors, Indo-Med and Mr Aksu. On the contrary, the email plainly demonstrates Mr Abidali speaking of a business then and there being run by the Baskans, albeit through Aksu, for example in the observation that:

"They are in peak of the season and this is the time that they must make the commitments or decide their limitations."

819. Mr Dahod replied on the 28^th October with a short email declining to commit further capital to the business, but enclosing a one page summary of his views entitled "Baskan Investment: First Year Perspective". Again, it is worth quoting in full:

"The relationship with Baskan was established to provide working capital to Baskan for all Hazelnut purchases by Ferrero. The investors (the various Dahod entities and Indo-Med) were to receive a fixed return of $140 per ton of Hazelnut purchases by Ferrero. The expected volume for Ferrero was a minimum of 30,000 tons. Consequently, an investment of $13,150,000.00 was expected to return $2,940,000.oo per year to the Dahod entities.

The investor's capital was considered to be relatively safe and liquidity was expected to be high except during the peak of the harvest. Also, the capital was expected to cycle through the investors during the year. The expectation was that as the capital cycled through the Dahod entities, they would have the opportunity to make commitments on an annual basis and also, change the relative contribution among them before the start of each season. Further, the cash flow was expected to be reasonably predictable to allow cash management by the investors. The major risk in the return was the total shipment to Ferrero.

While the total Ferrero booking (31,525 tons) was more than the minimum target, the shipment (27,144 tons) was below expectations. Consequently, the total return to the Dahod entities was only $2,660,112.00. However, the cash flow was very unpredictable and not only that the return has not been fully paid yet, but also, the capital was never cycled through the investors. The unpredictability of the cash flow forced the Dahod entities to execute unplanned liquidation of other assets to meet their cash commitments. Such unscheduled liquidation resulted in a significant penalty.

In reality, the events of the first year have dramatically altered the risk-reward profile of the investment as a result of the major reorganization of Baskan. The investors' capital is now supporting most of the shipment and not just the ferrero shipment since the investors' capital makes up $14,350,000.00 ($13,150,000.00 of the original investment plus the $1,200,000.00 of reinvested return from the first year) of the $15,000,000.00 working capital. As a result, the investors' capital is not liquid and cannot cycle through the investors. The return on the shipment to customers besides Ferrero is unknown and unpredictable. Consequently, the total return on the investment is not concomitant to the risk of the investment. The unresolved legal issues faced by Aksu further heighten this risk.

On the very positive front, the investors value the relationship that has been developed with Shaikh Shabbirbhai and Melih Baskan. Also, the Baskan family seems to be a very hardworking and tightly knit team with a very successful history in the hazelnut business. The potential for success remains to be very promising if the terms and conditions can be quickly modified to reflect the new reality. Also, the cash flow needs of the Dahod entities have to be considered to avoid continued unexpected and expensive liquidation of other assets."

820. It is to be noted that the 27,144mt of hazelnuts referred to as having been delivered during the first season, in the third paragraph of Mr Dahod's note, necessarily includes the whole of the deliveries to Ferrero by Baskan Gida, Aksu Gida and Baskan Yuksel. Far from being an aggregate of deliveries by a Baskan company, a Dahod company and a third company of unknown parentage, Mr Dahod described the deliveries as having been made during the period of a "major reorganization of Baskan" in which the profit sharing partnership with the Baskan family had continued both throughout, and successfully.
821. Nothing in Mr Abidali's or Mr Dahod's evidence about these documents in cross-examination detracted in any way from their force in demonstrating that the true relationship between them and the Baskans was as I have described it, rather than as portrayed in the sham documents. On the contrary, their refusal to face the realities demonstrated by these two documents went a long way to persuade me that both of them were entirely unreliable witnesses, and determined to lie whenever necessary.
822. I have considered whether the picture presented by this exchange between Mr Abidali and Mr Dahod was merely a private fiction which Mr Abidali practised upon Mr Dahod, so as to insulate him from what was happening between Indo-Med and the Baskans. While, as I have concluded, Mr Dahod was not shown any of the sham Baskan Gida/Aksu Gida transfer documents, and probably not told about the structure which they purported to create, there is no doubt that the Baskans themselves also regarded the substance of the profit sharing joint venture as being unaffected by the transfer, as appears from a draft of a letter which Melih Baskan proposed to send to Mr Dahod, but copied to Mr Abidali for his comments in February 2003. That draft clearly portrays the Baskans as owners and controllers of the hazelnut business throughout, and described the Baskan Gida/Aksu Gida transaction merely as one pursuant to which the Dahod Investors' "credit" was transferred by one company to the other, leaving the Dahods as major lenders to, rather than owners of, Aksu Gida.
823. By way of postscript, Mr Raniwala's understanding remained that the Indo-Med trading relationship remained in substance one with the Baskans during 2002, albeit subject to reorganisation of its structure. This is what he told Chantrey Vellacott when assisting them with the preparation of Indo-Med's accounts in April 2002, as he acknowledged in cross examination.

LAW AND ANALYSIS

Conspiracy to injure by unlawful means

824. This is the main cause of action advanced by the Banks against the Ferrero Defendants, and the only cause of action alleged against Mr Abidali.
825. This tort is commonly labelled "unlawful means conspiracy" so as to distinguish it from the related tort generally known as "conspiracy to injure". The label "unlawful means" is of some value, in the sense that it identifies unlawful means as an element in the cause of action. The label is unhelpful in drawing attention away from the continuing requirement in all types of conspiracy claim to demonstrate an intent by the defendant to injure the claimant. It is in truth a sub-species of the tort of conspiracy to injure, in which the ordinary requirement that such intent be predominant in the mind of the defendant is replaced by the requirement to show that unlawful conduct has been the means of the intentional infliction of harm to the claimant.
826. That this is so was trenchantly laid down by the House of Lords in Lonrho Ltd v. Shell Petroleum Ltd (No 2) [1982] AC 173, reaffirmed in Lonrho plc v. Fayed [1992] 1 AC 448 and spelt out in the clearest of terms more recently in OBG Ltd v. Allan [2008] 1 AC 1, in particular in the following passage from Lord Nicholls' speech at pages 57 to 58:

"164. I turn next, and more shortly, to the other key ingredient of this tort: the defendant's intention to harm the claimant. A defendant may intend to harm the claimant's business either as an end in itself or as a means to an end. A defendant may intend to harm the claimant as an end in itself where, for instance, he has a grudge against the claimant. More usually a defendant intentionally inflicts harm on a claimant's business as a means to an end. He inflicts damage as the means whereby to protect or promote his own economic interests.

165. Intentional harm inflicted against a claimant in either of these circumstances satisfies the mental ingredient of this tort. This is so even if the defendant does not wish to harm the claimant, in the sense that he would prefer that the claimant were not standing in his way.

166. Lesser states of mind do not suffice. A high degree of blameworthiness is called for, because intention serves as the factor which justifies imposing liability on the defendant for loss caused by a wrong otherwise not actionable by the claimant against the defendant. The defendant's conduct in relation to the loss must be deliberate. In particular, a defendant's foresight that his unlawful conduct may or will probably damage the claimant cannot be equated with intention for this purpose. The defendant must intend to injure the claimant.

167. I add one explanatory gloss to the above. Take a case where a defendant seeks to advance his own business by pursuing a course of conduct which he knows will, in the very nature of things, necessarily be injurious to the claimant. In other words, a case where loss to the claimant is the observe side of the coin from gain to the defendant. The defendant's gain and the claimant's loss are, to the defendant's knowledge, inseparably linked. The defendant cannot obtain the one without bringing about the other. If the defendant goes ahead in such a case in order to obtain the gain he seeks, his state of mind will satisfy the mental ingredient of the unlawful interference tort. This accords with the approach adopted by Lord Sumner in Sorrel v Smith [1925] AC 700, 742:

'When the whole object of the defendants' action is to capture the plaintiff's business, their gain must be his loss. How stands the matter then? The difference disappears. The defendants' success is the plaintiff's extinction, and they cannot seek the one without ensuing the other.'"

827. To the same effect Lord Hoffmann said this, at page 30:

"42 … people seldom knowingly cause loss by unlawful means out of simple disinterested malice. It is usually to achieve the further end of securing an economic advantage to themselves….

43. On the other hand, if the breach of contract is neither an end in itself nor a means to an end, but merely a foreseeable consequence, then in my opinion it cannot for this purpose to be intended. That, I think, is what judges and writers mean when they say that the claimant must have been "targeted" or "aimed at"."

828. Later, at page 35 he said this:

"62. Finally, there is the question of intention. In the Lumley v Gye tort, there must be an intention to procure a breach of contract. In the unlawful means of tort, there must be an intention to cause loss. The ends which must have been intended are different. South Wales Miners' Federation v Glamorgan Coal Co Ltd [1905] AC 239 shows that one may intend to procure a breach of contract without intending to cause loss. Likewise, one may intend to cause loss without intending to procure a breach of contract. But the concept of intention is in both cases the same. In both cases it is necessary to distinguish between ends, means and consequences. One intends to cause loss even though it is the means by which one achieved the end of enriching oneself. On the other hand, one is not liable for loss which is neither a desired end nor a means of attaining it but merely a foreseeable consequence of one's actions."

829. A useful if compressed statement of the essential elements of this form of conspiracy is to be found in the judgment of Nourse LJ in Kuwait Oil Tanker Co v. Al Bader & ors [2000] 2 All ER (Comm) 271, at page 312:

"A conspiracy to injure by unlawful means is actionable where the claimant proves that he has suffered loss or damage as the result of unlawful action taken pursuant to a combination or agreement between the defendant and another person or persons to injure him by unlawful means, whether or not it is the predominant purpose of the defendant to do so."

830. In the same judgment, at page 315, Nourse LJ approved the following dictum of Oliver LJ in Bourgoin SA v. Minister of Agriculture, Fisheries and Food [1986] QB 716 at 777:

"If an act is done deliberately and with knowledge of its consequences, I do not think that the actor can sensibly say that he did not "intend" the consequences or that the act was not "aimed" at the person who, it is known, will suffer them."

831. At first, this dictum is not easy to reconcile with the dicta of Lords Nicholls and Hoffmann in the OBG case to which I have referred, to the effect that knowledge that loss to the claimant is a foreseeable consequence of the actions of the conspirators is insufficient for the purpose of establishing the necessary intent to injure. To the extent that those dicta are irreconcilable, I am of course bound to follow the later observations of their Lordships in the OBG case. In truth, I doubt whether they are as irreconcilable as they might appear at first sight.
832. In Kuwait Oil, the common object of the conspirators was to misappropriate very substantial sums of money from the claimants by fraudulent means. Their predominant purpose was, of course, to line their own pockets, but an equivalent injury to the claimants was, in every sense, the "obverse of the coin" in the sense described by Lord Nicholls in the OBG case. Furthermore the dictum in Bourgoin relied upon by Nourse LJ was used for the limited purpose of answering a submission that the necessary intent to injure could never be inferred from the overt acts of the conspirators. As Nourse LJ himself put it:

"In the case of a conspiracy to defraud by wholesale misappropriation it would be absurd to argue that the conspirators did not intend just that."

833. The OBG case is not in terms a case about conspiracy, but rather about the closely related torts of inducing a breach of contract, and causing loss by unlawful means. Nonetheless, the dicta about intent to injure to which I have referred were treated as fully applicable to conspiracy to injure by unlawful means, in Meretz Investments NV v. ACP Ltd [2008] Ch 244, by the Court of Appeal.
834. Difficulties as to what constitutes the necessary intent are by no means the only aspects of Nourse LJ's convenient summary of the elements of the tort which call for deeper analysis. Of equal importance is the phrase "by unlawful means".
835. It has, since the decision of the House of Lords in Revenue & Customs Commissioners v. Total Network SL [2008] 2 All ER 413 been settled that unlawful means for the purposes of the tort of conspiracy may consist of conduct which is criminally unlawful, even if it gives rise to no private cause of action by the claimant. To this extent the phrase unlawful means has a wider meaning in conspiracy than in the related intentional harm tort. Nonetheless, the unlawful conduct must, if it is to qualify as an element in the tort, be in fact the means whereby the relevant loss in inflicted upon the claimant. As Lord Walker put it in the Total case, at paragraphs 95 to 96 of his speech, (at page 450):

"In my opinion your Lordships should clarify the law by holding that criminal conduct (at common law or by statute), can constitute unlawful means, provided that it is indeed the means (what Lord Nicholls of Birkenhead in the OBG case called … 'instrumentality') of intentionally inflicting harm…. What is important, to my mind, is that in the phrase 'unlawful means' each word has an important part to play. It is not enough that there is an element of unlawfulness somewhere in the story."

836. In the Meretz case, Arden and Toulson LJJ went further by requiring, as a condition of liability, that the defendant knows that the claimant's loss was to be caused by the use of unlawful means. As Toulson LJ put it at paragraph 174 (page 289):

"Just as the tort of conspiracy to induce breach of contract is not committed if the defendant believes that the outcome sought by him will not involve a breach of contract (the Mainstream case [2005] IRLR 964), so a defendant should not be liable for conspiracy to injure by unlawful means if he believes that he has a lawful right to do what he is doing. This is consistent with Lord Hoffmann's comment in the OBG case [2008] 1 AC 1 para 56, when considering the tort of causing injury by unlawful means, that the common law in this area is designed only to enforce basic standards of civilised behaviour."

837. The requirement that the defendant must have known that the relevant conduct was unlawful, and that the unlawful part of it would cause harm to the claimant, gives rise to additional complexity, in two respects relevant to these proceedings. The first concerns the quality of the requisite knowledge, and the second concerns its required extent. I shall address the quality issue first.
838. For different purposes, the law requires proof of a mental element in the defendant's conduct ranging from suspicion at one end to absolute certainty at the other. For the purposes of unlawful means conspiracy, the starting point is in my judgment the decision of the House of Lords in Manifest Shipping Co Ltd v. Uni-Polaris Insurance Co Ltd [2003] 1 Ac 469. Like the OBG case, it was not about conspiracy. There, the question was as to the quality of knowledge which had to be proved against a ship-owner, for him to be held to have been privy to the unseaworthiness of his vessel within the meaning of section 39(5) of the Marine Insurance Act 1906. The relevance of the decision is that the passage from the speech of Lord Scott to which I am about to refer was cited with approval by the Court of Appeal in Attorney General of Zambia v. Meer Care & Desai [2008] EWCA Civ 875, at paragraph 21, as an exemplary description of blind-eye or Nelsonian knowledge in a case specifically about dishonest assistance and unlawful means conspiracy. The passage in question consists of paragraphs 112 to 116 of Lord Scott's speech, which deserves reading in full. I shall content myself with two extracts. In paragraph 112, he began:

"Did the assured have blind-eye knowledge of the unseaworthiness of the Star Sea? It is as well to return to the language of the sub-section. What is required is "privity" on the assured's part of the unseaworthiness. "Privity" in its ordinary meaning connotes knowledge. "Blind-eye" knowledge approximates to knowledge. Nelson at the battle of Copenhagen made a deliberate decision to place the telescope to his blind eye in order to avoid seeing what he knew he would see if he placed it to his good eye. It is, I think, common ground – and if it is not common it should be – that an imputation of blind-eye knowledge requires an amalgam of suspicion that certain facts may exist and a decision to refrain from taking any step to confirm their existence."

Concluding at paragraph 116, he said this:

"In summary, blind-eye knowledge requires, in my opinion, a suspicion that the relevant facts do exist and a deliberate decision to avoid confirming that they exist. But a warning should be sounded. Suspicion is a word that can be used to describe a state-of-mind that may, at one extreme, be no more than a vague feeling of unease and, at the other extreme, reflect a firm belief in the existence of the relevant facts. In my opinion, in order for their to be blind-eye knowledge the suspicion must be firmly grounded and targeted on specific facts. The deliberate decision must be a decision to avoid obtaining confirmation of facts in whose existence the individual has good reason to believe. To allow blind-eye knowledge to be constituted by a decision not to inquire into an untargeted or speculative suspicion would be to allow negligence, albeit gross, to be the basis of a finding of privity."

839. Additional guidance is to be found in the judgment of Hobhouse LJ (although not referred to by the Court of Appeal in the Zambia case). At paragraph 25 (on page 486) he said this, by reference to dicta in earlier cases about privity:

"All these formulations reject the suggestion that even gross negligence will suffice. The use of the word "suspicion" and "belief" are indicative of the strength of the suspicion that is required. But perhaps the most helpful guide is to be found in what was said by Roskill and Geoffrey Lane LJJ about the reason for refraining from inquiry – "in the hope that by his lack of inquiry he will not know for certain" – "in order to avoid obtaining certain knowledge of the truth". It is probable that Lord Denning MR was saying the same thing when he used the phrase "So that he should not know it for certain". The illuminating question therefore becomes "why did he not inquire?" If the judge is satisfied that it was because he did not want to know for certain, then a finding of privity should be made. If, on the other hand, he did not inquire because he was too lazy or he was grossly negligent or believed that there was nothing wrong, then privity has not been made out."

840. If, as is clearly implied by the Court of Appeal in the Zambia case, "privity" within the meaning of the Marine Insurance Act 1906 is to be equated with knowledge for the purposes of dishonest assistance or unlawful means conspiracy, then those dicta suggest that an absence of inquiry may be probative of blind-eye or Nelsonian knowledge only if the reason for not inquiring is that the defendant believed to be likely (rather than merely suspected) that which he did not want to know, and refrained from inquiry in order to avoid learning the unwelcome fact for certain. Equally, it suggests that a suspicion that a counterparty may, vis-à-vis a third party, be up to no good, falls short of a suspicion firmly grounded and targeted on specific facts, of the kind necessary for subsequent lack of inquiry to give rise to blind-eye or Nelsonian knowledge.
841. I was referred by counsel to the earlier decision of the Court of Appeal in Heinl v. Jyske Bank (Gibraltar) Ltd [1999] Lloyd's LR 511. That was a case about dishonest assistance rather than unlawful means conspiracy, but if the quality of knowledge required in each type of case is (as suggested in the Zambia case) broadly equivalent, it is nonetheless of some assistance. Mr Heinl had been an unsatisfactory witness at trial, and had lied persistently in the witness box. Nonetheless, in concluding that the facts did not disclose a sufficient basis for the establishment against him of dishonest assistance, Colman J (with whom the other members of the court substantially agreed) said this at page 547:

"It has to be emphasised that it is not enough that on the whole of the information available to him he ought as a reasonable man to have inferred that there was a substantial probability that the funds originated from the Bank. It must be established that he did indeed draw that inference.

Taking fully into account the very unsatisfactory nature of much of Heinl's evidence, I have with some hesitation come to the conclusion that, even allowing that at the end of the day this is essentially a so-called jury question, there is insufficient evidence to support a finding of relevant dishonesty on his part. … If third parties are to be held accountable on the basis of accessory liability for breaches of trust committed by others the standard of proof of dishonesty, although not as high as the criminal standard, should involve a high level of probability. In this case I am not satisfied that such level has been established."

The liability of an unlawful means conspirator who has not himself committed the relevant unlawful act may be primary rather than accessory, but nonetheless the tort exists only to maintain basic standards of civilised behaviour, and the imposition of liability for what is not otherwise actionable requires a high degree of blameworthiness on the part of the defendant. The decision is also a useful reminder of the need to avoid jumping from a perception that a witness is a liar to a conclusion that he must have believed in the likelihood of the unwelcome facts into which he did not inquire.

842. Issues as to the extent of the knowledge necessary to be established against a conspirator are of particular importance in a case, such as the present, where primary fraudsters (here the Baskans) have invited others to assist them in their dishonest designs on a strict "need to know" basis. In the present case, I have concluded that while the Baskans identified separately to Ferrero and to Mr Abidali different aspects of that which they planned to perpetrate upon the Banks and upon Yapi Kredi, neither Ferrero nor Mr Abidali were informed about the entirety of the fraud which the Baskans planned, or, in particular about those parts of it which caused the Banks much of their loss. Put shortly, the issue is, how much of the principal fraudster's scheme must the defendant know before incurring liability for the whole, or part, of the consequences?
843. There are, inevitably, dicta pointing both ways on this issue. For example, in Kuwait Oil, Nourse LJ cited with approval the following dictum of Harman J in Huntley v. Thornton [1957] 1 All ER 234 at 251:

"No doubt it is not necessary that all the conspirators should join at the same time, but it is, I think, necessary that they should know all the facts and entertain the same object."

On the hand, it has been said that where two or more persons combine with a view to stealing from the claimant wherever possible, it is unnecessary to show that the defendant conspirators knew of, or planned, every specific subsequent theft: see Kuwait Oil (supra) at para 133.

844. Some indirect guidance is obtained from the analysis of the Court of Appeal in IS Innovative Software Ltd v. Howes [2004] EWCA Civ 275. In that case Mr Howes was a typical bit-player in a larger conspiracy in which he became involved on a need to know basis. Having resigned as managing director of the claimant company he was approached by three of its other employees who, fearing redundancy, sought to improve the financial consequences of being dismissed by a dishonest scheme whereby they hoodwinked the company into thinking that they had been employed on terms requiring notice to be given expiring on particular dates. For that purpose they persuaded Mr Howes dishonestly to sign letters of employment and to backdate them to a time when he had been managing director of the claimant company, so as to make it appear that they had been employed upon the terms which they sought to assert. Quite separately, and without Mr Howes knowing about it, they fabricated and then backdated a statement of general terms and conditions of employment which provided the requisite termination provisions, and then used the combination of the letters and the general terms to achieve their dishonest ends. Mr Howes thought that the letters which he had dishonestly backdated stated no more nor less than the terms of employment which had actually been agreed, albeit otherwise than in writing, at the material time.
845. Both the trial judge and the Court of Appeal rejected the company's claim that Mr Howes was party to that part of the other defendants' unlawful means conspiracy which caused the company's loss. Giving the judgment of the court, Neuberger LJ said this, at paragraph 44:

"In this case, Mr Howes agreed to sign the backdated letters, thereby appearing to give each of the four employees written contractual rights backdated to 1^st November 2000, whereas in fact they had no such written contractual rights. Accordingly, as the judge found, he was party to an unlawful means conspiracy. We would also accept that it could fairly be said that this conspiracy, and in particular the signing of the backdated letters was "directed to" the company, in the sense that the purpose of executing the letters in a backdated form was to induce the company to believe that they had been signed by Mr Howes at a time when he had power to bind the company."

Later, at paragraphs 50 and 51, he said this:

"50. Accordingly, in the light of the facts of this case we think that the judge was entitled, indeed correct, to conclude that Mr Howes should not be liable for the damages suffered by the company as a result of the unlawful means conspiracy, because the nature of the damage suffered arose from a provision not contained in the backdated letter wrongly executed by Mr Howes, but from a document which he did not know was going to be brought into existence, whose terms he did not know, and the existence of whose particular term (which actually caused the damage to the company) he had no reason to know of or even suspect.

51. The essential point is that he honestly believed that the terms of employment which were described in the backdated letters were to be the terms of employment which each of the four employees already had. That belief was not even mistaken. In other words, although Mr Howes was dishonestly taking part in an arrangement involving a backdating of the letters so that they would apparently be binding on the company because it would have been signed by him at a time when he was still managing director, he rightly and reasonably believed that those written terms were terms which were binding on the company in any event. Insofar as the further terms and different terms were included in the backdated Main Terms, Mr Howes knew nothing of them and no reason to know of them."

846. Mr Wardell submitted that this case was no more than an example of the principle that, without knowledge that the unlawful means in question would cause any loss to the claimant, and therefore without intent to injure, a defendant could not be made liable for unlawful means conspiracy. While the case can be analysed in that way, Neuberger LJ's analysis also suggests that where a bit-player in a multifaceted fraud knows only of one aspect of the fraud, and is ignorant of the others, he may not be liable for anything more than the loss properly attributable to that part of the fraud of which he is aware.
847. In my judgment there is no simple doctrinaire answer to the conundrum presented by such a case. The answer lies in a painstaking analysis of the extent to which the particular defendant shared a common objective with the primary fraudsters, and the extent to which the achievement of that objective was to the particular defendant's knowledge to be achieved by unlawful means intended to injure the claimant. Although neither I nor counsel have found an authority which specifically addresses this particular difficulty, the solution to it which I have described is reflected in the following passages in Nourse LJ's judgment in Kuwait Oil, at paragraph 111 (on pages 312-3):

"A further feature of the tort of conspiracy, which is also found in criminal conspiracies, is that, as the judge pointed out (at p.124), it is not necessary to show that there is anything in the nature of an express agreement, whether formal or informal. It is sufficient if two or more persons combine with a common intention, or, in other words, that they deliberately combine, albeit tacitly, to achieve a common end."

"Thus it is not necessary for the conspirators all to join the conspiracy at the same time, but we agree with the judge that the parties to it must be sufficiently aware of the surrounding circumstances and share the same object for it properly to be said that they were acting in concert at the time of the acts complained of."

848. I therefore approach the analysis of the facts, for the purpose of deciding whether the Banks have established a claim in unlawful means conspiracy against either Ferrero or Mr Abidali by asking, separately in relation to each of them, the following questions:

i) What objective did Ferrero or Mr Abidali share with the Baskans?

ii) Did the attainment of that objective, to their knowledge, involve the use of unlawful means?

iii) Did the use of such unlawful means for the attainment of that objective necessarily (in the sense of being the obverse of the coin) involve the infliction of injury upon the Banks?

iv) To what extent if at all was the Banks' loss attributable to an injury necessarily (in the same sense) caused by the attainment of that common objective by the use of those unlawful means?

The Conspiracy Claim Against Ferrero

849. The Banks' identification of the shared objective of the alleged conspiracy, as between Ferrero and the Baskans, went through a number of evolutionary stages during the preparation and the trial of these proceedings. At the end of his closing speech, Mr Wardell summarised the Banks' case as to the shared objective under two formulations, the wider of which embraced both Ferrero and Mr Abidali equally, and the narrower of which was directed, in response to my inquiry, at identifying the common objective shared between the Baskans and Ferrero.
850. In its wider formulation, the shared objective was defined as follows (day 73 page 13):

"The object of this one conspiracy was to procure monies from the Banks by deception with the intention that those monies would provide the seed corn of a front company which would continue Baskan Gida's hazelnut business under a new name… but under the same beneficial ownership, free of any liability to the Banks."

851. That objective related to the conspiracy primarily pleaded, said to have been undertaken in or before November 2001, and to have included as part of its implementation all the alleged misrepresentations (which I shall later describe) arising out of the making and confirmation of the Framework Contract, the placing of the Forward Contracts, the signing of re-issued and re-dated Baskan Gida contracts, the formulation of the December Acknowledgements and Trade Reference, the February re-confirmation of the Framework Contract and the transfer of all outstanding orders both to Aksu Gida and, later, to Baskan Yuksel. By describing the common objective as being to procure monies from the Banks by deception, it is framed so as to encompass all aspects of the Baskan's fraud, including the invoicing fraud, the trust receipt fraud and the fraudulent trading, in addition to the setting up of and trading with a front company.
852. Mr Wardell's narrower formulation, which addressed the risk that I might conclude that no relevant combination between Ferrero and the Baskans occurred before their meeting on 25^th January 2002, was phrased as follows (day 73 page 28 and 29):

"The common objective is effectively to save the Baskans' hazelnut business by allowing it to go on trading through a front company and thereby defeat the claims of creditors [on the basis that] the new front company will fulfil Ferrero's existing orders."

853. I shall address the Banks' wider formulation first. It was overwhelmingly the primary basis upon which the Banks both pleaded and argued their claim.
854. At the beginning of his oral closing submissions, Mr Wardell helpfully provided a list of more than 35 alleged overt acts by Ferrero which, if considered together, he submitted amply justified an inference that Ferrero had combined with the Baskans for the attainment of the objective identified by his wider formulation. As he rightly submitted, the court may have regard not merely to conduct of an alleged conspirator at the material time, but also to conduct thereafter, and in particular to conduct amounting to a cover-up, as being, at least, potentially probative of the alleged conspiracy.
855. Before dealing with the detail of Mr Wardell's lengthy list, I make it clear at the outset that I have not been persuaded by my review of those alleged overt acts, both separately and in the aggregate, that the evidence comes anywhere near justifying the inference that Ferrero combined with the Baskans in the obtaining of monies from the Banks by deception, whether with the intention that those monies would provide the seed corn of a front company under the Baskans' beneficial ownership or otherwise. On the contrary, and notwithstanding the shortcomings in the quality and even the integrity of the evidence of the principal Ferrero witnesses, I am persuaded on the evidence taken as a whole that at no time did Ferrero know (in the sense of having either actual or Nelsonian knowledge) before the end of February 2002 that the Baskans were or had been procuring monies from the Banks by deception, although I consider that during the period between 25^th January and the end of February 2002, the responsible employees of Ferrero, and in particular Mr Casale and Mr Rosa Brunet, probably began to suspect that, in some undefined way, the Baskans may have been up to no good vis-à-vis the Banks.
856. Before descending into the detailed analysis of the overt acts relied upon, I wish to emphasise at the outset that my reasons for rejecting this broad formulation of the Banks' conspiracy case against Ferrero is significantly (but not exclusively) based upon the following general considerations. The first arises from the need to appraise events in their historical context, rather than with the benefit of hindsight, in particular when addressing supposed inferences as to the state of mind of defendants alleged to have participated in a dishonest conspiracy.
857. The historical context in which the events relied upon by the Banks need to be set is that they occurred at the end of a 10 year commercial relationship between Ferrero and the Baskans throughout which (and this is not challenged) the Baskans had conducted their part in the relationship with complete integrity and trustworthiness. The aphorism that businessmen are in principle entitled to trust their counterparties may have uncertain force as between parties at the outset of a business relationship. It is of powerful force when that trust has been earned by mutual dealings over a very long period. To that may be added the fact (also unchallenged) that, quite apart from their relationship with Ferrero, the Baskans had a general reputation for integrity and trustworthiness in the conduct of what was, by 2001, a very large enterprise (by Turkish standards) which had been carried on as a family business for many years. It is therefore right to assume that Ferrero's starting point in addressing the question whether anything said, done or omitted by the Baskans during the relevant period suggested that they were up to no good, was that the Baskans had earned an entitlement to be given the firm benefit of any doubt.
858. The second general point relates to motive, as distinct from intent. Although motive is not, in that sense, a necessary element to be proved for the purposes of establishing a claim in conspiracy to injure by unlawful means, it is a vital ingredient in any rational assessment of the question whether a defendant, and Ferrero in this instance, acted dishonestly, as the broader formulation of the Banks' claim necessarily assumes. Furthermore, as Mr Strauss succinctly put it in his closing submissions, motivation in relation to dishonest conduct necessarily requires an assessment not merely of the incentives, but also of the disincentives to embarking upon the course of conduct alleged. The supposed incentives need to be balanced against the disincentives, which in the case of participation in a fraud include not merely a disinclination to do that which is immoral, but also the risk of incurring long-term damage to the commercial reputation of a business enterprise, and the personal risks to the careers and even liberty of the individuals involved, if found out and brought to justice.
859. Looking first at the incentives, it emerged with complete clarity during the trial that Ferrero had, as its relevant senior employees recognised, a powerful commercial motive for assisting the Baskans in the enhancement and preservation from perceived threats of their hazelnut export business, namely to preserve not merely a supplier with a track record of producing the highest quality hazelnuts, but also the supply chains behind that supplier, upon whom the Baskans depended, and which might be dissipated or otherwise lost if the Baskans were put out of business. An additional subordinate motive for supporting the Baskans' preservation of their business was the avoidance of being exposed to a single large supplier of Turkish hazelnuts, in the form of Oltan Gida, at a time when no other competing supplier appeared to offer Ferrero a wholly satisfactory service.
860. Recognising perhaps that such a motive would not easily be regarded as sufficient to justify Ferrero participating in a fraud on two well-resourced banks with offices within the EU, the Banks tried hard to identify further motives behind Ferrero's alleged participation in the conspiracy, and also to minimise the obvious disincentives by the submission that Ferrero believed that there was no risk of being found out. The main alternative motives consisted of the obtaining of cheap nuts, and a possible but unidentified private corrupt inducement to Mr Casale and Mr Rosa Brunet (the case against Mr Bolowich having been abandoned at the conclusion of his cross-examination) to ensnare an otherwise innocent Ferrero in the Baskans' scheme.
861. In my judgment, the "cheap nuts" motive fails, both on the facts, and as a matter of analysis. While of course Ferrero like any purchaser had the ordinary commercial motive to obtain its raw materials at the lowest price reasonably obtainable in the market, I am not persuaded that the particular prices which Ferrero obtained under the December 2001 and February 2002 Forward Contracts (being the main contracts relied upon for this purpose) were so low as to amount, on any analysis, to a significant motive for participation in a fraud. Furthermore, if viewed against a common perception by both the Baskans and Ferrero that spot prices in the Turkish hazelnut market were likely to fall by the contracted time for delivery, the proposition that the nuts obtained under those contracts were in any real sense "cheap" largely falls away. For different reasons, the criticism of the prices agreed under the late November re-grade contracts (namely that they were by reference to historic rather than current market prices) is also, for reasons which I have already given, misplaced.
862. As for the supposed private corrupt motive on the part of Mr Casale and Mr Rosa Brunet, the Banks have, quite simply, failed to unearth any evidence for this at all, despite strenuous efforts, in particular by a persistent examination of certain payments to a company connected with the Dahods' and Mr Abidali's religious community called Safe Holdings Ltd which, in the event, went nowhere. While there may be cases in which, even in the absence of any direct evidence, a private corrupt motive on the part of company employees may sometimes be inferred as the only explanation for what is otherwise a clearly proved fraud participated in by them, this is by no means one of those cases. On the contrary, whatever the shortcomings in the honesty of parts of the evidence of Mr Casale and Mr Rosa Brunet, their absolute loyalty to Ferrero was fully demonstrated by the evidence, taken as a whole. Indeed, I consider that it was that sense of loyalty which led them to commit themselves to Ferrero's untrue case about its lack of understanding that Aksu Gida and Baskan Yuksel were front companies for the Baskans, and to the protestation of an unrealistic level of ignorance about the meaning and implications of the word assignment, where used repeatedly in the relevant documentation which Ferrero received and (as I have concluded) carefully considered.
863. There remains Mr Wardell's point about a perception within Ferrero of a low risk of being found out if the Baskans' fraud on the Banks subsequently came to light. As to that, the level of complicity in the Baskans' fraud alleged against Ferrero under the Banks' wider formulation of its case makes it virtually inconceivable that, if Ferrero believed that the Baskans were planning and implementing a fraudulent scheme to obtain the Banks' money, that it would be otherwise than a mere matter of time before the fraud itself was found out.
864. The more difficult question is whether there is any substance in the submission that Ferrero could have thought that its own participation in that fraud would go undetected. That issue is closely bound up with the implications of the Ferrero witnesses' readiness to lie in the witness box in support of a false case by Ferrero designed to distance itself from the Baskans, and with the Banks allegations about destruction of relevant documents which I deal with elsewhere.
865. My conclusion is that while it is theoretically possible that, if Ferrero had been an active participant in the Baskans' fraud on the Banks in the manner alleged, it might have regarded the risk of being found out as low, nonetheless even a low risk of an entirely unacceptable consequence would remain a powerful disincentive to the participation alleged.
866. In my judgment the general analysis of the issue of motivation, as outlined above, leads clearly to the conclusion that the alleged participation either of Ferrero as an entity or of its employees Mr Casale and/or Mr Rosa Brunet in the Banks' broader formulation of the conspiracy is unsupported by any evidence of a motive which would afford a rational explanation for such grave and potentially self-destructive conduct. The absence of any such qualifying motive is therefore a substantial factor in the balance against an inference of participation by Ferrero in that conspiracy, quite apart from the factors which emerged from the detailed consideration of the overt acts relied upon, to which I now turn.
867. For the most part, my reasoning for concluding that, viewed separately, each of the overt acts relied upon failed to justify an inference of participation by Ferrero in the conspiracy will sufficiently appear from my findings of fact in relation to those events. In such cases, I will need to do no more than summarise that reasoning. A small number of others call for more detailed treatment.
868. The first group of overt acts relied upon relate to the period between 15^th October and 13^th December 2001, and consist of the making of the Framework Contract, Ferrero's involvement in the drafting of the Ferrero Industrial Acknowledgements of Assignment of 6^th and 13^th December and the different drafting of the Acknowledgements to be signed by the operating companies, the making of the December Forward Contracts and the signing of the December Letter. The gist of the Banks' case is that these acts display Ferrero deliberately concealing from the Banks what Mr Wardell described as the breakdown in the trading relationship between Ferrero and Baskan Gida during the same period, the deliberate provision of a falsely optimistic trade reference designed to enable the Baskans to obtain the facility by deception, the drafting of the Acknowledgements of Assignment in such a way as to enable the Baskans to commit the invoicing fraud without giving rise to any liability on the part of Ferrero, and the making of the December Forward Contracts on terms as to price sufficient to demonstrate to Ferrero the Baskans' dishonesty, and to make it clear that those contracts were to be used purely for the raising of money from the Banks, rather than for any genuine commercial purpose.
869. It will be apparent from my findings of fact that none of those events, taken singly or together, justify or even support an inference of dishonest participation by Ferrero in a conspiracy to defraud the Banks. Looked at in the round, the single most important event is constituted by the making of the three December Forward Contracts. Together they constituted an order in an aggregate amount exceeding 6,000mt, for delivery during periods ending between March and June in the following year. Although they were priced below the then spot price, and represented a substantial departure from the ordinary pattern of trading between Ferrero and Baskan Gida, I have concluded after detailed analysis that they were genuine commercial contracts negotiated to meet Ferrero's real needs (both in terms of production and the building up of a buffer stock), at a negotiated price which reflected both parties' broadly congruent perceptions as to the likely development of the market during the delivery period.
870. Since the aggregate tonnage ordered represented a very substantial part of Ferrero's outstanding requirements in relation to the 2001 season, the placing of those orders with Baskan Gida of itself demonstrates Ferrero's continued confidence in Baskan Gida as their principal (or by then joint principal) supplier. It shows both that the relationship between Ferrero and the Baskans had not broken down and that Ferrero expected the relationship to continue well into 2002, during the period agreed for delivery of the hazelnuts ordered under those contracts.
871. That analysis of the making of the December Forward Contracts necessarily deprives the other overt acts relied upon during that period of any force as material from which Ferrero's participation in the alleged conspiracy could possibly be inferred. There was, for separate reasons already given, nothing sinister in Ferrero's signature of the Framework Contract, nor in its involvement of the drafting of the Acknowledgements of Assignment. Although the 13^th December Trade Reference contained a negligent misrepresentation, I have concluded that it did not involve dishonesty by Ferrero or by Mr Casale personally. In my judgment neither he nor anyone else at Ferrero knew, believed or even suspected at that stage that Baskan Gida intended any dishonest or deceptive conduct as against the Banks, still less that it was planning to obtain money from the Banks by deception.
872. The second group of alleged overt acts consist of steps taken by Ferrero from becoming aware of the grant of the Facility during the remainder of 2001 and up until, but not including, attending the meeting on 25^th January 2002. The events consist of Ferrero's countersigning re-numbered Baskan Gida contracts on 19^th December, Mr Casale and Mr Rosa Brunet's lack of response to Mr Bolowich's 3rd January email, and the signing of Baskan Gida contracts 5088 and 5089 with altered delivery dates in January 2002. The Banks rely in particular on the allegation that Mr Casale forged Ferrero France's signature on contract 5089.
873. Again, I have dealt in detail with each of these events in my chronological findings of fact. While the suspicious construction of Ferrero's conduct which the Banks placed on these events is understandable, and while I have been obliged to treat with reserve Mr Casale and Mr Rosa Brunet's evidence about them, due to their clearly demonstrated unreliability as witnesses in relation to other matters, I have concluded on balance, and in particular when these events are set in their historical context, that they do not disclose dishonest assistance by Ferrero in the Baskans' scheme to defraud the Banks, or overt acts from which a concurrent or prior combination by Ferrero with the Baskans in such a scheme can properly be inferred.
874. Furthermore, the single contemporary documentary reference to the 25^th January meeting between Ferrero and Melih Baskan immediately thereafter, namely Mr Abidali's email to Mr Dahod on 8^th February, unmistakably portrays Ferrero as, at least prior to that meeting, an outsider to the Scheme then being concocted, rather than an existing party to a conspiracy.
875. There follow a pair of overt acts, which stand together as a distinct group, consisting of Ferrero's attendance at the 25^th January meeting with Melih Baskan, followed by the decision shortly thereafter to cooperate with the Baskans' request to transfer outstanding orders to Aksu Gida, if later asked to do so.
876. I have dealt with these two events in great detail in my findings of fact, and concluded that, although they demonstrated to Mr Casale and Mr Rosa Brunet that the Baskan family were, if the scheme proceeded, going to remain in de facto control of Aksu Gida while pretending the contrary to Yapi Kredi, the manner in which the Baskans presented the scheme to them did not cause them either to suspect, still less believe or intend, that the Scheme with which they then decided to cooperate involved any deception of the Banks, let alone the obtaining of any money from the Banks by deception. Thus, those events do not support an inference that, while agreeing to participate in the Scheme for the setting up of what Ferrero probably did understand was to be a front company, Ferrero combined with the Baskans in any plan to obtain money from the Banks by deception, to serve as the seed corn for the business to be conducted by the front company.
877. As a prelude to the events which followed however, the 25^th January meeting did serve to call into question Ferrero's assumption that the Baskans' trustworthiness and integrity in their business dealings could be regarded as beyond doubt, since it revealed a readiness on their part to deceive Yapi Kredi, in pretending that the proposed front company was not beneficially owned or controlled by them.
878. The next group of overt acts focus on the period between Ferrero's decision to cooperate, if requested, with the Baskans request to transfer outstanding contracts to Aksu Gida, and the Banks' discovery of the fraud. They consist of the making of the 28^th January contract renewals (contracts 5100 and 5101), the making of the Early February Contracts, the signing of the amendment to the Framework Contract and the simultaneous transfer of outstanding orders to Aksu Gida, without informing the Banks of the latter, Mr Bolowich's and Mr Casale's continuing contacts with Melih Baskan and the approval, without communication to the Banks, of the split invoices. I have addressed each of these events in detail earlier in this judgment, so that little needs to be added by way of analysis at this stage.
879. The most important item in this group of events is the making of the February Forward Contracts. The Banks' case was that Ferrero had no need for a further 6,000mt of nuts at this late stage in the season, had no reason in any event to buy them all from Baskan Gida rather than, for example, the more efficient Oltan Gida, that the price payable was so far off the spot market price as to be inherently uncommercial and therefore suspicious, so that it should be inferred that the sole purpose of the making of them was, to Ferrero's knowledge, to put into the Baskans' hands documents enabling them to draw down very large sums of money from the Banks which they had no intention that Baskan Gida, or anyone else, would ever repay.
880. This is a formidable case, all the more so because by February 2002 Ferrero could no longer attribute the qualities of unquestioning integrity and trustworthiness to the Baskans, having regard to their apparent readiness to mislead Yapi Kredi about the real ownership and control of Aksu Gida, should their plan to transfer the business to that company proceed. For reasons already given, that case misses its target, but only by a small margin. I have concluded that Mr Casale and/or Mr Rosa Brunet probably did ask themselves how (if Baskan Gida raised monies on those contracts) it would repay the Banks if the transfer of the underlying orders to Aksu Gida went ahead, but did not pursue inquiries on the basis that it was none of Ferrero's business.
881. In my judgment however, the state of mind of Mr Casale and Mr Rosa Brunet in relation to the Early February Contracts fell short of blind-eye or Nelsonian knowledge of the Baskans' intended fraudulent use of them, so that Ferrero's participating in them fails to justify an inference that it had by then combined with the Baskans with a shared objective of enabling them to obtain money by deception from the Banks, whether as the seed corn for Aksu Gida's business or otherwise.
882. My main reason for that conclusion is that, contrary to the central thrust of the Banks' case about them, the February Forward Contracts did have, and were perceived by Ferrero to have, commercial purposes for both parties. For the Baskans they secured a virtual monopoly of the supply of Ferrero's further requirements for the 2001 season, to the exclusion of their major competitors, and on terms that would enable those contracts to be transferred so as to constitute Aksu Gida's initial order book if the Scheme went ahead. They did so at a price which, however unfavourable in terms of the then spot price, was one which both the Baskans and Ferrero thought reasonable having regard to their shared perception that the market would thereafter fall during the delivery period (as indeed it did). For Ferrero the contracts secured their requirements (both for production and buffer stocks) from suppliers and underlying supply chains of unquestionable quality, coupled with an assurance that the Baskans would, if necessary, honour them through Aksu Gida if the transfer went ahead. Ferrero had its own commercial interest in providing a substantial order book for Aksu Gida, so as to ensure that it would have sufficient business as a purchaser within Turkey to preserve the Baskans' supply chains, and the price offered was, again, favourable but not unrealistically cheap, having regard to Ferrero's perception that the market was likely thereafter to fall.
883. Since those contracts had a commercial purpose for both parties, there was in my judgment no reason for Ferrero to suppose that they would be used simply to raise money from the Banks which the Baskans did not intend should be repaid. Ferrero had no way of knowing at what time before the intended supply of any particular consignment of nuts the Baskans needed to raise working capital, or whether therefore the February Forward Contracts would be used to raise money by Baskan Gida prior to the transfer, or by arrangement with the Banks, by Aksu Gida after the transfer. Ferrero simply did not know what the Baskans proposed to do by way of funding the working capital requirements of Aksu Gida, whether with the Banks or otherwise, and, in sharp contrast with the position of Mr Abidali, those matters were none of Ferrero's business.
884. The essence of Nelsonian knowledge (as is vividly depicted by Manifest Shipping, and by the episode at the Battle of Copenhagen from which it takes its name) is that a failure to inquire about matters giving rise to suspicion qualifies as the basis for an inference of knowledge only if the reason for the failure to inquire is that the defendant fears that he will learn for certain that which he already considers likely. In my judgment, no-one at Ferrero reached that level of suspicion as to the fraudulent use to which (in fact) the Baskans proposed to put part of the value of one of the February Contracts, and their absence of inquiry about it is to be attributed not to a fear that they would discover an intended fraud, but rather to a belief that it was none of Ferrero's business.
885. The next large group of alleged overt acts consists of those aspects of Ferrero's conduct following the Banks' discovery of the fraud which they criticised as characteristic of persons anxious to conceal their wrongdoing, coupled with Ferrero's simultaneous participation in trading with Aksu Gida, acquiring nuts from Aksu Gida which the Banks claim belonged to them, and cooperating in the further sham business transfer from Aksu Gida to Baskan Yuksel. Again, I have already reviewed this part of Ferrero's conduct in detail. I have concluded that Ferrero knew throughout that the Baskans were in de facto control both of Aksu Gida and of Baskan Yuksel, so that they were knowingly trading with the Baskans throughout that period. I have also concluded that, for most of that period, Ferrero's suspicion about the integrity the Baskans' conduct had risen to a level where it believed it probable that they had in some way been defrauding the Banks. In both those respects I have rejected Ferrero's case, and the evidence of its witnesses, to the contrary.
886. Nonetheless I have also concluded, contrary to the Banks' case, that Ferrero had sufficient commercial reasons for continuing to trade with the Baskans during the rest of 2002, notwithstanding Mr Gun's advice to the contrary, and that its uncooperative attitude towards the Banks when asked for information about its dealings with the Baskans' companies was also sufficiently attributable to motivation other than a desire to cover up prior participation in the Baskans' fraud, that it would be wrong to treat that conduct, however objectively unsatisfactory, as justifying an inference of participation in a conspiracy of the broader type alleged.
887. The final group of overt acts relied upon by the Banks consist of aspects of Ferrero's conduct of litigation about the matters in issue, both in the criminal complaint brought in Italy, and later in these proceedings.
888. The Italian criminal complaint was, as was evident from the unsuccessful attempt of Ferrero's witnesses to distance themselves from any responsibility for it, a singularly unattractive and, in the event, unsuccessful application of the principle that, sometimes, attack is the best form of defence. It does not however follow that the aggressive defence constituted by the criminal complaint was designed to fend off a counterattack alleging participation in a fraudulent conspiracy, still less does it justify an inference that Ferrero thereby demonstrated a sense of vulnerability to such a claim. The central focus of the criminal complaint was the alleged use by the Banks of the forged Acknowledgements of Assignment. In my judgment the criminal complaint was specifically aimed at putting out of action before it could even be launched a feared claim by the Banks that Ferrero was liable pursuant to the Assignments to pay the Banks in relation to hazelnuts for which it had already paid the Baskans: i.e. the Banks' contractual claim. That was the claim made in the Banks' letters before action which immediately preceded those proceedings.
889. Much more substantial in my judgment are the Banks' criticisms of Ferrero's conduct, and that of its witnesses, in these proceedings. Putting forward a false case and then supporting it by lying in the witness box must always give rise to serious questions about that which it is sought thereby to conceal. So, if it were true, would deliberate destruction of relevant documents, or the deliberate withholding of them in breach of a disclosure obligation. This is a serious allegation, to the details of which I must now turn.
890. The Banks have, both in interim applications before the trial, and on numerous occasions during it, maintained a strenuous claim that Ferrero has been guilty of one or both of those forms of misconduct in relation to documents. Indeed, the time and effort applied to the prosecution and defence of that claim would itself have been sufficient to constitute a substantial piece of litigation in its own right. Nor was it so lacking in substance that it would be fair or appropriate for me to deal with it summarily, or by recourse to the traditional rules whereby the affidavit of a solicitor that his client has no further documents to disclose is taken as being conclusive.
891. Before getting into the detail, three general points deserve emphasis. The first is that the yardstick by reference to which misconduct in relation to disclosure of documents is to be measured needs to take account of the fact that, in the three European countries where the Ferrero Defendants are incorporated and carry on their business, disclosure of documents plays an altogether less prominent role in civil litigation than it does in England or, for that matter, in the USA. It is therefore reasonable to suppose that, whereas from the moment when an English company appreciates that matters may become the subject of litigation, steps are expected to be taken to ensure the preservation in safe custody of relevant documents, the same perception may not apply, at least with the same force, in Italy, France or Germany. Mr Wardell for the Banks acknowledged that this was a relevant consideration, at least in relation to Italy, where most of the relevant documents might have been expected to be found.
892. The second general point is that the purpose of my analysis of this question is, at least at this stage in these proceedings, which are not concerned with orders for costs, strictly limited. The question is whether misconduct by or within Ferrero in relation to the disclosure or destruction of documents supports an inference that attempts were being made to cover up participation in the alleged conspiracy. It follows that the analysis is properly to be limited to allegations of the deliberate withholding or destruction of documents known to be relevant, rather than to routine destruction, mindless as to the consequences, or delays or failures in producing documents attributable to carelessness, indolence or a lack of an appropriate sense of the diligence required.
893. The third general point is that in any case of alleged deliberate destruction or withholding of relevant documents, it is necessary to identify the real culprit, and in particular whether the alleged misconduct is that of Ferrero as an institution or of a particular employee, and in each case, whether the likely motive is the concealment of damaging material from the Banks, and therefore from this court, or the concealment by an employee of material from Ferrero, on the basis that it might lead to damage to his career.
894. The Banks' written closing submissions (in which 20 pages were devoted to this issue) jumbled together criticisms of all the various types which I have just described, and the more precise focus called for by the limited task which I have identified only emerged during Mr Wardell's closing speech, to a significant extent, in response to my specific inquiries. That said, I shall follow the categorisation of the Banks' allegations in relation to disclosure set out in paragraph 54 of their written closing submissions.
895. The first concerns Ferrero Italy's and Ferrero Industrial's originals or copies of nine of what the Banks describe as the forty key contractual documents in the case, including in particular the ten Baskan Gida contracts which, by countersigning them, Ferrero is alleged to have assisted the Baskans to use for the obtaining of money by deception. Apart from those contracts, the missing key documents also included Ferrero Industrial's copy of the KBC faxes dated 21^st December 2001, and Ferrero Italy's original letters from KBC of 24^th April and 1^st May 2002, making claims in respect of unpaid invoices. The allegation was that in the absence of any satisfactory explanation for these documents being missing from Ferrero Italy or Ferrero Industrial's files, the obvious inference is that they have been destroyed.
896. The idea that these documents might have been destroyed by Ferrero to keep them from coming to the attention of the Banks seemed fanciful, since by their very nature, the Banks would have been assumed by Ferrero to have been likely to have their own copies for use in any proceedings, as indeed they did. It emerged from Mr Wardell's closing speech that the allegation of destruction in relation to this category is destruction by Mr Casale, to keep the documents from coming to the attention of his employers, rather than institutional destruction by Ferrero. The suggestion was that Mr Casale destroyed them fearing that their examination by Ferrero during its investigation in 2002, after the discovery of the Baskans' fraud, would lead to him being identified as a co-conspirator with the Baskans.
897. I regard that allegation as no more persuasive than a suggestion that these documents had been destroyed by Ferrero. In relation to all of them, Mr Casale would have appreciated, as a highly intelligent man, that destruction of the only copies of those documents to which he had access, namely those in Ferrero Industrial and Ferrero Italy's files would at best merely postpone the date upon which, if as damaging as the Banks suggest, they would in due course have come to the attention of his employers. The countersigned contracts were, of course, all held in copy form by the Banks, as he would have appreciated if, as alleged, he had signed them only to enable the Baskans to draw down money on them. The faxes from the Banks of December 2001, April and May 2002 were part of sets also sent to the other Ferrero Operating Defendants (and thereby out of Mr Casale's control) and in any event retained in copy form by the Banks.
898. Furthermore, this allegation comes nowhere near displacing my clear impression that, although prepared to lie from time to time, Mr Casale remained wholly loyal to his employers throughout. Furthermore, as Mr Strauss observed, there are numerous examples of missing copy contracts in Ferrero's files for the relevant period. In fact, more went missing in relation to the 2000 year than in relation to the year in question. It is also destructive of any notion that there existed a privately motivated conspiracy between Mr Casale and the Baskans, unknown to Ferrero that, upon request later in 2002, the Baskans sent copies of a number of the relevant contracts to Ferrero, upon its inquiry, which were in due course disclosed in these proceedings.
899. The second category of which complaint was made was Ferrero's email archives for the period 2001/02, a failure which was described in the Banks' written closing submissions as "incredible", the apparent implication being that this was being deliberately done to avoid disclosure in these proceedings, or that Ferrero's evidence that email archives were routinely destroyed to free up spare computer capacity was simply untrue. In this case Mr Wardell eventually confirmed in his closing speech that the allegation here was one of negligence rather than the deliberate destruction of documents perceived to be relevant and harmful. Accordingly, while that routine destruction was, to an English eye, both surprising and unhelpful, it contributes nothing to the inference of participation in any conspiracy.
900. I shall pass over the next category, namely missing cover sheets for faxes sent by Baskan Gida, since it was not developed with any specificity, nor was an allegation that these were deliberately destroyed out of a perception that they were damaging, seriously pursued.
901. The next two categories are of real importance, namely documents which might be expected to have been created in relation to relevant management issues within Ferrero at the material time, and general management documents and board minutes. In relation to these two categories, Mr Wardell submitted that there had been a deliberate institutional withholding of these documents by Ferrero telling their lawyers falsely that they had no such relevant documents in their possession.
902. I have not found this allegation easy to determine. It is in my judgment striking that the events with which these proceedings are concerned, and in particular the investigation of the Banks' claims in and after April 2002, appear to have generated such a paucity of internal management records within Ferrero. Mr Wardell was also able to point to certain categories of documents, relating to production targets and information about quality, where documents survive from the 2000 period but are absent during the relevant period, a trend for which Mr Do's description of his hands-on face-to-face 'talk about rather than write about it' management style provided only a partially satisfactory explanation. Also in favour of the Banks' case under this heading is the fact that in relation to certain material issues, Ferrero has advanced a false case and supported it by lying witnesses. Litigants who behave in that way can hardly complain if the court approaches an allegation of destruction of relevant documents with something less than the usual disinclination to believe such a serious allegation without convincing proof.
903. Against a finding of deliberate destruction or withholding of documents of these types are the following factors. First, there is, as Mr Wardell acknowledged, the likelihood inherent in the Ferrero Defendants' places of business that an altogether less rigorous attitude to the preservation of documents for litigation purposes than applies in England may have been a large, but not sinister, contributor to the surprising paucity of documents of this type. Secondly, the Ferrero Defendants have in other respects displayed a conscientious approach to the disclosure of many documents, including for example documents which for the purposes of their inquiries they obtained from the Baskans, and then disclosed. There is not demonstrated any uniform picture probative of deliberate destruction or withholding, but rather one which appears possible only in patches.
904. Above all, there is, quite simply, no convincing proof of this allegation, but rather a reasonably grounded suspicion that destruction or withholding may have occurred. While convincing proof in the form of an admission to that effect is extremely rare, it is less uncommon to find proof in the form of other documents which by cross-reference point to documents which have not been produced, in circumstances where concealment or destruction is more likely than mere accidental loss. No such paper trail exists in relation to the categories of which complaint is here made. In certain instances the Banks attempted to suggest otherwise, in particular in relation to a supposed written request by Melih Baskan for Ferrero to countersign contracts 5094 to 5097 and 5100 to 5101. But, as it seems to me, those requests could equally have been made orally, without giving rise to any document, or made in documents which, because of the routine destruction of emails, have disappeared otherwise than because of an intent to destroy them to conceal relevant evidence.
905. As for board minutes, Ferrero's answer has not been that there are no such documents, but that if and to the extent the matters in issue in these proceedings were discussed at board level within Ferrero the minutes do not record any relevant record of those discussions. While again, this might appear surprising in relation to a large English group of companies, I do not consider that it would be right to draw an inference that relevant minutes have either been destroyed or withheld, in circumstances where other companies may organise their affairs so that management discussions about sensitive matters take place outside formal board meetings, or within them, but on the basis that the discussions are not then minuted, precisely for the preservation of their confidentiality.
906. My conclusion is that there is on balance insufficient material from which I can properly conclude that there has been deliberate destruction or withholding of relevant material by Ferrero in relation to these two categories, despite the existence of reasonable grounds for a suspicion to that effect. Even if there had been such destruction or withholding, the question would nonetheless remain whether that justified an inference that Ferrero was seeking to cover up documents probative of its participation in the conspiracy alleged rather than, as I consider more likely, documents adverse to Ferrero's case that, for example, there was an insufficient understanding of the meaning of the word assignment in the December 2001 documentation, or that Ferrero did not appreciate that the Baskans beneficially owned and controlled Aksu Gida and Baskan Yuksel. In other words, even if I had concluded that destruction or withholding of relevant documents was probable, it would not have served to do more than reinforce my conclusions as to the reasons why, in certain respects, Ferrero has advanced a false case, and its witnesses told deliberate untruths. As exemplified in Heinl v. Jyske Bank, not every lying witness is thereby shown to be a conspirator, nor, in my judgment, every destroyer or withholder of relevant documents.
907. I can pass quickly over the remaining three categories of complaint about disclosure summarised in the Banks' written closing submissions. The sixth was concerned with inadequate reasons for redacting documents, and a delay in disclosing a letter from the Baskans to Michele Ferrero. In both cases, those shortcomings reflected misunderstandings in the nature of Ferrero's disclosure obligations, and give rise to no sinister inferences. The final two categories concern delays in Ferrero's production of stock records and a shipping data-base, all of which would have been of great assistance to the forensic experts if disclosed earlier, and no doubt have saved very substantial preparation costs. Nonetheless, the reasons for their late disclosure give rise to no inferences probative of Ferrero's participation in the alleged conspiracy, so I need say no more about them at this stage.
908. I must briefly mention a repeated complaint by the Banks about Ferrero's non-disclosure of the Soremartec Standards. In a short judgment given during the course of the trial I ruled that the documents corresponding to that label were not disclosable by any of the Ferrero Defendants, not being documents in their possession or control, but rather in possession and control of their associated company Soremartec and, arguably, of the ultimate Ferrero parent company, neither of which are defendants in these proceedings. In response to my observation at the end of that judgment that, nonetheless, production of the Soremartec Standards would be of assistance to the court, those companies persisted in their previous decision not voluntarily to disclose them, in the interests of preserving confidentiality in documents containing significant business secrets. This was despite my indicating that the court had procedures whereby such documents, even if disclosed and used in evidence, can be preserved from coming into the public domain.
909. While I can understand the Banks' frustration in not obtaining sight of these undoubtedly relevant documents, it would not be appropriate for me to criticise, still less to draw adverse inferences of a serious nature from, a decision not voluntarily to produce the Soremartec Standards, any more than it would be appropriate to draw adverse inferences from a claim to privilege in respect of an otherwise disclosable document.
910. A final aspect of Ferrero's conduct of this litigation which has been relied upon by the Banks as the basis for inviting the court to draw adverse inferences has been Ferrero's not calling certain potentially relevant witnesses. They include Mr Caroly of Ferrero France, Mr Schulz of Ferrero Germany, Messrs Altare, Scavino, Bertone and Gagliasso from Ferrero Industrial, Mr Carelli of Ferrero Italy and unspecified but theoretically available members of Ferrero Italy's accounts department. Yet again, this issue provoked lengthy debate between counsel, and copious reference to authority, extending even into written reply submissions.
911. It is, again, necessary to focus upon the relevance of this alleged omission to the matter in question, namely whether it should thereby be inferred that Ferrero knew about and participated as a conspirator in the fraud on the Banks. In that context it is relevant to note the concession contained in the following final paragraph of the Banks' written submissions on this issue:

"Although it is not suggested that anyone at Ferrero other than Mr Casale and Mr Rosa Brunet knew about the fraud in late 2001/early 2002, it is submitted that Ferrero as a whole is at fault for having closed ranks around these two men, and conducted these proceedings in such a deliberately tight-lipped way."

I consider that this concession was rightly made. The participation in the conspiracy by Ferrero is not alleged by the Banks to have extended beyond Ferrero Industrial and Ferrero Italy, specifically in the persons of Mr Casale and Mr Rosa Brunet. It follows that witnesses from the other Ferrero Operating Defendants, or even from the departments within Ferrero Italy with which Mr Casale was not concerned (such as the accounts department) would be unable to offer evidence going directly to that question.

912. Nonetheless, Mr Wardell vigorously pursued arguments based on absent witnesses in relation to certain discrete aspects of the facts relevant to the conspiracy issue, such as the absence of Mr Caroly in relation to the allegation that Mr Casale forged Ferrero France's signature on Contract 5089, and the absence of Messrs Altare and Scavino in relation to the questions first whether Ferrero was dissatisfied with Baskan Gida's performance in late 2001, and secondly whether Ferrero knew that the Baskans were running Aksu Gida and Baskan Yuksel in 2002.
913. I have already addressed those discrete questions, and the inferences if any which I should draw about the absence of potentially relevant witnesses. The question at this stage is whether Ferrero's attitude towards the calling of relevant witnesses justifies any wider inference of a perceived vulnerability in relation to the conspiracy claim. In my judgment it goes no further than to support the inferences which, specifically in relation to the absence of Mr Altare and Mr Scavino, I have already drawn. For reasons which will be apparent, those fall well short of an inference that Ferrero participated in the alleged conspiracy.
914. The result of my analysis of the overt acts relied upon by the Banks is that they fail, at the first hurdle, to establish as against Ferrero the wider formulation of their alleged conspiracy. Put shortly, Ferrero did not at any time share with the Baskans the objective of procuring money from the Banks by deception, whether to provide the seed corn for a front company or otherwise.
915. I turn therefore to the narrower formulation of the Banks' conspiracy case against Ferrero in which the common objective was:

"To save the Baskans' hazelnut business by allowing it to go on trading through a front company and thereby defeat the claims of creditors [on the basis that] the new front company will fulfil Ferrero's existing orders." (Mr Wardell QC at day 73 pp. 28 to 29, in closing speech)

Perhaps because this was a very subordinate alternative to his main case, Mr Wardell did not set out in anything like the same detail the basis of overt acts from which I should infer that more limited common objective, but it was specifically directed to the risk that I should find (as I have) that nothing done by Ferrero justified the inference of any conspiracy to injure the Banks before the 25^th January meeting with Melih Baskan. It is therefore to the conduct of Ferrero on and after that date, in the historical context of events which had not previously given rise to any reason for Ferrero to suspect the Baskans of any impropriety, that it is necessary to turn.

916. For the reasons which I have already given in detail, I consider that from the moment when Ferrero decided to accede, if later requested, to an invitation to transfer outstanding orders to Aksu Gida, it shared with the Baskans the common objective to save their hazelnut business by cooperating with the Baskans transferring it to a front company, on the basis that the new front company would fulfil Ferrero's existing orders. The critical question is whether Ferrero shared with the Baskans as part of that objective the defeating of the claims of Baskan Gida's creditors, or whether, put more conventionally, the objective of saving the Baskans' hazelnut business involved, to Ferrero's knowledge, the use of unlawful means, necessarily involving the infliction of injury upon the Banks, as one of those creditors. In such an event, there would remain also the question whether the attainment of that objective by the unlawful means thereby identified, caused all or any part of the loss actually suffered by the Banks.
917. The starting point is to ascertain from whom, or from what, Ferrero perceived that the Baskans' hazelnut business needed to be saved. The story given to Ferrero by Melih Baskan on 25^th January was that it needed to be saved from the unwelcome attentions of Yapi Kredi due to a developing dispute between the Baskan family and Yapi Kredi about the biscuit business. The Baskans did not want Ferrero to know that Baskan Gida was hopelessly insolvent (although it was), and although I have concluded that Ferrero probably did not distinguish between Yapi Kredi as a creditor of Baskan Gida and as a creditor merely of the members of the Baskan family, the evidence does not suggest that Ferrero realised, at least prior to the end of February, that Baskan Gida was hopelessly insolvent, or that this was the real threat to the continuation of the Baskans' hazelnut business. Ferrero simply assumed that, as a major Turkish lender to the Baskan family businesses, probably including Baskan Gida, Yapi Kredi could in a dispute with the Baskans use its position in a way which could threaten the hazelnut business, whether by refusing to make further loans or by calling in existing loans, or by seeking to appropriate the Baskans' shareholding in Baskan Gida as a means of enforcing debts owed by the family directly, in connection with the biscuit business. I consider it unlikely that Mr Casale or Mr Rosa Brunet gave any conscious thought to the identification of the precise threat facing the hazelnut business within that range of alternatives, save to understand that the threat emanated from Yapi Kredi. Ferrero was not, and had not been since it ceased making pre-payments in 2000, a creditor of Baskan Gida, and had no reasons connected with credit risk to give any considered thought to Baskan Gida's ability to pay its debts.
918. For the same reason, although Ferrero believed by 25^th January 2002 that the Banks had become a major creditor of Baskan Gida by reason of the grant and implementation of the Facility, Ferrero had no way of knowing, nor reason to inquire, whether the Banks were to any significant extent unsecured creditors or whether, if secured creditors, they faced a shortfall if the cessation of Baskan Gida's trading caused them to have to call in their security. These were, quite simply, not matters with which businessmen like Mr Casale and Mr Rosa Brunet, accustomed to trading with Baskan Gida on terms which afforded no credit, had any reason to concern themselves.
919. Yet further, on the basis that, as I have concluded, nothing had happened prior to 25^th January to give Ferrero reason even to suspect let alone believe that Baskan Gida had entered into the Facility with the Banks with a view to obtaining money by deception, or for any other improper purpose, I consider it likely that, if Mr Casale and Mr Rosa Brunet gave the matter any thought, they would have assumed that the plan being described to them by Melih Baskan at their meeting was one which either accommodated the Banks' rights, in the sense of not running counter to the Banks' exercise of them, or one which was already or was to be a matter of discussion between the Baskans and the Banks, so as to ensure that, unlike Yapi Kredi, the Banks represented no threat to the continuation of the hazelnut business, whether or not transferred to Aksu Gida.
920. I have concluded, notwithstanding their denial, that Mr Casale and Mr Rosa Brunet probably did know that the objective of saving the hazelnut business involved an element of subterfuge and deception vis-à-vis Yapi Kredi, in creating a false picture distancing the Baskans from continuing beneficial ownership and control of the hazelnut business. If Yapi Kredi's threat to the business resided purely in its status as a creditor of Baskan Gida, it is not easy to see how that deception would fend off the threat. By contrast if Yapi Kredi's threat consisted of its ability to execute against assets of the Baskans by virtue of their personal liability to that bank, then it is easy to understand why the deception might be essential. Armed with a false cloak of invisibility as owners and controllers of the hazelnut business, the Baskans would be able to say to Yapi Kredi in any negotiations: "Sue us all you like, but it will not get you the hazelnut business". It follows that the element of unlawfulness (i.e. fraud) inherent in a deliberate deception of Yapi Kredi was not, in the way that the Scheme was presented to Ferrero by Melih Baskan, one which if implemented would necessarily injure Baskan Gida's creditors as a body, let alone the Banks. Injury to the Banks was simply not the obverse of a coin which had deception of Yapi Kredi on its face.
921. The transfer of a trading business by one company to a front company controlled by its owners may or may not defeat the claims of the creditors of the first company. It is only an inevitable consequence if the transfer is carried out at an undervalue (or for no value), or if it is done on terms which prefer the transferee as a creditor, or if once the decision to transfer has been made, so that it is known that the first company will shortly go into insolvent liquidation, it continues to incur debts knowing that they will not be repaid in full: i.e. fraudulent trading. Fraudulent trading is of course unlawful (both in England and in any civilised country) but preferential payments of debts and transfers at an undervalue are not, at least in England. Subject to its vires, a company may deal with its assets as it thinks fit, and pay its debtors in such order as it wishes to. If it then goes into insolvent liquidation, liquidators and (sometimes) creditors have statutory powers to unwind such transactions: see generally National Westminster Bank v Jones [2001] 1 BCLC 98 and Dora v. Simper [2000] 2 BCLC 561.
922. Dora v. Simper is instructive since the allegation was that, by transferring assets at an undervalue from an insolvent company to a new company managed and part-owned by its directors, the defendant directors thereby committed an unlawful means conspiracy against creditors of the insolvent company. Buckley J treated the claim as virtually untenable because, in the absence of any assertion that the directors had deceived the receivers who effected the sale, or that the receivers intended to put assets beyond the reach of the insolvent company's creditors, and it being unlikely that the directors' predominant motive was to injure the creditors (rather than line their own pockets), there was neither a conspiracy to injure, nor a conspiracy to injure by unlawful means, since the statutory remedy arising from transactions at an undervalue did not render those transactions unlawful. It is fair comment that it was conceded before Buckley J that "unlawful means" in conspiracy meant unlawful at the suit of the claimant (the Total case having not by then been decided). Nonetheless I consider that the outcome would have been the same in the light of Buckley J's conclusion, at page 566g that:

"Section 423 does not, of itself, render anything unlawful."

923. In the present case, Ferrero were told virtually nothing about the terms upon which the hazelnut business was to be transferred from Baskan Gida to Aksu Gida if negotiations with Yapi Kredi failed, save that the Baskans were to remain in control, and that the unperformed part of outstanding orders from Baskan Gida would be performed, on the same terms, by Aksu Gida. They knew nothing of the detailed and carefully prepared sham documents which were to be used to implement the transaction, nothing about the creation of a false impression of creditor pressure, and were not told (by contrast with Mr Abidali) that the objective was to pay off one of Baskan Gida's smaller creditors in kind, the implication being that its larger creditors would be left with a shortfall.
924. It follows in my judgment that while Ferrero did understand that the transaction whereby Aksu Gida was to take over the Baskans hazelnut business as a front company was to be structured in a way that would conceal the Baskans' continuing beneficial ownership of the hazelnut business from Yapi Kredi, they were not given to understand that any deception would be used so as to conceal from Baskan Gida's creditors as a body the whereabouts of what had formerly had been Baskan Gida's assets. In that sense, Mr Do's half-informed assumption that the transaction was to be open and transparent was not incorrect. No attempt was made at any stage, whether by the Baskans, Ferrero or anyone else, to conceal the fact that the asset constituted by the hazelnut business had been transferred, virtually lock stock and barrel, by Baskan Gida to Aksu Gida.
925. The evidence does not therefore show that, as at 25^th January 2002, Ferrero knew that the proposed transfer to Aksu Gida involved anything unlawful other than a deception of Yapi Kredi, or that the attainment of the shared objective of preserving the hazelnut business from Yapi Kredi's unwelcome attention would necessarily defeat the claims of Baskan Gida' creditors as a class, still less the claims of the Banks. While I accept that Ferrero may well have thought it foreseeable that the transfer would take place in circumstances where some of Baskan Gida's creditors (and in particular its unsecured creditors) might not get paid in full, that perception falls critically short of anything sufficient to give rise to an intent to injure those creditors, as a necessary corollary of achieving the shared objective. Nor did the manner in which the Scheme was presented to Ferrero on 25^th January necessitate a conclusion that such shortfall as might be incurred by any creditors of Baskan Gida after the transfer would arise as a result of the use of any unlawful means in connection with its implementation.
926. I have concluded that, during the period between 25^th January and the end of February 2002, events occurred which did cause Ferrero to suspect that the Banks might, in addition to Yapi Kredi, be victims of the Baskans' readiness to deceive, in particular when Ferrero was asked to and did enter into large forward contracts with Baskan Gida which, if the Scheme proceeded, would be performed not by that company, but by Aksu Gida. Nonetheless I have not been persuaded that the suspicion thereby engendered ever reached a level at which it could properly be said that Ferrero had blind-eye knowledge of that aspect of the Baskans' dishonest plans, still less that Ferrero had an intention thereby to injure the Banks.
927. That is sufficient to dispose of the Banks' more narrowly formulated conspiracy claim against Ferrero. The common objective simply did not include the defeating of the claims of Baskan Gida's creditors as a class, so as necessarily to include the Banks. I should add that it is by no means clear how much if any the Banks' loss was caused by the transfer of the business to Aksu Gida, as opposed to the invoicing fraud, trust receipt fraud and fraudulent trading which preceded it.

The Conspiracy Claim against Mr Abidali

928. The Banks devoted considerable time and effort to a portrayal of Mr Abidali as a co-conspirator with the Baskans in 2001, dating his involvement from the early November meeting between him, Melih Baskan and Mr Dahod, on the basis that it was contemporaneous with the first manifestation of the 45 Day Lie, coincided with the fictitious date of the Protocol Agreement, and followed what the Banks claim to be Indo-Med's assistance to the Baskans in defrauding their Turkish banks.
929. For reasons set out fully in my findings of fact, I reject the claim that Mr Abidali was involved in any kind of conspiracy with the Baskans against the Banks prior to his visit to Turkey in early January 2002.
930. In my judgment, the case against Mr Abidali is to be addressed by reference to what he knew and did during and after that first visit to Turkey in early January 2002. Moreover, I have also concluded that Mr Abidali did not personally commit himself to participation in the Baskans' Scheme until after his second visit, from the 14^th to 17^th January, when it was explained to him in detail, and when the drafting of the sham documents began in earnest.
931. In my judgment the common objective which Mr Abidali shared with the Baskans from then on was to protect the Baskans' hazelnut business, including the Dahod Investment and Indo-Med's profit share in it, from Yapi Kredi and from Baskan Gida's other major creditors, by a transfer of its business and in particular its substantial stock of hazelnuts to a front company (Aksu Gida) to be beneficially owned and run by the Baskans. The end result sought to be achieved was to leave the Dahod Investors and Indo-Med as the profit sharing investors in the whole of the Baskans' hazelnut business, in circumstances where, freed from the claims of Baskan Gida's major creditors, that investment could be expected to yield the very profitable returns originally contemplated under the Dahod Agreement, with the added prospect, albeit not then finally agreed, that the Dahod and Indo-Med profit sharing entitlement would be extended to cover the whole of the Baskans' exports, rather than merely those to Ferrero.
932. The attainment of that objective, to Mr Abidali's knowledge, involved the use of the following means. First, the commission of a large and deliberate preference, by Baskan Gida effectively repaying in full the Dahod Investors' loan, primarily by means of the transfer of hazelnuts at Indo-Med's direction to Aksu Gida. Secondly, it involved the deliberate creation by the use of sham documents of a false appearance that the transaction was brought about by the exercise of commercial pressure by Indo-Med as an unpaid creditor, and that the effect of the transaction was the transfer of Baskan Gida's business and current assets, including its stock of hazelnuts, to a company owned as to 80% by Indo-Med, in which the Baskans had no beneficial share, control nor managerial involvement. Thirdly it involved a period during which the Transfer Scheme was to be prepared in secret, while Baskan Gida continued to trade, and to incur debts to its major creditors, including the Banks, which the Baskans did not intend should be repaid in full once the Scheme was fully implemented. During that period the nuts identified for transfer to Aksu Gida, or part of them would be being acquired by Baskan Gida with the use of funds representing both the Dahod Investors' original contribution and money drawn down under the newly completed Facility with the Banks.
933. I have concluded that there is no basis for attributing to Mr Abidali knowledge that the Scheme involved transferring to Aksu Gida hazelnuts belonging beneficially to the Banks (in the sense of being either pledged to or held on trust for them), nor that the Scheme involved the perpetration by the Baskans of the Invoicing Fraud. Nonetheless, I have found that Mr Abidali knew enough about the terms of the Scheme (namely that it was to involve the transfer of substantially the whole of Baskan Gida's current assets, leaving behind only land charged to Yapi Kredi) for it to be clear to him that debts incurred to the Banks during the period while the Scheme was being prepared, but prior to its becoming publicly known, were debts which the Baskans did not intend should be repaid in full. Furthermore, the way in which the Scheme was presented to Mr Abidali from early January 2002 onwards was, in my judgment, such as to demonstrate to him that Baskan Gida was insolvent, in every relevant sense.
934. Anticipating that there was likely to be a lively debate about whether the means which I have described were unlawful, and that in substance they consisted of dealings in Turkey with the assets of an insolvent Turkish company, it seemed to me that Turkish insolvency law might be of real relevance to the issue. The Banks' and Ferrero's Turkish law experts had not at that stage been instructed to assist in that regard, but at my request they prepared supplemental reports which helpfully presented a useful summary of the relevant provisions, about which they were, happily, in almost total agreement. I can therefore summarise it very briefly.
935. Turkish insolvency law is mainly derived from its Execution and Bankruptcy Code. It was substantially amended in 2004, but the experts' summary stated the law in force as at 2002. Sections 278 to 280 contain provisions providing for the avoidance of certain types of disposal of assets by insolvent persons, including gifts, transactions at an undervalue, and transactions reducing the insolvent's assets in a manner not expected of a prudent person acting in good faith, provided that application is made within stated time limits from the onset of insolvency. Transactions by an insolvent person with intent to cause damage to its creditors may be set aside at any time, provided that the recipient of the assets is shown to have known of the insolvent person's intention.
936. Turkish insolvency law contains no equivalent provisions for the setting aside of preferences as are to be found in English law, but the making of a preferential payment to an unsecured creditor who is not pressing for payment may constitute evidence of a transaction falling foul of Article 280 (that is, imprudent transactions otherwise in than in good faith), since the underlying principle of Turkish insolvency law is that the assets of an insolvent person ought to be applied, as between unsecured creditors, pari passu.
937. Leaving aside transactions intended to injure creditors, the provisions in the Turkish Execution and Bankruptcy Code for the avoidance of transactions by an insolvent company do not appear to categorise the transactions themselves as unlawful, any more than do the corresponding provisions in English law, to which the analysis in Dora v. Simper (supra) was applied. A person may pay his creditors in Turkey in whatever order he pleases, just as in England.
938. It follows that the obvious preference afforded to Indo-Med and the Dahod Investors by the Scheme did not, of itself, render the Scheme unlawful in Turkey, nor would an undervalue inherent in the terms upon which the hazelnuts were transferred (had it been proved) have done so either. Nonetheless I consider that the obvious trouble and expense to which the Baskans' legal team headed by Mr Alpay went in the construction of the fictitious appearance of an overdue debt, dishonoured promissory notes and the application by Indo-Med of creditor pressure, as reflected in the sham documents and the collusive bankruptcy proceedings, demonstrates a perception by them that, in the absence of an appearance of creditor pressure, the Scheme was vulnerable to being set aside under Article 280. The creation of a false picture of creditor pressure was therefore a deception by the perpetrators of the Scheme designed to mask its vulnerability to being set aside under Article 280. That deception was unlawful not because of any provisions of Turkish insolvency law, but simply because of its inherently fraudulent nature. Mr Werbicki was in my judgment correct in his concession that, deliberately to create an untrue impression of creditor pressure as a means of disguising a preference would be unlawful means (day 84 pp. 150 to 152).
939. The other main element of deception involved in the Scheme was the false portrayal of Indo-Med and Mr Aksu as the true owners and therefore controllers of Aksu Gida after the transfer of the hazelnut business, and the concealment of the Baskans' continuing beneficial ownership and control. Again, and despite Mr Abidali's evidence to the contrary, I have concluded that this was indeed a deliberate deception, known to Mr Abidali to be untrue, both at the stage of the planning of the Scheme, and in its implementation.
940. Again, it seems to me that that element of deception was written into the Scheme specifically to fend off a perceived vulnerability to dissatisfied creditors, who would be likely to seek to have a transfer by Baskan Gida to a Baskan owned front company set aside, either under Articles 278 to 280, or more generally as a transaction intended to injure creditors. Again, that element of deception was by its fraudulent nature also unlawful.
941. The Turkish experts were not invited to, and therefore did not address the question whether there is a Turkish equivalent of fraudulent trading, that is deliberately incurring debts knowing and intending that they are not to be repaid in full. Nonetheless, since this is in substance simply a particular species of fraud, there is no reason to suppose that it is any less unlawful in Turkey than in England. Since the Scheme involved Baskan Gida continuing to trade normally, and thereby to incur debts to the Banks under the Facility until after the business transfer to Aksu Gida had been fully implemented, it did involve fraudulent trading by the Baskans, as a third element of the unlawful means employed for the achievement of the Baskans' and Mr Abidali's common objectives.
942. I have concluded that Mr Abidali knew of the three aspects of unlawfulness which the Scheme embodied. He knew, of course, that neither Indo-Med nor the Dahod Investors were pressing creditors, so that the presentation of creditor pressure by the sham documents involved a deliberate deception. He was aware that the Baskans were to continue beneficially to own and to control the hazelnut business once transferred to Aksu Gida, so that the whole structure of the sham documents presented a deliberate falsehood. Finally, he knew enough about the way in which hazelnut purchases by Baskan Gida had been funded (by a mixture of Dahod Investors' money and draw-downs under the Facility from the Banks) for it to have been obvious that some of Baskan Gida's hazelnut stock (substantially all of which was to be transferred to Aksu Gida) had been funded using the Banks' money, at a time when the Baskans were arranging for Baskan Gida to borrow from the Banks money which, as Mr Abidali knew, they did not intend should be repaid in full.
943. The final question relevant to Mr Abidali's liability as an unlawful means conspirator, is whether he had the requisite intent to injure the Banks. In my judgment he did, for the following reasons. First, his knowledge that, while the Scheme was being prepared in secret, Baskan Gida was probably continuing to borrow in the usual way under the Banks' Facility money which the Baskans did not intend that it should repay in full necessarily involved, as the obverse of the coin, injury to the Banks in the same way that every victim of fraudulent trading is injured. The injury consists in losing money lent on a mistaken assumption, dishonestly induced, that the borrower intends to repay it. Secondly, the deceptions inherent in the structure of the Scheme documents, designed to reduce or eliminate its vulnerability to attack under Articles 278 to 280 of the Turkish Execution and Bankruptcy Code also involved injury to the Banks, since they were major creditors of Baskan Gida whose Turkish statutory rights to seek to set aside the transfer of Baskan Gida's assets and in particular its hazelnut stock, would be adversely affected if the deception were believed. To that extent the injury which is the obverse of the coin is only the consequential reduction in the amount of the Banks' claim in the liquidation of Baskan Gida attributable to an inability to rely upon Articles 278 to 280, where, but for the deception, those remedies would otherwise probably have been available.
944. I was much pressed in argument by Mr Werbicki with a 'lack of motive' submission, just as I had been by Mr Strauss on behalf of Ferrero. Why, he submitted, should Mr Abidali have been any more motivated than Ferrero to become involved in a fraudulent conspiracy against the Banks? Although much travelled, Mr Abidali's business base was in London, and he knew that the Banks' offices (for the purpose of the Facility) were also in London, having visited them with the Baskans. By becoming involved in a fraud on the Banks, he risked ruin and even loss of liberty even more clearly than did Ferrero. He had not invested his own or Indo-Med's money in Baskan Gida and was therefore, it was submitted, interested in the outcome of the Scheme only to the extent of Indo-Med's share of profits under the Dahod Agreement.
945. In my judgment Mr Abidali's motivation was as follows. First, the prospect of retaining rather than losing Indo-Med's profit share on the whole of the Baskans' exports to Ferrero for the 2001 crop and thereafter was, for a business of Indo-Med's size, significant. Secondly, Indo-Med depended upon the continuing health of the Baskans' hazelnut business for its own supplies, and for the continuation of the profit sharing venture which Mr Abidali already enjoyed under his arrangement whereby Indo-Med acquired hazelnuts from the Baskans at cost, and shared with the Baskans the profits of its on-sale of them, on a 50/50 basis, without having to pay his debts to the Baskans before his customers paid him. Thirdly, Mr Abidali owed the Baskans a favour, for having in effect bailed Indo-Med out of the large financial difficulties caused in the late 1990s by its involvement in sales to Russia. Fourthly, Mr Abidali had an obligation to his co-religionist Mr Dahod to protect the large investment in the Baskans which Mr Dahod had undertaken entirely on his advice.
946. Whether combined motivation of that kind might in the abstract be expected to be sufficient to encourage a hitherto respectable trader into a dishonest conspiracy, Mr Abidali's participation in the deceptions constituted by the sham documents is so obvious that the motivation which I have identified must have been sufficient. Mr Werbicki made much of the fact that the Scheme and accompanying documents had been approved by Turkish and English lawyers, so that Mr Abidali could not have thought there was anything improper in it. In my judgment Mr Abidali can gain no comfort from Mr Trotter's assistance. Either Mr Trotter was grossly negligent or worse, or (more probably) Mr Abidali managed to deceive him as to the underlying reality. As for Mr Alpay, he appears to have been the progenitor of the dishonest sham. The only comfort which Mr Abidali may have gained was that the dishonest Scheme would probably work, if the deceptions inherent in it were believed, which to an extent they were in Turkey.
947. As for the fraudulent trading by the Baskans as against the Banks during the period while the Scheme was being prepared and implemented in secret, that was an aspect of the unlawful means used in respect of which I doubt that Mr Abidali appreciated that he incurred any risk of liability, whether for himself or for Indo-Med. It was not something about which Mr Trotter appears to have known or advised.
948. The final question is to what extent, if at all, was the Banks' loss attributable to an injury necessarily caused by the attainment of the Baskans' and Mr Abidali's common objective by the use of the unlawful means which I have identified. At a late stage in the trial, the parties by agreement invited me to determine any quantum question on the basis that, only if my conclusion was that a particular defendant was liable either for the whole or for none of the Banks' loss, should I decide quantum now. By contrast, if my conclusion was that a particular defendant was liable only for some part of the Banks' loss, I should first invite additional submissions as to the precise proportion, in the light of the facts as found.
949. It is well settled that liability for participation in an unlawful means conspiracy does not extend to losses suffered by the victim prior to the date upon which the relevant defendant "joined" (in that sense) the conspiracy. It is therefore first necessary to determine when Mr Abidali should be taken to have joined it. In my judgment he did so on 19^th January 2002, having together with Melih Baskan obtained Mr Dahod's approval to go ahead with the Scheme (regardless how far they mis-described it to Mr Dahod for that purpose). Although the detailed drafting of documents had by then started, I consider that it would be wrong to treat Mr Abidali as having unequivocally decided to commit Indo-Med to the Scheme without Mr Dahod's approval, so that the drafting prior to that date was, however expressed, in substance subject to contract. The fact that the Scheme remained contingent upon Ferrero's approval, which was only obtained about a week later, does not affect the date of Mr Abidali's joinder, but may affect the date from which further trading by Baskan Gida should properly be regarded as being (to Mr Abidali's knowledge) fraudulent trading, since it was only at the point when the Scheme became unconditional that it could be seen to be only a matter of time before Baskan Gida ceased trading, leaving its major creditors unpaid. Mr Abidali's liability for loss caused by Baskan Gida's fraudulent trading vis-à-vis the Banks arises only because of his participation in a Scheme of which that was to his knowledge a necessary element.
950. It follows that, on any basis, Mr Abidali cannot be liable for the whole of the Banks' loss. The first draw-downs under the Facility had occurred before Mr Abidali joined the conspiracy, and the first six draw-downs had occurred before the Scheme became unconditional due to Ferrero's agreement to play its part in it. By then, the Banks had lent Baskan Gida in excess of €10 million, and may (although I need not decide that question at this stage) have incurred an irrecoverable loss of that amount for which Mr Abidali would not be liable.
951. Quite separately, Mr Abidali's participation in the false appearance of creditor pressure and independent ownership and control of Aksu Gida created by the sham documents would cause the Banks loss only to the extent that the value of their rights as unsecured creditors in Baskan Gida's liquidation were thereby adversely affected. That element of loss arising out of the unlawful means constituted by that deception necessarily requires quantification, since it is clear that an unwinding of the Aksu Gida transfer under Articles 278 to 280 of the Turkish Execution and Bankruptcy Code (even if sought within the relevant time limits) would not have led to the Banks being repaid in full, or anything like it. In such a liquidation, the Dahod Investors would through Indo-Med have had been able to prove for their full investment, and the Banks would also have faced competition from secured and unsecured claims of Baskan Gida's Turkish banks.
952. My conclusion at this stage therefore is that Mr Abidali is liable for a proportion of the Banks' loss, such proportion to be the subject of further enquiry.
953. I had at an early stage of the trial thought that an additional complication in the quantification of any claim by the Banks against Mr Abidali was that the extent (if any) of any contractual claim of the Banks against Ferrero pursuant to the Assignments remained to be determined in the Italian courts, and that Mr Abidali's liability could only be for the net loss suffered after giving credit for that contractual claim.
954. In his closing speech Mr Wardell for the Banks submitted that where a claimant has suffered loss by reason of a commission of a tort, in respect of which he might have an uncertain contractual or propriety claim against a third party, the quantum of the tort claim ought not to be regarded as reduced by the application of some notional value to the outstanding contract claim. He relied upon the well known case of Pilkington v. Wood [1953] Ch 770, where the quantum of the plaintiff's claim against his negligent solicitors was not treated as reduced by the fact that he retained a difficult claim under his vendor's covenants for title which he had not pursued. Again, I consider that this is a matter which ought to be resolved after hearing full argument, including in particular argument on Mr Abidali's behalf.

The Banks' Proprietary Claim against Ferrero

955. This claim arises out of the Banks' allegation that a proportion of the hazelnuts sold to Ferrero by Aksu Gida and Baskan Yuksel originated in Warehouse 2, so that they were, prior to release, held subject to Baskan Gida's pledge to the Banks and, following release, held upon trust for the Banks pursuant to the Master Trust Receipt. Putting on one side the formidably difficult factual question as to just how many hazelnuts sold by Aksu Gida or Baskan Yuksel to Ferrero did originate in Warehouse 2, it is convenient to address the legal issues first, assuming (as I do) that some hazelnuts fell within that class.
956. The Banks put their claim in two ways. The first is a claim in conversion, based upon an allegation that, under Turkish law, the pledge in favour of the Banks was not extinguished, either by the release of the hazelnuts from Warehouse 2, by Baskan Gida's transfer of them to Aksu Gida and to Baskan Yuksel, or by their subsequent sale of them to Ferrero. The case in conversion is therefore that Ferrero received the hazelnuts while they were still subject to the Banks' rights as pledgee.
957. The alternative claim is that when Ferrero received the hazelnuts as purchaser from Aksu Gida and Baskan Yuksel, it became liable for knowing receipt, since they remained subject to the Banks' rights under the Master Trust Receipt.

Conversion – Turkish Law

958. The Conversion claim gives rise to no difficulties under English law. The problems relate entirely to the relevant provisions of Turkish law, about which Professors Tekinalp and Ansay gave expert evidence.
959. The Turkish law issues between the parties and their experts largely arose from their differing interpretations of the following provisions of the Turkish Civil Code, all of which are closely modelled on equivalent provisions of the Swiss Civil Code, and usually interpreted by the Turkish courts in conformity with Swiss jurisprudence:

Article 939

"With certain exceptions provided by law, moveable chattels can be given in pledge only by delivery of the chattel to the creditor. Where the creditor bona fide takes possession of the chattel he acquires the right of a pledge over it even when the pledgor has no right to dispose of it, subject always, however, to the right of the third parties derived from their prior possession.

No pledge can be constituted so long as the pledgor retains exclusive control over the chattel."

Article 943

"A pledge is extinguished when the pledgee ceases to have possession of the chattel and cannot demand it back from the third parties in possession.

The pledge is suspended while the chattel is in the exclusive possession of the pledger with the consent of the pledgee."

Article 988

"Where a moveable chattel is transferred with a view to passing ownership or some other right to the transferee, and he takes possession in good faith, his right over the chattel must be protected, even where the transferor had himself no authority to alienate it."

Article 3(2)

"Bona fides is presumed whenever the existence of a right has been expressly made to depend on the observance of good faith.

No person can plead bona fides in any case where he has failed to exercise the degree of care expected from him by the circumstances."

Article 6

"In the absence of a special provision to the contrary, the burden of proving an alleged fact rests on the party who bases his claim on that fact."

960. The pledge of hazelnuts in Warehouse 2 was achieved pursuant to the terms of the Master Pledge Agreement, made between Baskan Gida as Pledgor and KBC as Pledgee. Clause 3.2 is as follows:

"The Pledgor hereby undertakes to transfer from time to time the possession of the Products to the Pledgee or to SGS, a custodian appointed by the Pledgee pursuant to the provisions of Article 853 of the Turkish Civil Code, against each Warehouse Receipt."

Clause 5, which constituted the cornerstone of the Banks' case, is as follows:

"The Security created hereunder shall be in addition to and independent of every other security which the Pledgee may at any time hold for any of the Secured Obligations, and shall remain in full force and effect as a continuing security unless and until the Secured Obligations have been irrevocably and unconditionally discharged in full…."

The Secured Obligations were all monies and other liabilities now or at any time becoming due and owing by the Pledgor under the Facility. The Master Pledge Agreement is expressed to be governed by Turkish law.

961. Professor Tekinalp 's opinion may be summarised as follows:

i) Clauses 3.2 and 5 of the Master Pledge Agreement reflected the common intention of the Banks and Baskan Gida that, provided money was still owing under the Facility, the release by SGS of hazelnuts from Warehouse 2 to Baskan Gida for the purposes of processing and onward sale to Ferrero merely suspended rather than extinguished the pledge, within the meaning of the second sentence of Article 943.

ii) Since neither Baskan Gida had authority from the Banks to transfer the hazelnuts to Aksu Gida, and since Aksu Gida knew of Baskan Gida's lack of authority, then Ferrero could only have obtained good title to the hazelnuts on a purchase from Aksu Gida (or Baskan Yuksel) if, pursuant to Article 988, it took possession in good faith.

iii) Since it made no careful or reasonable inquiries when acquiring the hazelnuts from Aksu Gida or Baskan Yuksel, Ferrero cannot have acted in good faith within the meaning of Article 3(2).

iv) In any event, regardless of the application of the second sentence of Article 943, the requirement of good faith is such a fundamental principle of Turkish law that a failure to take the degree of care expected by the circumstances within the meaning of Article 3(2) would disable Ferrero from obtaining title to the hazelnuts purchased from Aksu Gida, even if the pledge had been extinguished by the release of the hazelnuts from Warehouse 2.

962. Professor Ansay's opinion may be summarised as follows:

i) Notwithstanding clause 5 of the Master Pledge Agreement, the effect of the release to Baskan Gida of hazelnuts from Warehouse 2 was, in every case, to extinguish rather than merely suspend the pledge, because the second sentence of Article 943 was applicable only to a temporary release of the chattel to the possession of the pledgor, pending its return in due course to the pledgee.

ii) Even if the effect of the release of hazelnuts from Warehouse 2 to Baskan Gida was only to suspend the pledge, the effect of suspension is pro tem to remove the pledgee's rights, which revive as rights in rem only if and when the goods are returned to the possession of the pledgee. In the meantime the pledgee has merely a personal right against the pledgor to recover possession of the goods.

iii) Accordingly, whether the pledge was suspended or extinguished upon release of the hazelnuts from Warehouse 2, the pledgee's rights were ineffective as rights in rem when Baskan Gida sold the nuts to Aksu Gida and when Aksu Gida sold them to Ferrero, so that Aksu Gida had full authority to sell to Ferrero. The same analysis would apply to sales by Baskan Yuksel.

iv) A purchaser of a chattel from a person with authority to sell it need not show good faith.

v) If contrary to his opinion, Aksu Gida or Baskan Yuksel lacked authority to sell the hazelnuts to Ferrero, he expressed no opinion whether or not Ferrero took possession of them in good faith within the meaning of Article 988, leaving that to be determined by the court.

963. It became very clear during the experts' cross-examination that their differing views on these issues stemmed from two fundamentally opposed points of departure. Professor Tekinalp's starting point was that the relevant provisions of the Civil Code should be interpreted so as to uphold and enforce the requirements of good faith, so much so that every transaction by a possessor of goods under a personal contractual restraint as to the disposal of them could be impugned if a lack of good faith within the meaning of Article 3(2) was shown. By contrast, Professor Ansay's starting point was what he described as the "publicity principle", pursuant to which third parties who receive goods should normally be protected if the person transferring them has the appearance of being the owner. Closely related to that principle was his understanding that the whole system of property in movables under Turkish or Swiss law is based upon possession, such that a person in possession of goods is presumed to be the owner of them, and has authority to sell them by virtue of his possession. Thus he regarded Article 943 as reflecting in relation to pledges the underlying publicity principle, so that a person receiving originally pledged goods is unaffected by the pledge if he acquires them either from the pledgor to whom they have been returned or from a third party in possession.
964. In my judgment both the competing principles separately relied upon by Professors Tekinalp and Ansay can be identified as playing a part in the relevant provisions of the Civil Code. Neither entirely prevails against the other, and both experts tended to overplay the fundamental principle upon which each relied.
965. Both experts recognised as authoritative, and relied upon, the writings of a number of distinguished academic commentators, including in particular Professor Thomas Bauer in the Basler Kommentar on Swiss law, writing in particular on the meaning and effect of Article 888 of the Swiss Civil Code which is the Swiss equivalent of Article 943 of the Turkish Civil Code. I shall use the Turkish numbers throughout, although Bauer does not.
966. Bauer distinguishes for the purposes of Article 943 first, between transfer of possession by the pledgee to a third party (dealt with by the first sentence) and return of the goods to the exclusive possession of the pledgor (addressed by the second sentence). The present case is of course concerned with return of the goods to the pledgor. He described the second sentence of Article 943 as an exception to the general rule (reflected in the first sentence) that a pledge is lost whenever the pledgee ceases to have possession of the goods. He says that the second sentence is therefore to be narrowly interpreted, and he distinguishes between a purely temporary and short-term return of the goods to the pledgor, and a long-term or unconditional return of the goods, concluding that:

"Under sub-section 2, a short-term, unconditional duty of return will alone justify the only temporary inefficacy of the lien."

The translation available to me is less than ideal, but the meaning is clear. Anything other than such a short-term return of the goods to the pledgor, with an unconditional duty to hand them back to the pledgee, will lead to the extinguishment of the pledge. He gives as an example of a short-term return to the pledgor with an unconditional duty to return them to the pledgee, the occasional surrender of the pledged item or keys to the warehouse for maintenance, service or use by the pledgor.

967. In the present case, having due regard not merely to the terms of the Master Pledge Agreement, but to the Facility documents as a whole, Baskan Gida obtained the release of hazelnuts from Warehouse 2 not for some temporary purpose, following which they were to be returned to Warehouse 2, but permanently, so that they could be processed to the required standard and exported to Ferrero. Upon release, Baskan Gida owed the Banks no unconditional duty to return the nuts to Warehouse 2. It follows on Bauer's analysis (with which Professor Ansay agreed) that release of nuts from Warehouse 2 extinguished rather than suspended the pledge.
968. Under the heading "legal consequences" Bauer provides the following description of the effect of a temporary suspension of the lien under the second sentence of Article 943, to the Swiss equivalent of which he refers as sub-section 2:

"Sub-section 2 provides for the temporary suspension of the lien. Unlike in sub-section 1, the lien continues to exist. It is however ineffective so long as the pledgor exercises sole authority. If the item is pledged or the pledgor petitions for bankruptcy during this time, the creditor will be unable to assert his lien. The lien will be deemed not to exist….

The lien will be ineffective against third parties. With respect to the pledgor, only the right in contract to restoration of possession will remain… since the lien is ineffective during the critical phase, the pledgee cannot assert a reduced right in rem against the pledgor. The question is therefore of little practical importance…."

969. To the same effect is the following passage from Professor Dr M. Kemal Oguzman, writing on Turkish property law in 2006:

"Should the pledged thing have been returned with the pledgee's consent to the pledgor for a limited period and the opportunity thus arises for him to enjoy sole practical disposal over the thing, the pledge right is not extinguished but the provisions concerning the right of pledge are suspended (CC article 943/para 2). For example, if the chattel is temporarily returned to the pledgor for repair of maintenance purposes, the provisions concerning the right of pledge are suspended. For the duration of this period, the creditor may not apply for the pledge to be realised. As of the time at which the pledgee regains possession of the pledged thing, the provisions concerning pledges take effect once again. The suspension of the provisions concerning pledges does not affect the priority of the pledge. The pledge retains the degree of priority that it had when it was first created."

970. Bauer's analysis tended to support Professor Ansay's opinion on two of the critical matters in dispute. First, his distinction between short-term temporary return of the goods to the pledgor and a more permanent return unaccompanied by any duty to hand them back to the pledgee suggests that, on the present facts, the pledge was extinguished in relation to any batch of hazelnuts once released by SGS from Warehouse 2 to Baskan Gida for processing and sale. This was, coincidentally, Pekin & Pekin's understanding when drafting the Master Pledge Agreement and advising the Banks, and they added the Master Trust Receipt as a hoped-for alternative means of security in relation to the nuts once released from Warehouse 2, but without any particular confidence in its efficacy, on the assumption that the release would extinguish the pledge.
971. Secondly, Bauer's analysis of the legal effect of suspension, supported by Oguzman, tended to accord with Professor Ansay's opinion, that a suspended pledge is ineffective unless and until the goods are returned to the pledgee. Since in the present case nuts once released from Warehouse 2 never were returned to the possession of the Banks' agent SGS (whether in Warehouse 2 or otherwise), the pledge was ineffective as against either Aksu Gida, Baskan Yuksel or Ferrero in relation to any released hazelnuts.
972. As for clause 5 of the Master Pledge Agreement, while it appears at first sight to provide for the continuing efficacy of the pledge, I consider it to be better explicable as a general provision providing for the continued effect of the pledge, but only in relation to nuts remaining in Warehouse 2, for as long as anything should be owing under the Facility. It is by no means necessary to read it even as attempting to provide for the efficacy of the pledge in relation to nuts once released, and such a conclusion would militate against the commercial efficacy of the security structure, since processing and sale to Ferrero could not sensibly proceed while the nuts remained pledged. Even if it purported to do so, it was in my judgment ineffective for that purpose, on the true interpretation of Article 943.
973. Mr Wardell attempted to rely upon an example given by Bauer of the temporary release of a pledged antique to a third party restorer for repair. If the restorer sells the antique to a fourth party, or transfers it to the original pledgor, the fourth party will obtain title only if acting in good faith, while the pledgor will not escape the pledge because he is receiving it in bad faith. In my judgment the reason why good faith is relevant in that example is that the restorer was never the owner, and as a mere bailee from the pledgee never had any authority to alienate the chattel. In the present case by contrast, Baskan Gida was the owner throughout, whose authority to alienate was qualified only by the pledge.
974. It is nonetheless necessary to deal with Professor Tekinalp's broader opinion that a purchaser of chattels must show good faith wherever his title is derived from a sale in breach of a purely private restriction contractually imposed by a third party upon the owner. There being no such express provision to that effect in the Code, Professor Tekinalp's opinion to that effect depended critically upon his view that good faith is a universal requirement in Turkish civil law, the absence of which may undermine almost any transaction.
975. In that respect, I found that Professor Tekinalp's view was not supported by the academic texts. In particular, Dr Peter Jaggi, writing on the Swiss equivalent of Article 3 in Obligationenrecht (1973) said this:

"Like Article 2, Article 3 does not contain a principle which determines all legal relationships. It only refers to the provisions of law, which connect "a legal effect to good faith"…. Article 3 forms a general section for these. Its two legal clauses add to the individual provisions regarding good faith: para 1 is a rule of evidence subject to Article 8. Para 2 excludes the person not paying sufficient attention from protection by good faith…. Article 3 does not have any further content. In particular it does not paraphrase the term of good faith… or tacitly contain the clause that good faith is always protected."

976. To the same effect, Professor Dr Rona Serozan, of the Istanbul University Law Faculty, writing in 2004, said this:

"Under circumstances in which the law makes a legal result conditional on good faith, the presence of good faith is presumed (CC3I)."

The clear implication is that Article 3 is only triggered where some other provision requires good faith to be shown. In the present case Article 988 does so in terms, but only where the transferor of the chattel had himself no authority to alienate it.

977. In the present case, subject only to the Banks' pledge, Baskan Gida was the owner of the hazelnuts both before, during and after their period of storage in Warehouse 2. Thus, whether the effect of the release of hazelnuts from Warehouse 2 into Baskan Gida's possession was to extinguish or suspend the pledge, Baskan Gida recovered its full authority to sell as owner, such that neither Aksu Gida, Baskan Yuksel nor, a fortiori Ferrero when purchasing from either of them, was required to demonstrate good faith in connection with that acquisition. The authority of an owner in possession of a chattel to alienate it is inherent in his ownership. A contractual restriction on alienation merely exposes the owner to personal liability if he alienates the chattel contrary to that restriction. It does not affect his authority, unless the restriction takes the form of a right in rem, such as a pledge.
978. My conclusions on the Turkish law issues may therefore be summarised as follows:

i) The pledge was extinguished upon the release of each batch of hazelnuts to Baskan Gida.

ii) Even if, contrary to (i), the pledge was only suspended, it was ineffective as a right in rem in relation to released hazelnuts when transferred subsequently to Aksu Gida, Baskan Yuksel and Ferrero.

iii) Each of Baskan Gida (as owner), Aksu Gida and Baskan Yuksel (as transferees from an owner) therefore had authority to alienate the hazelnuts.

iv) Ferrero therefore obtained good title to them, free from any pledge, and without the need to plead good faith.

979. My analysis of the Turkish law issues makes it unnecessary for me to decide whether Ferrero did act otherwise than in good faith when purchasing hazelnuts from Aksu Gida which originated from Warehouse 2. Since however the Turkish experts both agreed that this is a fact sensitive question, I shall briefly express my conclusions about it, in case a higher court should take a different view about the Turkish law issues which I have already decided. The question is, "whether Ferrero failed to exercise the degree of care expected from it by the circumstances". I should note that the translation of Article 3 of the Turkish Civil Code produced by Professor Tekinalp used the phrase "required by the circumstances" but in cross-examination he agreed with Professor Ansay's view that the phrase "expected from him (or it) by the circumstances" was a better reflection of the Turkish text.
980. Although the experts differed in their conclusions as to the application of Article 3 to the assumed facts, I could discern no real difference between them as to its legal meaning. Where for any purpose expressly provided for by the Civil Code a party needs to demonstrate good faith (for example to obtain ownership of a chattel on a purchase from someone without authority to alienate it under Article 988), then the test imports something more than mere honesty. Where the facts about a transaction would put a reasonable purchaser on inquiry as to the vendor's authority or title, then the purchaser must make such inquiries as to the vendor's title or authority as, taking into account all relevant facts, are reasonable in the circumstances. The relevant facts include the qualifications and skills of the purchaser, but the test is otherwise largely objective. If the purchaser fails to make such reasonable inquiries, then he will be affected by such adverse rights or want of authority which those inquiries would have revealed. Conversely, if reasonable inquiry would not have revealed the adverse rights or want of authority, the purchaser will take free of them.
981. The nearest analogy to that Turkish law concept seems to me the English law doctrine of constructive notice, whereby the purchaser of a legal estate in land will be affected by equities affecting that estate if he fails to make such reasonable inquiries of the vendor's title as are called for by transactions of the relevant kind. Failure to make those inquiries leaves the purchaser with constructive notice of such adverse rights as reasonable inquiries would have revealed: see Megarry & Wade's Law of Real Property (7^th edition) at paragraph 8-018. Professor Ansay, whose evidence on the meaning of the good faith in Article 3 I found more impressive than that of Professor Tekinalp, warned that reasonable inquiries in relation to the purchase of land were regarded by the Turkish courts as inherently likely to be more extensive than reasonable inquiries in relation to the purchase of a chattel, due to the greater propensity for land to be subject to adverse rights. The doctrine of constructive notice is of course of particular application to the purchase of land in English land law, and the English courts have shown a similar reluctance to extend the doctrine into commercial transactions about goods.
982. The factual question whether Ferrero's conduct met the Turkish law standard of good faith under Article 3 has to be addressed separately in relation to two periods during which it acquired hazelnuts from Aksu Gida. The first is the period from the implementation of the business transfer on 13^th February 2002 until Mr Byles' telephone conversation with Mr Casale on or about 4^th March. The second is the period from then until the cessation of supplies to Ferrero by Aksu Gida and Baskan Yuksel.
983. The critical difference between the two periods which I have identified is that it was on 4^th March that Mr Casale told Mr Byles that Ferrero was in the process of buying hazelnuts from Aksu Gida, but Mr Byles decided not to inform Mr Casale that the Banks claimed proprietary rights in relation to hazelnuts originally stored in Warehouse 2, to the extent that they might have been transferred upon release to Aksu Gida for use in connection with its business. Otherwise, it seems to me that the relevant facts are the same in relation to both periods. Ferrero made no inquiries, either of Aksu Gida or of the Banks before purchasing and taking delivery of hazelnuts from Aksu Gida or Baskan Yuksel.
984. The first question therefore is whether the circumstances were such that any inquiries were to be expected of Ferrero. The second is whether reasonable inquiries would have revealed the Banks' rights (on the hypothesis which I have thus far rejected that the Banks had any continuing property rights sufficient to deprive Aksu Gida of authority to sell the nuts, under Turkish law).
985. In my judgment the circumstances were not such as to give rise to any obligation on Ferrero to make inquiries under Article 3. My reasons are as follows. First, Ferrero knew nothing whatsoever about the Master Pledge Agreement, still less about the terms upon which any nuts released from Warehouse 2 were thereafter held by Baskan Gida. All Ferrero knew was that at some stage during the life of the Facility, it had been necessary for third party consent to be obtained in relation to certain nuts in advance of processing, and that there may have been third party security rights in relation to some of the nuts stored from time to time by Baskan Gida at Giresun. Secondly, Ferrero knew nothing at all about the sources from which Aksu Gida or Baskan Yuksel were obtaining nuts for on-sale to Ferrero, after the business transfer, save that Mr Casale and Mr Rosa Brunet no doubt hoped and assumed that Aksu Gida would take over Baskan Gida's supply chains.
986. My third reason relates to the possibility that careful thought might have suggested that some of Aksu Gida's early deliveries had been made from hazelnut stocks already obtained by Baskan Gida, if only because of the apparently smooth transition from the one company to the other during the period immediately following the transfer requests on 13^th February. Although I have concluded that during this first period Ferrero probably entertained a growing suspicion that, in some undefined way, the Baskans might not have been acting properly or frankly vis-à-vis the Banks, I have concluded that Ferrero had no reason even to suspect that the Baskans were abusing any continuing property rights of the Banks in relation to hazelnuts offered for sale by Aksu Gida, even though Ferrero might have suspected that some part of those nuts had been acquired with the use of the Banks' money.
987. Fourthly, I have no reason to think that reasonable inquiries either of Aksu Gida or of the Banks would have brought to Ferrero's attention any want of authority in Aksu Gida to sell the nuts. As for enquiries of Aksu Gida, I have little doubt that the Baskans would have lied to Ferrero about it. As for the Banks, Mr Byles' evidence was that when Mr Casale told him that Ferrero was buying nuts from Aksu Gida, it did not occur to him that there was any possibility that Aksu Gida held nuts which were still subject to the Banks' security. During the first period, prior to Mr Casale's conversation with Mr Byles, I see no reason why inquiries by Ferrero should have extended to the Banks, at least until Ferrero was notified on 27^th February of outstanding payments due to them. Thereafter, for the reasons which I have given, I consider it unlikely that inquiry of the Banks would have put Ferrero on notice of any property rights.
988. For those reasons, the Banks' claim in conversion fails.

Knowing Receipt

989. There is no doubt that, to the extent that Baskan Gida transferred nuts released from Warehouse 2 to Aksu Gida, it acted in breach of the Master Trust Receipt which affected all nuts released from Warehouse 2 and that, since Aksu Gida through the Baskans knew all the relevant facts, it received those nuts as constructive trustee, such that any of those nuts on-sold by Aksu Gida to Ferrero were transferred in breach of constructive trust. The same applies to sales to Ferrero by Baskan Yuksel.
990. The insuperable difficulty with the Banks' knowing receipt claim is proof of the necessary knowledge on Ferrero's part. Ferrero had no more knowledge of the Master Trust Receipt than it had of the Master Pledge Agreement. In BCCI (Overseas) Ltd v. Akindele [2001] Ch 437, the Court of Appeal held that liability for knowing receipt depended upon the recipient having knowledge that the assets received were traceable to a breach of trust or of a fiduciary duty, and identified as a single test of knowledge in that context that it must be such as to make it unconscionable to retain the benefit of the receipt: see per Nourse LJ at page 455 E to G. Knowledge for that purpose is to be distinguished from mere notice, and requires at least "a clear suspicion": see Uzinterimpex JSC v Standard Bank plc [2008] EWCA Civ 819, per Moore-Bick LJ at para 44.
991. In my judgment the application of that test to the facts which I have found produces a negative conclusion as to Ferrero's knowledge. It knew neither of the constructive trust, nor of the breach of it, nor that, whether or not there had been a breach of trust, the hazelnuts acquired from Aksu Gida had originally been subject to any trust in favour of the Banks. The same insuperable difficulty affects any claim in relation to hazelnuts which had been purchased with the proceeds of sale of nuts held on constructive trust.
992. That short conclusion is sufficient to dispose of the Banks' knowing receipt claim.

The underlying factual question

993. My negative conclusions on the legal aspects of the Banks' claims in conversion and knowing receipt make it unnecessary for me to decide the large and very difficult question to what extent nuts sold by Aksu Gida or Baskan Yuksel to Ferrero did originate from Warehouse 2. As with almost every issue in this case, enormous quantities of time, effort, cost and tenacity were applied to this question. The Banks and Ferrero used distinguished accountants as forensic experts to present their respective arguments by reference to numerous lever arch files of supporting documentation. For his part, Mr Abidali advanced his own arguments, based in particular upon Aksu Gida's inspection records deriving from an apparent stock-take conducted at the supposed time of the transfer of the hazelnuts from Baskan Gida at the end of January 2002.
994. I am acutely conscious that a trial judge should not leave issues of fact unresolved merely because, on his analysis, they are unnecessary for his decision, if there is a real risk that a higher court might take a different view of the law, or of other facts forming part of the analysis. I have however concluded that no useful purpose would be served in the present case by my attempting to determine this very difficult question with any precision. My reasons are as follows. First, for the question how many of the hazelnuts released from Warehouse 2 formed part of Aksu Gida or Baskan Yuksel's sales to Ferrero to matter it would be necessary for a higher court to form a different view not merely of the applicable principles of Turkish law, but of my factual conclusions that Ferrero knew neither of the Master Trust Receipt, nor of the Master Pledge Agreement, such that knowing receipt cannot be proved, nor the absence of good faith within the meaning of Article 3 of the Turkish Civil Code. Secondly, if a higher court was to take a different view from mine as to those two factual questions, it seems to me very likely that a different view would also have been taken of Ferrero's liability as an unlawful means conspirator. It is only if I were upheld on conspiracy, but reversed on knowing receipt or good faith under Article 3, that it would be necessary to decide this outstanding factual question.
995. Thirdly, the answer to this question does not depend to any material extent upon my view of the relative credibility of witnesses who were cross-examined. Neither of the forensic experts were cross-examined, mainly because pursuant to directions given at a case management conference, and upheld during the trial, no permission was given for the opinions of the forensic experts to be admissible as evidence, although I was invited to read them as helpful encapsulations of the parties' respective submissions on the issue. Similarly, Mr Abidali's case on this issue was set out in a verified statement of case rather than in a witness statement largely because, again, he had no admissible first hand evidence of his own to contribute. Mr Aslan's evidence about the day to day handling of receipts and releases from Warehouse 2 was mainly uncontentious, and none of the relevant employees of Baskan Gida or Aksu Gida were called. The result is that the determination of the issue depends upon the weighing of argument based upon documents of widely varying degrees of reliability. It follows that, in the unlikely event that this issue needs to be determined, it can be done by means of an inquiry based upon submissions and documents, without the need for oral evidence, and without the risk of there having to be a retrial.
996. Finally, the issue is one of quite extraordinary complexity and difficulty, the precise resolution of which would significantly extend both the length and preparation needed for delivery of this already very long judgment. To illustrate the difficulties, I need do no more than quote from the forensic experts' joint statement, under the heading:

"The Destination of Hazelnuts Released from Warehouse 2

11. The release instructions from Warehouse 2 identify only that the destination of the hazelnuts is Baskan Gida, but no other information that allows identification of destination or customer.

12. A lack of key documentation, such as the underlying accounting records of Baskan Gida, Aksu Gida and Baskan Yuksel or documentation that would allow the identification and distinction of specific bags of hazelnuts, means that it is not possible to directly trace the sales of specific hazelnuts by Baskan Gida, Aksu Gida and Baskan Yuksel to Ferrero back to the hazelnuts released from Warehouse 2. …

13. If Baskan Gida did hold further stocks outside of Warehouse 2 as at 31^st December 2001 or after this date, then these could have formed part of what was sold to Ferrero.

14. There is no evidence, either way, on whether any hazelnuts were deposited in Warehouse 2, released and then re-deposited in Warehouse 2."

997. Where necessary, the court will not baulk at deciding issues even where forensic experts, as here, agree that decision is impossible. Nonetheless, in the absence of such a necessity, my resolution of this question would necessarily involve a large degree of what accountants would probably call speculative guesswork. In my judgment that process is, at least at present, neither necessary nor appropriate.

Misrepresentation – Deceit and Negligence

998. The Banks make claims in deceit and negligence in relation to four categories of documents. They are:

i) The July and December Letters.

ii) The Framework Contract, in its original form signed on 15^th October 2001, and as amended and resigned on 13^th February 2002.

iii) The Ferrero Industrial Acknowledgements signed on 6^th and 13^th December.

iv) A group of Baskan Gida contracts countersigned by Ferrero, namely Contracts numbered 5094-7, 5088-9, 5100-1 and 5104.

999. The legal principles applicable generally to the Banks' misrepresentation claims were not in significant dispute, and may be summarised as follows. First, the definition of the tort of deceit:

"Where a defendant makes a false representation, knowing it to be untrue, or being reckless as to whether it is true, and intends that the claimant should act in reliance on it, then in so far as the latter does so and suffers loss the defendant is liable for that loss." (Clerk & Lindsell on Torts (19^th edition) paragraph 18-01.)

1000. It is no defence that the maker of the false statement acted without a criminally dishonest motive, for example to assist the bringing about of a business transaction which he believed would be beneficial to both parties: see Standard Chartered Bank v. Pakistan National Shipping Corporation (No 2) [2000] 1 Lloyd's Rep 218 at 224 (reversed by the House of Lords, but on other grounds at [2003] 1 AC 959).
1001. In determining whether there has been a misrepresentation, the meaning of the statement in question is to be "judged objectively according to the impact that whatever is said may be expected to have on a reasonable representee in the position and with the known characteristics of the actual representee" (per Mance LJ in Primus Telecommunications plc v. MCI Worldcom International Inc [2004] EWCA Civ 957, at paragraph 30, applying Investors Compensation Scheme Ltd v. West Bromwich Building Society [1998] 1 WLR 896 at 912.
1002. Nonetheless, the representor is not liable in deceit if he believed that the statement was true in the sense in which he understood it, even if that sense differs from its true meaning, objectively construed: see Cartwright on Misrepresentation, Mistake and Non-Disclosure (2007) at paragraph 5.18 and Angus v. Clifford [1891] 2 Ch 449, in particular per Lindley LJ at 466 where, commenting on observations of Lord Herschell in Derry v. Peek (1889) LR 14App. Cas 337, he said:

"The passages about knowledge – knowingly making it, and making a statement without believing its truth, are based upon the supposition that the matter was really before the mind of the person making the statement and, if the evidence is he never really intended to mislead, that he did not see the effect, or dreamed that the effect of what he was saying could mislead, and that the particular part of what he was saying was not present to his mind at all, that I should say is proof of carelessness rather of fraud."

That was a striking case in which a company prospectus contained a statement signed by the directors that reports by engineers about a mine which the company proposed to acquire and work had been "prepared for the directors". The reports were attached to the prospectus and spoke in glowing terms of the mine, but had in fact been prepared for the intending vendors of the mine rather than for the purchasing company. The Court of Appeal, reversing Romer J, held that since it had never occurred to the directors that the prospectus would be read by intending investors as meaning that the directors had themselves commissioned the report (which, objectively, it plainly did mean) they could not be held liable for deceit. Their conduct was described as grossly negligent, but in those days that gave rise to no liability. Negligence may therefore consist not merely of a failure to apply due care in checking the truth or otherwise of a representation made, but also in a careless failure to appreciate the objective meaning of the statement made.

1003. The Banks allege that the statements made both in the July Letter and in the Framework Contract (in its original form) were continuing representations such that, even if true when made, they could give rise to a liability in deceit or negligence if Ferrero became aware subsequently that they had become untrue, and failed to correct them before the completion of the Facility. The question whether a representation has that continuing character depends in every case upon the context in which it is made. An inference to that effect will readily be drawn where the representor is seeking to induce a contract between himself and the representee: see for example With v. O'Flanagan [1936] Ch 575, in particular where the claim is for rescission of the contract subsequently made. Examples of continuing representations giving rise to a liability in tort (i.e. where, as here, the representor is not to be a party to the contract under negotiation) are less common, but not inherently impossible: see Slough Estates plc v. Welwyn Hatfield District Council [1996] 2 EGLR 219 at 236 to 7 per May J.
1004. Misrepresentation, whether giving rise to a claim in deceit or negligence, is only actionable if it induces the claimant to act in a manner that caused him to suffer the relevant loss. In the language of the tort of deceit, the false statement must be a material inducement. In the language of negligence, the claimant must rely upon it. Sometimes the difference in language is important: see per Millett LJ in Bristol & West Building Society v. Mothew [1988] Ch 1 at 10-11. Usually, the result is the same, as appears from the following passage in the judgment of Stephenson LJ in JEB Fasteners v. Marks Bloom & Co [1983] 1 All ER 583, at 588 to 9:

"In such a case the cause of action is the same as in all claims for damages for misrepresentation. The representation must be false, and it must induce the plaintiff to act on it to his detriment. If he does, he relies on it; if he does not, he does not. He may, of course, rely on other things as well. What operates on his mind, or motivates him, or influences him to act as he does, may be a number of things, some operating more or less strongly, one perhaps predominating… But, as long as a misrepresentation plays a real and substantial part, though not by itself a decisive part, in inducing a plaintiff to act, it is a cause of his loss and he relies on it, no matter how strong or how many are the other matters which play their part in inducing him to act."

To much the same effect is the following passage from Edgington v. Fitzmaurice (1884) 29 Ch D 459, at 483, per Bowen LJ:

"But such a misstatement was material if it was actively present to his mind when he decided to advance his money."

1005. Where a claimant suffers loss in reliance upon a misrepresentation, it is not necessary for him to prove either in deceit or in negligence that he would have acted differently if he had been told the truth: see Downs v. Chappell [1997] 1 WLR 426, at 433 (in relation to deceit) and Bristol & West Building Society v. Mothew (supra) at 11E (in relation to negligence). It is sufficient if the plaintiff proves that, had the false statement not been made to him, he would not have acted as he did.
1006. Thereafter, and mainly for policy reasons, the law as to the consequences of deceit and negligence in relation to misrepresentation diverges. In deceit, it is simply a question of causation, there being no defences of remoteness, contributory negligence or what is commonly mis-called scope of duty.
1007. Nonetheless, a defendant to a claim in deceit may rely upon a novus actus interveniens, if that consists of a third party's fraud which occurred after the claimant's entry into the transaction induced by the deceit: see Smith New Court Ltd v. Scrimgeour Vickers [1997] AC 254 at 285F to H, applying Twycross v. Grant 2 CPD 469 at 544 to 545, by analogy. On its facts, Smith New Court was a case about loss caused by an antecedent concealed fraud, which was only discovered after the purchase of the shares induced by the defendant's deceit. Accordingly, although the defendant was unaware of the antecedent fraud, it afforded no defence to liability for that loss.
1008. In negligence, the position is very different. First, the defences of remoteness and contributory negligence of course apply. Logically anterior to issues arising from those defences is the question whether the particular kind of loss suffered falls within the extent of the liability which the law imposes for breach of the particular duty owed. Where a claimant seeks information about a particular aspect of a contemplated transaction, (rather than general advice as to its wisdom), the defendant is, in general, not liable for loss which the claimant would still have suffered even if the information provided had been true. His responsibility is only for the consequences of the information being wrong: see South Australia Asset Management Corp v. York Montague Ltd [1997] AC 191 ("the BBL Case") at page 214 C to D per Lord Hoffmann.
1009. The giving of a trade reference to a bank contemplating lending to the subject of the reference is a paradigm example of the provision of information about, rather than advice as to the wisdom of, a contemplated transaction, to which the BBL principle applies. It is sometimes called the scope of duty principle, but even Lord Hoffmann has now recognised that this label is less than ideal: see (2005) 121 LQR 592, at 596. The BBL principle simply limits the extent of the loss for which the maker of the negligent statement is held by the law to be liable by excluding all that loss caused by the recipient's reliance upon the negligent statement which the recipient would have suffered even if the statement had been true. Sometimes the limitation operates by way of type of loss, sometimes by limiting its amount. It is sometimes described as a cap on the claimant's loss by reference to the liability which the maker of the negligent misstatement would have incurred, had he made it by way of contractual warranty. The BBL case was itself about liability for breach of a contractual duty of care. Nonetheless the BBL principle has been applied by analogy to cases of purely tortious liability, most recently in Calvert v. William Hill [2008] EWCA Civ 1427. It would be extraordinary if the voluntary provider of a gratuitous trade reference was subjected to a wider liability than a person who provided the same information for reward.
1010. Closely related to the BBL principle are the rules as to remoteness: see Platform Home Loans v. Oyston Shipways Ltd [2000] 2 AC 190 at 208 per Lord Hobhouse. Loss which is not the reasonably foreseeable consequence of the breach of duty is, as a matter of policy, outwith the scope of the defendant's liability.
1011. It is unnecessary in this case to enter into the debate whether, in cases where tortious negligence causes pure economic loss, the contractual rules as to remoteness should displace the ordinary rules applicable in tort. Both Mr Wardell and Mr Strauss accepted that, on the facts of the present case, the application of the rules as to remoteness were unlikely to give rise to any narrowing of the scope of Ferrero's liability beyond that which would be caused by the application of the BBL principle.
1012. Armed with those legal principles, I now address directly the specific alleged misrepresentations relied upon by the Banks. Again, as with the instances of alleged overt acts in the conspiracy claim against Ferrero, I have covered much of the relevant ground in my findings of fact.

The Trade References

The July Letter

1013. I have already concluded that, measured against the facts as they stood at the time, the 31^st July Letter contained no misrepresentation at all. Nonetheless the Banks claim that it was a continuing representation, and became untrue later in 2001, so that Ferrero Industrial came under a duty to correct it, with which it failed to comply.
1014. In my judgment, the July Letter did not contain continuing representations. The context in which it was given was that the Banks and the Baskans were negotiating a detailed loan facility, and neither had undertaken any obligation to keep Ferrero informed of the progress of the negotiations. Ferrero had simply been asked to give, and had given, a trade reference at a particular point in time, and then been left largely in peace, until approached again in connection with the Framework Contract.
1015. To treat one-off trade references as continuing representations would give rise to risks of grave injustice to the providers of those references, since they would be required, as it were, to remember thereafter what they had said, to keep a track on the underlying facts relevant to the reference, and to monitor the progress of the negotiations, so as to know whether a duty to alert the representee to a change of facts still had any practical content. That would in my judgment impose an entirely unreasonable burden on the provider of a gratuitous trade reference.

The December Letter

1016. I have concluded in my findings of fact that this trade reference contained a single misrepresentation, namely that Baskan Gida provided shipments to the Ferrero operating companies on a timely basis, and that this misrepresentation was made negligently, rather than fraudulently, by Mr Casale because, like the defendant directors in Angus v. Clifford (supra) he failed to appreciate that, objectively construed, the reference in the letter to a timely basis for deliveries meant timely in accordance with Ferrero purchase orders, rather than timely in terms of meeting Ferrero's production and stock requirements. This being therefore a case of negligence rather than deceit, I must deal with the consequential issues of reliance, causation, the BBL principle, remoteness and contributory negligence. I shall first deal with all the other allegations of misrepresentation, before returning to this one for that purpose.

The Framework Contract

The Framework Contract as signed on 15^th October 2001

1017. I have already concluded that, apart from an immaterial small mis-description of the precise periodicity of the average rate of orders in clause 3.2, the Framework Contract contained no misrepresentations at all. Since there was no material misrepresentation, questions of fraud or negligence in connection with the making of the Framework Contract do not arise.
1018. Yet again, the Banks claim that the Framework Contract contained continuing misrepresentations, which became untrue thereafter, such that Ferrero Industrial incurred a duty to correct them, with which it failed to comply. In my judgment the Framework Contract did not contain continuing representations. The context in which it was made was that Mr Casale had rejected a much more detailed draft contract, with elaborate provisions about variation, termination and arbitration, on the basis that Ferrero Industrial was prepared to do no more than sign a brief statement setting out the basis of Ferrero's then contractual arrangements with Baskan Gida, provided that it involved Ferrero entering into no new obligations or altering the way in which it then did business with Baskan Gida. In substance therefore, the Framework Contract was to constitute no more than a snapshot as at the date of its signature of the way in which Ferrero and Baskan Gida did business, leaving the parties entirely free to vary their arrangements thereafter, or to alter the pattern of their trading, even if that involved no alteration in its underlying terms and conditions.
1019. Furthermore, there are two reasons why the Banks cannot have regarded the Framework Contract as containing continuing representations, such that they expected any alteration of the terms or pattern of trading to be notified to them by Ferrero thereafter. The first is that the 45 Day Lie contained a description of a significantly different pattern of trading than that set out in the Framework Contract, yet it was unaccompanied by any notification to the Banks from Ferrero that any such change had occurred. Secondly and more importantly, when the Banks sought during the negotiation of the Acknowledgments of Assignment to impose an obligation on Ferrero not to change its terms of trading with Baskan Gida without the Banks' consent, this was roundly rejected by Mr Casale. Ferrero thereby made it clear to the Banks that it reserved the right to change the terms of trading with Baskan Gida at any time, so the Banks cannot have entered into the Facility in reliance upon any assumption, one way or the other, as to the terms or pattern of trading which was then, or which might during the currency of the Facility be agreed upon between Ferrero and Baskan Gida, by way of departure from the snapshot summarised in the Framework Contract.
1020. It follows in my judgment that although of course the pattern of trading between Ferrero and Baskan Gida changed substantially between 15^th October 2001 and the completion of the Facility, and continued to change thereafter, this gave rise to no notification duty on Ferrero's part. Even if it had done, the Banks could not have relied upon the absence of such notification, from the moment when Ferrero rejected the Banks' attempt to tie them down to the terms of the Framework Contract, during the negotiation of the Acknowledgements, before the completion of the Facility.
1021. The following passage in the judgment of Ferris J in WPP Group plc v Paul Reichmann (unrep. 23 August 2000) at paragraph 63, rejecting an analogous claim of continuing representation, captures the air of unreality about this part of the Banks' case:

"There is, in my judgment, a complete artificiality about an argument which starts with a statement which appears to amount to a promise, accepts that such promise has no contractual effect, proceeds to extract an implied statement of fact out of the promise, treats that as a statement that, unless corrected, the fact continues to exist, and concludes by stating that the legal effect is substantially the same as if the promise had been enforceable in the first place. As a basis for a claim in fraud it is, in my judgment, unsustainable. "

I consider it to be equally unsustainable as the basis of a claim in negligence.

The Framework Contract as amended on 13^th February 2002

1022. I have already concluded that by countersigning the relevant page of the Framework Contract containing the completed details of the Banks' Collection Account, Ferrero made no misrepresentation, let alone a fraudulent one. The questions of negligence, materiality, and reliance do not therefore arise. At the highest, the countersignature by Ferrero of the completed clause 6.2 merely asserted that, as at the 13^th February, Ferrero still considered itself obliged to pay money due on Baskan Gida invoices into either the Collection Account, or into such other account as Baskan Gida should direct. That was true, and Ferrero continued to do so during the short period after 13^th February while deliveries from Baskan Gida continued.

The Ferrero Industrial Acknowledgements of Assignment

1023. I have thus far dealt in detail with the facts about Ferrero's negotiation and signature of the Acknowledgements of Assignment and, in particular, rejected the Banks' case that Mr Casale had a sinister intent as a motive for the manner of his participation in the drafting. I must now address the Banks' case that the Ferrero Industrial Acknowledgements contained misrepresentation.
1024. The Banks allege that both Acknowledgements (signed on 6^th and 13^th December) contained implied representations that:

i) The Framework Contract contained a fair summary of the trading relationship between Baskan Gida and Ferrero on those two dates in December; and

ii) Ferrero expected to continue with the trading relationship thereafter on the terms of the Framework Contract.

In my judgment the Ferrero Industrial Acknowledgements made no such implied representations. My reasons follow.

1025. Leaving aside the all-important undertaking to pay invoices into the Collection Account, the only representation made in either of the Ferrero Industrial Acknowledgements was that it had received the notice of assignment of the Framework Contract, from Baskan Gida. The obtaining of that statement and the accompanying undertaking were the only purposes for which the Acknowledgements had been sent for signature by Ferrero Industrial.
1026. During the course of negotiation of the Ferrero Industrial Acknowledgement, Ferrero had by its rejection of the Banks' attempt to obtain an undertaking not to change the terms of the Framework Contract without their consent, made it clear that it did not regard itself as bound in any way to the Banks to carry on business with Baskan Gida in accordance with the terms or pattern of trading set out in the Framework Contract as at 15^th October, or at all.
1027. Bearing in mind that (subject to negotiation by Ferrero) the Acknowledgments were prepared by the Banks, I consider that if they wished to obtain confirmation from Ferrero that the terms and pattern of trading with Baskan Gida were, on the eve of the completion of the Facility, still in accordance with the Framework Contract, and were expected to continue to be in accordance with it thereafter, it was as a matter of common-sense and fairness incumbent on the Banks to seek and obtain that assurance in express terms. In fact, the Banks knew (or thought they knew) from the receipt of the 45 Day Lie, that the pattern of trading had changed, or was expected to change, to a significant extent.
1028. In his closing submissions, Mr Strauss described the Banks' attempt to identify implied misrepresentations in the Ferrero Industrial Acknowledgments as an after-the-event intellectual parlour game. I agree. All the Banks were seeking by asking for Ferrero Industrial's signature of the Acknowledgments was the construction of a documentary paper trail which would provide an effective assignment to them of Baskan Gida's rights to payment for hazelnuts supplied to Ferrero. Whether the Banks got that or not is a matter for the Italian Courts.

The Countersigned Contracts

1029. I have dealt with the circumstances in which each of the relevant contracts were countersigned in my findings of fact, and concluded that Ferrero, and Mr Casale and Mr Rosa Brunet in particular, did not by signing them knowingly assist the Baskans in the furtherance of their dishonest designs against the Banks. I must now address the separate question whether by signing these contracts, or any of them, the relevant Ferrero defendants made any fraudulent or negligent misrepresentations.
1030. The relevant contracts fall into four discrete groups and, although the misrepresentations alleged to have been made in each of them overlap, they are not identical. It is therefore necessary to deal with each group separately, subject to the following preliminary observations.
1031. Each of the relevant contracts is expressed to be made between Baskan Gida and one or other of the Ferrero Operating Defendants. None of them make any reference to the Banks, and none of them were sent by Ferrero to the Banks. The question therefore arises whether Ferrero thereby made any representations to the Banks in any of these contracts. Since Mr Casale eventually acknowledged in evidence that he expected that all contracts which Ferrero was asked to sign after the completion of the Facility were likely to be used by the Baskans for the obtaining of funds under the Facility, I consider that, exceptionally, these contracts may be taken as documents containing representations not merely to the counterparty, Baskan Gida, but also to the Banks. Furthermore, Mr Strauss sensibly acknowledged that in such circumstances Ferrero incurred a duty of care to the Banks, with regard to material representations in those countersigned contracts.
1032. Basing himself on dicta of Moore-Bick LJ in Kingscroft v. Nissan Fire & Marine (No 2) [1999] Lloyd's Rep 603 at 628, Mr Strauss submitted that the only representations capable of being conjured out of a buyer's signature of a contract for the purchase of goods is that, as at the date stated on the face of the contract, he honestly intends to perform the obligations therein stated, that is to take the goods and pay the contract price, if delivered within the delivery period specified. In my judgment, as a general proposition, that submission is correct.
1033. By contrast, the representations alleged to have been made by Ferrero in the relevant contracts are of the following three types (although not all are alleged to have been made in each relevant contract). The first is that the relevant contract related to new orders being placed for the first time, as opposed to old orders in respect of which deliveries were badly overdue. The second was that the relevant contract misrepresented the true delivery dates applicable to the contract between the parties in respect of the order to which the relevant contract related. The third is that the relevant contract had been entered into in accordance with the method of trading described in the Framework Contract.
1034. Taking each of those alleged representations in turn, it does not seem to me that, in the absence of some special context, the signature by a buyer on a purchase contract bearing a specified date says anything about the date of the purchase order which preceded it, or about the question whether that contract does or does not follow on from, or repeat (possibly on different terms) some earlier unperformed contract in relation to the same specification of goods. It merely states that, whatever may have been the pre-history, the parties thereto consider themselves bound to each other as at the date of the contract, respectively to buy and sell the specified goods on the specified terms.
1035. In my judgment there is nothing in the context of the signing of the relevant contracts to give rise to any other conclusion. The context was only that shortly beforehand, Baskan Gida and the Banks had entered into a Facility Agreement pursuant to which Baskan Gida would be likely to wish to raise money on the strength of contracts entered into after completion of the Facility, so that it might wish to re-make an earlier unperformed contract so as to bring it within the range of contracts pursuant to which, under the Facility, the Banks were prepared to lend. In those circumstances it seems to me quite immaterial whether the apparently new contract followed on from some earlier unperformed contract, provided that, as conventionally represented by the new contract, the parties to it intended, as from the date which it bore, to comply with the obligations therein stated.
1036. As for the second representation, I accept that by signing a written contract containing a stated delivery period, Ferrero represented to the Banks that this was the delivery period contained in the bargain between the parties. Accordingly, if notwithstanding the signed contract, the parties secretly agreed that they would continue to adhere, as between themselves, to some quite different period, then the signed contract might be capable of containing a misrepresentation. The question in each of the cases where a misrepresentation as to delivery period is alleged, is whether following the signing of the relevant contract, any previously agreed different delivery period was regarded by the parties as remaining part of their bargain.
1037. As for the third alleged representation, I can see no basis upon which, or special context as the result of which, any of the relevant contracts contained a representation, express or implied, that they had been concluded in accordance with the Framework Contract. In short, no such representation is made expressly, and there is no basis upon which it could reasonably be implied, in particular because, prior to the date of the earliest relevant contracts, Ferrero had made clear to the Banks that it reserved the right, without their consent, to depart from the Framework Contract in any future dealings with Baskan Gida.
1038. It follows from the above general analysis that the only one of the three misrepresentations alleged which calls for specific treatment, in relation to each group of relevant contracts, is the second, namely mis-statement of delivery periods. In case however a different view might be taken of my analysis of the first alleged misrepresentation, I will deal with the relevant underlying facts, to the extent that I have not already done so.

Contracts 5094 – 7

1039. Contract 5094 replaced Contract 5071, which had been placed in September 2001, and in respect of which only 168mt out of a contract quantity of 360mt had been delivered by 19^th December. The delivery period for the balance expired at the end of October 2001. Contract 5094 provided, in relation to that balance of 192mt, a new delivery period of December 2001 through to March 2002. It referred by number to the original Purchase Order from Ferrero Italy, which had been dated 21^st September 2001.
1040. As between Ferrero and Baskan Gida, I consider that as a matter of contract, Ferrero was after signing Contract 5094 bound to accept delivery of the outstanding balance of 192mt at any time during the December to March period thereby prescribed. However careless Mr Casale may have been in not checking the delivery period, Ferrero Italy could not have been heard to say that it could refuse to accept delivery of any of that balance, upon the basis that the delivery period originally specified in September had already passed. Accordingly, there was in my judgment no mis-statement of the delivery period in Contract 5094 and, since it relates only to a balance outstanding from an earlier contract, it cannot be regarded, objectively, as representing by implication that it was the consequence of some fresh purchase order. The only representation made in it was that, as at 19^th December, Ferrero Italy was obliged to take delivery of 192mt of hazelnuts of the type and quality specified in the contract, if tendered at any time in the period December 2001 to March 2002. That, in my judgment, was contractually correct as between Ferrero and Baskan Gida. The misrepresentation claim in relation to contract 5094 therefore fails on the facts.
1041. The same analysis applies to contract 5095, although the quantities and dates are slightly different. It replaced Contract 5074, made on 1^st October 2001, for delivery of 912mt by the end of that month. By 19^th December, 624mt remained to be delivered and that balance was made the subject of Contract 5095, for delivery during the period December 2001 to February 2002. Again, by signing Contract 5095, Mr Casale committed Ferrero Italy to taking delivery of that balance within the period specified, and it therefore made no misrepresentation.
1042. Contract 5096 had different antecedents. It originated in a Ferrero Italy contract dated 13^th September for delivery of 1,008mt by the end of that month. Having been amended in mid-October, 216mt remained outstanding by 26^th November, when it was re-graded to paste and made the subject of Contract 5084, for delivery by the end of November. 192mt remained outstanding by 19^th December, when it was made the subject of Contract 5096, for delivery between December and February.
1043. Notwithstanding those different antecedents, the underlying analysis is in my judgment the same. It was a contract for an outstanding balance undelivered under previous contracts, and after 19^th December Ferrero Italy could not have refused delivery or payment of that balance if tendered between December 2001 and February 2002. Contract 5096 therefore made no relevant misrepresentations.
1044. Contract 5097 had antecedents similar to Contract 5096 save that, again, the amounts and precise dates were different. Its progenitor was Contract 5063 made on 9^th September for delivery of 1,008mt to Ferrero Italy by the end of that month. On 26^th November the outstanding balance of 480mt was re-graded to paste under Contract 5085, for delivery by the end of that month and the balance of 216mt was the subject of Contract 5097, for delivery between December 2001 and February 2002. Again, Ferrero Italy could not have refused to take delivery during that period, and accordingly Contract 5097 made no relevant misrepresentations.

Contracts 5088 and 5089

1045. These contracts were both countersigned, by Ferrero Germany and France respectively at Mr Casale's direction on 16^th and 17^th January 2002. Contract 5089 is the subject matter of the Banks' allegation of forgery by Mr Casale, which I have rejected. They were two of the three Early December Forward Contracts, both dated 4^th December 2001, but signed shortly thereafter, for delivery in each case of 2,016mt. Unlike contracts 5094-7, these were not re-numbered contracts for outstanding balances due under earlier contracts which had expired, but simply copies of the original December contracts with certain alterations. The contract numbers, dates, quantities, specification and price all remain the same. The alterations relied upon by the Banks related to the delivery period. For Contract 5088 it was originally "during March 2002/April 2002 from origin" and was changed to "from December 2001 through March 2002". For Contract 5089 it was originally, as to 600mt during April 2002, 720mt during May 2002 and 696mt during June 2002 from origin. It was altered, in the version actually signed, to "from December 2001 through June 2002".
1046. The relevance of those changes to the delivery periods under these two contracts is that the delivery periods originally agreed were both so far forward that they would not qualify until after the elapse of some considerable time for inclusion as the basis for applications to draw down funds under the Facility.
1047. The Banks' case is that, notwithstanding those changes, the originally agreed delivery periods nonetheless remained binding as between Baskan Gida, Ferrero Germany and Ferrero France respectively. In support of that allegation, the Banks pointed to the fact that when the subject matter of these two contracts was reordered from Aksu Gida, the original delivery periods were reinstated.
1048. The question remains however whether Ferrero France and Ferrero Germany respectively could have refused delivery of the subject matter of Contracts 5088 and 5089 during the altered periods specified in those versions of contracts signed on 16^th January. They were in fact signed not by Mr Casale but by the respective responsible officers of the two operating companies, on Mr Casale's instructions and, on the basis that there survives no allegation of dishonesty against either of them, the natural consequence is that they thereby committed their respective companies to taking delivery of the subject matter of those two contracts within the altered periods specified. It is as a practical matter inconceivable that either company would have refused delivery within the earlier period rendered permissible by the alteration, and the altered contracts did not in any event compel Baskan Gida to make delivery at any time earlier than the last day of the altered period under each contract. Nor would the Banks, if acting prudently in accordance with the terms of the Facility, have lent money on either of these contracts unless and until supported by pro-forma invoices in respect of deliveries planned within the 45 day period between advance and repayment specified by the Facility Agreement.
1049. It follows in my judgment that there was not in fact any misrepresentation in Contracts 5088 and 5089 of the contractual provision as to delivery in force from 16^th January onwards between Baskan Gida and the relevant Ferrero operating companies, in the absence of any evidence of an agreement between them that the altered periods should not be of contractual effect, and there is no such evidence.
1050. The other misrepresentation alleged in relation to these two contracts is that they had been made in accordance with the method of trading described in the Framework Contract. For the reasons which I have already given, no such representation was made.

Contracts 5100 and 5101

1051. These were two Ferrero Germany contracts, signed by Mr Bolowich on 28^th January 2002 in respect of the outstanding balances due under two earlier contracts, namely 5072 and 5076. To that extent they were similar to Contracts 5094-7, save that in this case, neither of the delivery periods had yet expired. The delivery period under Contract 5100 was about to expire, at the end of January, and the period specified in each of the new contracts was from January to March 2002.
1052. I have already concluded in my findings of fact that there can have been no dishonest intent in the making of fresh contracts for the balances outstanding under Contracts 5072 and 5076, with revised delivery periods, because in fact almost all the whole of those balances was delivered by Baskan Gida within the revised periods.
1053. Again, as with Contracts 5094 to 5097, Ferrero Germany could not have refused to take delivery within the revised period of the subject matter of Contracts 5100-1 once Mr Bolowich had signed them, so that neither of them contained any misrepresentation of the true contractual position between the parties. Again, it is in my judgment neither here nor there (if it be the case) that Mr Casale instructed Mr Bolowich to sign these two contracts without addressing his mind to the fact that the delivery dates had been changed. Baskan Gida desired that change, Ferrero Germany agreed to it, and that change became embodied in their contract.

Contract 5104

1054. This was one of the Early February Forward Contracts and the only misrepresentation alleged is that it had been made in accordance with the method of trading described in the Framework Contract. For the reasons already given, I have concluded that there was no such misrepresentation.
1055. My review of the alleged misrepresentations in countersigned contracts thus produces a nil return for the Banks. For completeness, I should add that even if I had been minded to think that there had been misrepresentations (and that they had been made fraudulently or negligently) there was no evidence whatsoever from the Banks that they either induced any detrimental conduct, or were relied upon. In deciding to lend money on the strength of those contracts, the Banks appear to have been entirely unconcerned whether the contracts were the result of entirely new purchase orders, or rather for the balance outstanding under earlier contracts, whether the delivery periods were different from periods earlier agreed, or whether they had or had not been made in conformity with the pattern of trading described in the Framework Contract. For that reason, these misrepresentations claims would have failed to give rise to any liability, even if one or more misrepresentations had been identified.

The Negligent Representation as to Timely Delivery in the December Letter

Reliance

1056. The Banks' elaborate misrepresentation claims have thus given rise only to one misrepresentation which, because I have concluded that it was made negligently, gives rise to a potential liability for Ferrero. I need therefore to follow through the consequential stages in the analysis of the cause of action in negligence only in relation to the misrepresentation that, as at 13^th December 2001, Baskan Gida made timely deliveries.
1057. A misrepresentation as to timeliness of delivery in a trade reference given for an exporter to a bank contemplating the provision of trade finance is on the face of it, material and in principle likely to be relied upon. That is, if for no other reason, because a reputation for tardy deliveries, in relation to contemplated trade finance where, (as Ferrero knew) the Banks contemplated receiving payment from the off-taker would be likely to disrupt the process of timely repayment of monies lent, and be an adverse factor in the balance for or against making trade finance available.
1058. Nonetheless, there is a real issue as to whether the Banks in fact relied upon the December Trade Reference, even if, which is also challenged, they received it or had its receipt reported to them before committing themselves to the Facility. I have considered the issue of receipt in my findings of fact relating to the relevant period, and concluded on a narrow balance of probabilities that Mr Byles was told before the Banks committed to the Facility that the 13^th December had been received from Ferrero, on the basis that Lovells would not otherwise have confirmed that the condition precedent had been satisfied.
1059. The absence of any contemporaneous copy of that letter in the Banks' files, or any reference to it during the internal audit process which followed the discovery of the fraud hardly suggests that it was regarded within the Banks of an item of much significance, before it came to be used for the purposes of this litigation. Nonetheless I am just satisfied that the Banks relied upon the trade reference in the December Letter as a whole, in the sense that they required to be satisfied that, from Ferrero's perspective, nothing that had happened since the July Letter would undermine it, by December.
1060. Mr Strauss submitted that the reference to timeliness was, on a fair reading of both letters, something of an afterthought in a sentence primarily about quality. He also obtained the admission from Mr Lewis that, even if Ferrero had qualified its trade reference by saying that it believed that Baskan Gida was in financial difficulty, but that if the Banks lent to it, Ferrero would be happy to continue trading with it, then the Banks would probably have lent. The implication from that exchange is that the Banks were looking to Ferrero's Trade Reference as an indication that it was happy to continue trading with Baskan Gida, rather than looking to each individual part of it, as a separate factor relevant to the decision whether or not to lend.
1061. There is therefore evidence to support a conclusion that, although the Banks relied on the December Trade Reference, the precise truth of every small part of it was not material to their decision whether or not to lend, so long as they had the comfort that Ferrero intended to continue trading with Baskan Gida, as undoubtedly it did.
1062. The question whether the Banks can be said to have relied upon the representation as to timeliness in the December Letter is complicated by the fact that it was the second of two, and that the representation was true when originally made in July, but untrue when repeated in December. If one asks whether it would have made any difference to the Banks' decision if there had been no reference to timeliness in either of the two Letters, but merely a truthful reference to quality in both, the answer is, as a matter of factual inference, that it probably would not. Conversely, if the question is whether the removal from the December Letter of an earlier plaudit about timeliness in the July Letter would have made any difference, a careful comparison between the two would probably have set alarm bells ringing, if only to a muted extent. The application of the test "did the claimant have the misrepresentation actively in his mind when making his decision" would thus produce a negative answer under the first analysis, but possibly a positive answer under the second.
1063. Not without some hesitation, I have come to the conclusion that, in committing itself to the Facility, the Banks relied upon the clear and truthful statement in the December Letter that Ferrero intended to continue trading with Baskan Gida at the stated 30,000mt level, but that the statement that its deliveries were made on a timely basis was not actively present in the Banks' corporate minds, because its absence would not have undermined the substantial truth of the positive trade reference which they thereby received.
1064. Against the risk that a higher court may take a different view on the question of reliance, I shall set out my analysis of the remaining stages of the cause of action in negligence.

The BBL Principle

1065. I consider that this principle presents an insuperable barrier to the Banks in relation to any cause of action based upon a negligent statement by Ferrero as to the timeliness of Baskan Gida's deliveries. This being a paradigm case of the provision of information rather than advice as to the merits of the Banks' contemplated transaction, the question is: would the Banks have suffered the same loss if the representation as to timeliness had been true? The Banks' loss was caused not by any tardiness in the making by Baskan Gida of deliveries of hazelnuts to Ferrero, but by the Baskans' commission of a series of deliberate premeditated frauds. They began with the invoicing and trust receipt fraud, included fraudulent trading, and were completed by the transfer of Baskan Gida's remaining assets (other than its freehold land) to a front company masquerading as independent from the Baskans, by means of a preference which was dishonestly concealed by the false creation of an impression of creditor pressure.
1066. None of that fraudulent course of conduct had anything to do with Baskan Gida's record prior to 13^th December for timely or tardy deliveries of hazelnuts to Ferrero, save in the loose sense that the Baskans' financial problems may have been a common cause of both Baskan Gida's delivery failures and of their decision to embark upon defrauding the Banks. Mr Wardell tried hard to persuade me that, in addition to that common cause, the truth about Baskan Gida's delivery performance would have revealed the company's financial difficulties as its underlying cause, and thereby presented to the Banks a picture of a company inherently more likely to be affected by management fraud, than would a company in financial health.
1067. I am not persuaded that, either in principle or upon analysis of the facts, that submission affords an answer to the difficulties presented by the application of the BBL principle to this case. First, the question is not whether the Banks would have been put on inquiry as to a risk of fraud if Ferrero had told the truth, but rather whether if Baskan Gida had maintained a satisfactory delivery record against contractual obligations during the two months prior to 13^th December, the fraud on the Banks would not otherwise have occurred. I regard that as an inherently unlikely analysis.
1068. Secondly, the evidence does not establish that Ferrero knew that Baskan Gida's deteriorating delivery performance in the period immediately prior to the 13^th December 2001 was attributable to financial difficulties, still less difficulties of a gravity which would materially affect the propensity of its owners and management to resort to fraud. In this context it is important to bear in mind that the negligence proved consisted not of any failure by Ferrero to conduct some deeper investigation into the underlying causes of Baskan Gida's deteriorating delivery performance, but in Mr Casale's failure to appreciate that the statement in the December Letter as to timeliness had an objective meaning other than that which he intended to convey. For those reasons, the Banks' limited case in negligent misrepresentation would fall at the BBL hurdle, even if it were to survive the application of the test of reliance.

Causation – Novus Actus Interveniens

1069. Contrary to Ferrero's submissions, I am not persuaded that the fraud which caused the Banks' loss was a novus actus. Although its perpetration followed the grant of the facility, it was at least in outline planned in advance, in and after November 2001, so as to be ready for implementation almost as soon as the ink dried on the Facility Agreement. It follows that if the other elements of a claim in negligent mis-statement had been established, so that the Banks' entry into the Facility had been reliant upon the negligent mis-statement as to timeliness, and the liability had fallen within the scope of the responsibility assumed by Ferrero in giving the December Trade Reference, then a case in causation would have been established.

Remoteness

1070. I consider that if, for some reason, the BBL principle had not been a fatal obstacle to the Banks' case, it would nonetheless have failed for remoteness. Whatever the precise test applicable in the circumstances, it seems to me that it was not reasonably foreseeable to Ferrero that a mis-statement as to the timeliness of Baskan Gida's deliveries of hazelnuts would lead to a loss attributable to the premeditated frauds which I have described. Since the BBL principle and the rules to remoteness are so closely interrelated, I need say nothing more about that issue.

Contributory Negligence

1071. On the findings of fact and the analysis which I have so far made and applied, the question of contributory negligence does not arise. Yet again however, it is a fact intensive issue which might arise if a higher court took a different view about the true analysis of the Banks' negligence claim, and its resolution is dependent in part upon the weight of oral evidence of witnesses cross-examined at length, and upon the relative weight to be attributed to the different opinions of the parties' banking experts. Its outcome is therefore dependent upon the trial process in a more fundamental way than the issue as to the destination of the hazelnuts released from Warehouse 2, and if the contributory negligence question were to matter, it would require a re-trial of a part of this case which lasted several weeks before me. I consider it therefore appropriate, even if not strictly necessary, to deal with the contributory negligence issues, upon the hypothesis that the Banks have, contrary to my view, established a vicarious liability of Ferrero Industrial for Mr Casale's negligent mis-statement in the December Letter.
1072. Section 1(1) of the Law Reform (Contributory Negligence) Act 1945 provides:

"Where any person suffers damage as the result partly of his own fault and partly of the fault of any other person or persons… the damages recoverable in respect thereof shall be reduced to such extent that the court thinks just and equitable having regard to the claimant's share in the responsibility for the damage…"

1073. In applying this test the court must have regard both to blameworthiness and to what has been called causal potency: see Arab Bank plc v. John D Wood [2000] 1 WLR 857, Corr v. IBC Vehicles Ltd [2008] UKHL 13 at paragraph 44, and Calvert v. William Hill [2008] EWCA Civ 1427 at paragraph 69. In a case where the claimant's conduct is both blameworthy and causally potent, the court must have regard to the respective responsibility of each of the claimant and the defendant (or defendants as a group if more than one) for the claimant's loss: see Clerk and Lindsell (op. cit.) at paragraph 3-70.
1074. That comparative exercise is rather artificial in the present case, since I have already concluded that, due to lack of reliance, Ferrero Industrial's negligent mis-statement did not cause the loss, and because, applying the BBL principle, Ferrero Industrial did not incur responsibility to the Banks for the loss actually suffered. The apportionment required by the application of section 1(1) of the 1945 Act requires me to assume hypothetically, contrary to my conclusion, that Ferrero Industrial both caused and incurred legal responsibility for the Banks' loss, but that hypothetical approach provides little assistance in any assessment of Ferrero's proportionate responsibility, causal potency, or blameworthiness. The best I can do is to identify those blameworthy respects in which the Banks caused their own loss, to identify the degree of blame which ought to be attached, and the causal potency of that conduct in absolute rather than relative terms, and then leave it to any higher court which might form a different view about Ferrero's responsibility to carry out the necessary apportionment.
1075. Ferrero's allegations of contributory negligence by the Banks were pleaded under four main headings, twenty three sub-headings, and occupied no less than forty pages of their Re-Re-Re-Amended Defence. Examination of the Banks' witnesses of fact, whose evidence was overwhelmingly directed to the contributory negligence issues, took up almost the whole of November 2008. The cross-examination of the banking experts took a further three days in January 2009.
1076. In the end, and largely due to the very sensible narrowing of their differences of opinion by the banking experts during joint meetings and telephone discussions, Mr Strauss narrowed Ferrero's attack in his closing written submissions to four items, two relating to due diligence failures before money was advanced under the Facility and two relating mainly to failures thereafter. Mercifully, the parties confined themselves to written closing submissions on these issues. I shall deal with each of Ferrero's four surviving points in turn.

Audited Accounts

1077. Both Banks' credit committees made it a condition of the grant of the Facility that they should each be provided with Baskan Gida's most recent audited accounts. Furthermore, the Facility Agreement provided, at clause 26(t) and 27.1(b), that failure by Baskan Gida at any time to maintain a minimum net worth of US$3 million was to be an Event of Default. Under clause 27.1(h)(ii) an Event of Default was also constituted if the value of Baskan Gida's assets was less than its liabilities. The occurrence of an Event of Default entitled the Banks both to refuse to make any or any further lending, and to call in any loan already made.
1078. The Banks made it clear to Baskan Gida that the Facility would not proceed without the provision of its latest audited accounts, and on 14^th October 2001 PRI sent to KBC a certified translation of Baskan Gida's financial statements and auditor's report for the year ending 30^th June 2001. It appears likely that BTM did not receive a copy of those accounts until after the grant of the Facility, but they were received on 20^th December 2001, before any money was advanced to Baskan Gida.
1079. The audited accounts showed that it was the opinion of Baskan Gida's auditor (a Mr Ahmet Çakir of the Ankara Chamber of Chartered Financial Consultants) that, as at 30^th June 2001, Baskan Gida had a deficiency of assets against liabilities of some 9.6 trillion Turkish Lira, (approximately US$7.5 million), and a net loss for the year of some 14.3 trillion Turkish Lira.
1080. Ferrero's case is that the Banks either failed to read the audited accounts, or failed to appreciate that this was the auditor's opinion, thereby failed to realise that, far from having a minimum net worth in excess of US$3 million, Baskan Gida was in fact balance sheet insolvent and that, had the Banks realised this, the Facility would never have been granted. Alternatively in relation to BTM, Ferrero claims that had it read and understood the audited accounts received on 20^th December, no money would have been advanced under the Facility, even though by then granted.
1081. The Banks' response was one of confession and avoidance. For KBC, Mr Lewis and Mr van Broekhoven said that the audited accounts had been given a cursory review, which did not reveal that the auditor's opinion was as I have described. For BTM, Mr Byles accepted that there had been a failure to press for the audited accounts before the grant of the Facility, and a failure to study them once received.
1082. More generally, the Banks advanced a strenuous defence along the following lines. First, in structured trade finance the financial position of the borrower is of limited importance, reliance being placed on repayment from the off-taker (here Ferrero), and that had the audited accounts been studied with sufficient care, it would have been ascertained with the assistance of further inquiry that the apparent net asset deficiency and trading loss were in fact the illusory consequence of the application of exchange differences to accounts in Turkish Lira, occasioned by the devaluation of the Lira against the US dollar, with no substantive adverse consequence for Baskan Gida's cash flow, since it both sold its merchandise for, and borrowed in, US dollars. Accordingly, a careful appreciation of the implications of the audited accounts would not have led to the Facility being refused or to the calling of an Event of Default once granted.
1083. In my judgment, the Banks' failure properly to study the audited accounts, and BTM's failure even to call for them before committing itself to the Facility, was seriously blameworthy. As to causative potency, I consider that, had the audited accounts been studied properly, so as to bring home to the Banks the auditor's opinion, the Facility would not have proceeded unless the Banks had on further inquiry, both of Baskan Gida and independent Turkish accountants, satisfied themselves that the apparent balance sheet deficiency and trading loss represented no real threat to Baskan Gida's ability to continue trading throughout the period contemplated by the Facility, i.e. for the rest of the 2001 season. Since no such inquiries were made, and the evidence included no qualified expert opinion as to what their outcome might have been, it is a matter of speculation whether the Banks would, on further inquiry, have been so satisfied. In my judgment, and having regard to the fact that by then Baskan Gida was insolvent, and to the less than perfect security structure available, I consider that the prospect that the Banks would have been satisfied is slightly less than even. My reasons follow.
1084. Despite Mr Lewis's doctrinaire attempt to suggest first that in a trade finance facility the borrower's financial position hardly matters, and secondly that the audited accounts did not in terms, or with sufficient clarity, disclose a deficiency or a loss, I am satisfied that it was the duty of the Banks, in their own interest as intending lenders, properly to study the audited accounts, rather than merely to skim-read them, and that a proper study would have revealed that the auditor's opinion was as I have described.
1085. The audited accounts were undoubtedly presented in a manner which is unfamiliar to an English trained eye. They began by presenting the accounts as determined by Baskan Gida itself, which indeed showed both a modest net asset surplus and a very small net profit. Each page of the company's version of the accounts had at its foot a signed stamp, but that on any careful reading was no more than the stamp of the official translator. The document ended, under a section headed "RESULT" (to which reference is made on the front page under the heading "CONCLUSION"), with statements by the auditor that the accounts as settled by the company did not conform with the General Directives of Accounting Applications and Accounting Standards (applicable in Turkey), and reference was then made to tables showing what, if adjusted so as to conform with those standards, the relevant figures should have been. Those tables were contained in an earlier section headed "ACCOUNTING CONTROL", clearly identified in the Result section, and were preceded by a more detailed explanation of the reasons for the adjustments which the auditor considered it necessary to make. It is readily apparent from the Accounting Control section of the accounts that the main reason for the downward adjustments which the auditor regarded as necessary was the inappropriate treatment in the company's accounts of financial expenditure, including exchange losses incurred on foreign currency loans.
1086. Mr van Broekhoven accepted, much more readily than Mr Lewis, that a reasonably careful study of the audited accounts would have revealed that Baskan Gida's auditor had required those downward adjustments, such that in his opinion the company had a negative net worth, and had incurred a trading loss. Notwithstanding his evident sympathy for the Banks, having to deal with accounts in an unfamiliar foreign form, Mr Bryant also readily accepted that the Banks' failure to appreciate that the audited accounts did show negative net equity and a trading loss in the auditor's opinion was negligent.
1087. Turning to the question of causative potency, there is much force in the Banks' case, ably supported by Mr Lewis, Mr Graham and, but to a lesser extent, Mr Bryant, that in structured trade financing of this type, the finance position of the borrower does not go directly to the credit risk inherent in the transaction where, as here, reliance for payment is placed on the financial strength of the off-taker. Nonetheless, a serious financial weakness in the borrower creates the operational risk that it may go out of business during the currency of the Facility, thereby bringing to a halt the stream of merchandise, the delivery of which triggers the off-taker's payment liability. If the music stops in that way during the currency of the facility, it is almost inevitable that the lender will suffer a loss, unless fully secured at all times, which in the present case, the Banks were not, even in the absence of fraud.
1088. Thus the question posed to any prudent bank by a proper reading of the audited accounts was whether the auditor's gloomy opinion as to balance sheet solvency and profitability really did call into question Baskan Gida's ability to trade during the intended period of the Facility, or seriously undermined the presentation already given (largely by PRI) of Baskan Gida's rather tenuous financial health, based on its earlier audited and management accounts.
1089. The Banks' witnesses went to considerable length to demonstrate (and Mr Rex to similar length to refute) a thesis that, properly understood, the auditor's adjustments to otherwise just about satisfactory statements of net assets and profits were simply the consequence of the strict application of Turkish accountancy standards to exchange differences in relation to a company which was in reality trading and borrowing in dollars, without any real adverse consequence for Baskan Gida's cash flow or, therefore, its ability to trade. Since the necessary expert advice from an independent Turkish accountant had not been obtained, even by the time of trial, and since none of the Banks' witnesses, including Mr Bryant, however distinguished, had the necessary specialist Turkish accounting expertise, I was provided with no reliable information upon which to assess the strength or weakness of that analysis. It certainly sounded persuasive in the mouth of Mr Graham, who was in all respects a most impressive witness, but even he acknowledged that the outcome of any sufficiently diligent inquiry as to the implications of the auditor's opinion had to be a matter of speculation. He readily acknowledged that, in the absence of a satisfactory outcome to that inquiry, after a proper study of the audited accounts, the Facility would not have proceeded.
1090. The strength of the case that the auditor's opinion would not, after inquiry have been adjudged as destructive of Baskan Gida's commercial solvency lies in the fact that the adjustments were heavily dependent upon improper treatment of financial expenses, including currency losses, at a time of severe devaluation of the Turkish Lira against the US dollar, in relation to a company which both traded and borrowed in dollars, with a cash flow prima facie unlikely to be adversely affected by a devaluation of the local currency in which its audited accounts had to be expressed. The weakness of that case arises from the undoubted fact that Baskan Gida was by then commercially, if not hopelessly, insolvent, having by October 2001 resorted to defrauding its Turkish banks. It seems to me that, consistent with Mr Graham's own view, it would be wrong to treat the likelihood of a satisfactory outcome to that inquiry as being more than speculative.

The 30% Point

1091. The essence of the security structure under the proposed Facility was that the Banks would lend 100% of the cost price to Baskan Gida of hazelnuts required for delivery for Ferrero, subject to a cap of 90% of the amount due from Ferrero under associated purchase orders: see clause 5.1(g) of the Facility Agreement. Thus the Banks' primary security was to consist of hazelnuts in Warehouse 2 with a cost price approximately equivalent to the amount advanced followed, during the short period necessary for processing, by a trust of the nuts released for that purpose. The Master Trust Receipt provided in terms that the nuts would be sold only to a buyer to which Baskan Gida was not indebted, with the entire proceeds of such a sale being paid immediately to the Banks, or (if Baskan Gida was indebted) only on terms that the buyer would pay the Banks directly without deduction or setoff. The intent was that the nuts would only be sold to Ferrero, as is evidenced by the fact that upon any release, Baskan Gida had to provide pro-forma invoices to Ferrero "which must domicile payment to the Collection Account": see Schedule 10 to the Facility Agreement, setting out the operating procedures.
1092. The Facility structure therefore contemplated that the two successive forms of security (pledge and trust receipt) would bite in relation to the whole of a consignment of hazelnuts purchased, deposited in Warehouse 2 and then released for processing, in order to satisfy any particular Ferrero purchase order. But on about 12^th September 2001, Mr Byles learned that, on average, only 70% of any consignment purchased by Baskan Gida and pledged to the Banks would be of sufficient quality for on-sale to Ferrero, and that the remaining 30% would have to be swapped or sold in the market by Baskan Gida in order to obtain replacement nuts (or the money to buy replacement nuts) of sufficient quality to complete the Ferrero order. The 30% would therefore fall out of the security structure and, since they would be sold or swapped on terms which did not lead to immediate payment of the proceeds of sale to the Banks, would perforce have to be dealt with otherwise than in accordance with the Master Trust Receipt. Mr Byles' manuscript notes on the email which first conveyed this information leaves no doubt that he understood how, in practice, the 30% was to be dealt with.
1093. Mr Byles discussed this information with Mr Lewis, and they spoke to Melih Baskan about it. Mr Lewis could not remember that conversation, but Mr Byles said that Melih Baskan told him that this procedure had been operating for years, without adverse consequences, and he came away personally satisfied that the Baskans were sufficiently expert traders to deal with the 30% problem in a way that did not adversely the Baskans' profit margin or the Banks' security.
1094. It is clear from the evidence of Mr van Broekhoven and Mr Graham that neither Mr Byles (at BTM) nor Mr Lewis (at KBC) took the 30% point any further. Mr van Broekhoven did not learn of it until he read Mr Rex's expert report, and when he did so he said that "I couldn't believe my eyes there, to be honest." (day 17 page 43). Mr Graham confirmed that the point was not reported to BTM's credit committee, as it should have been, and that if it had been it might have caused the Facility negotiations to be aborted, if no satisfactory solution could be found.
1095. Ferrero's case is that Mr Byles' and Mr Lewis's failure to raise the 30% issue with their respective credit committees, and the Banks' failure to do anything about the inevitable departure from the security structure which the sale or swapping of the 30% necessarily involved, were both seriously blameworthy. As to causative potency, Mr Strauss submitted if the 30% problem had gone to either credit committee, it would have caused the Facility to be aborted, because of the unacceptable damage which its implications would have done both to the Banks' intended security structure, and to the assumed Baskan Gida profit margin upon which the commerciality of the proposed facility was based.
1096. Apart from some insubstantial pleading points, the Banks' main response, supported by Mr Bryant, was that this was a practical operational matter within the Banks' judgment to accept, that the supposed difficulties with the security structure and profit margins were illusory and that, had the matter been reported to credit committee, modest adjustments to the security structure could have been adopted, such that the Facility would have proceeded, without any additional protection thereby being provided against the Baskans' fraud, which was the real cause of the Banks' loss.
1097. In my judgment, the failure by Mr Byles and Mr Lewis carefully to consider the implications of the 30% point, in particular upon the proposed security structure, or to report the point to their respective credit committees (or, in Mr Lewis's case to Mr van Broekhoven) was blameworthy, and it was not for them simply to accept Melih Baskan's blithe assurances that it would cause no problem, without deeper investigation.
1098. As to causative potency, it seems to me that, viewed on its own, the 30% issue would upon a deeper analysis at credit committee level not have been regarded as a deal-breaker. Rather, adjustments would have been made to the security structure including, probably, a reduction in the loan to value ratio sufficient to give the Banks some protection against the consequences of the Baskans' need to sell or swap sub-standard hazelnuts in order to meet Ferrero's specifications.
1099. Viewed in conjunction with the concerns which would have arisen from a proper appreciation of the audited accounts, I consider it rather more likely that the 30% issue would have contributed to an abandonment of the negotiations. This is because its main impact was upon the adequacy of the Banks' security structure, and I consider it likely that a derogation from the security structure coupled with increased concern about Baskan Gida's commercial solvency would in combination have been fatal. My reasons follow.
1100. As to blameworthiness, the combined effect of the evidence of Mr van Broekhoven, Mr Graham, Mr Rex and (in cross-examination on the point) Mr Bryant was that the 30% problem should have been reported by Mr Lewis and Mr Byles to higher authorities in their respective banks. Furthermore, since the security structure in the facility as eventually agreed was not adjusted to accommodate the problem, the result was that, even in the absence of a fraud, Baskan Gida would have had to embark upon the process of buying, processing and selling hazelnuts in a manner which inevitably conflicted with the terms of the Facility Agreement, and the Master Trust Receipt in particular. Plainly, that ought not to have happened, and the fact that it did inevitably reflects blame upon the Banks in failing fully to appreciate the manner in which the transactions which they intended to finance were on a day to day basis and had for many years been conducted.
1101. As to causal potency, the sheet anchor of Ferrero's case was that, since Baskan Gida's reported domestic sales were insufficient to accommodate Turkish re-sale of 30% of all consignments purchased for Ferrero, the 30% would have had to be exported, on 60 day credit terms, with the consequence that the proceeds of sale to be used to buy further Ferrero standard hazelnuts would not have been available until long after the necessary processing and repayment periods. The evidence as to the way in which the 30% problem was actually presented, in particularly orally by Melih Baskan to Mr Byles and Mr Lewis, together with the application of common-sense, suggests that the 30% problem would not in practice have been addressed by Baskan Gida in that mechanical way. A trader knowing that a 100mt of hazelnuts purchased for processing and resale to Ferrero would leave only 70mt of the requisite specification would not buy a 100 tonnes to meet a 100 tonne order. He would increase his purchase to accommodate the necessary attrition attributable to sub-standard nuts.
1102. I can envisage no insuperable obstacle to accommodating a pattern of trading of that kind by an amendment of the security structure of the type which I have described above, and I consider it likely that, if the 30% problem had been dealt with thoroughly, such an amendment would probably have ensued.
1103. On that footing, if the Facility had proceeded as amended, it would have afforded no additional protection to the Banks against a fraud by the Baskans, although it is theoretically possible that an adjusted loan to value ratio might slightly have reduced the amounts drawn down under the amended Facility.

Failure to Monitor Deliveries and Payments

1104. In relation to sales to Ferrero, the Facility allowed 45 days for repayment, following each specific advance: see clause 12.1(b)(i) of the Facility Agreement. Compliance with this 45 day limit was rigorously monitored by the Banks, and it was in the end Baskan Gida's non-compliance with that limit in relation to repayment of the second and subsequent draw-downs that led to the Banks' discovery of the fraud. Ferrero's case is that the Banks should have, but failed to, put in place a system of monitoring of the progress of nuts from the moment of release from Warehouse 2 until delivery to Ferrero, the progress of the delivery of final (rather than pro-forma) invoices to Ferrero and the progress of Ferrero's payments pursuant to those invoices. It is common ground that the Banks did not do so. This arose in part from the absence of any monitoring provisions to that effect in the Facility documentation, and in part from the fact that, even in the absence of contract, the Banks did nothing internally to put in place procedures which would have enabled that monitoring to take place, once draw-downs started to be made under the Facility. That involves, incidentally, no criticism of those responsible for the day to day process such as Mr Parello and Mr Wenden of BTM, who were simply following orders. Rather, the criticism is that no procedures were put in place which, had they been complied with, would have led to satisfactory monitoring of deliveries and payments to take place.
1105. Ferrero's case is that, had such procedures been put in place and operated diligently, the Banks would have been appraised of a relevant non-payment by Ferrero in respect of deliveries made by about 17^th January 2002, in sufficient time to prevent all but the first three draw-downs under the Facility, which amounted only to just under €4 million in aggregate. Blowing the whistle might, in addition, have yielded a larger amount of secured hazelnuts still in Warehouse 2, but that is speculative.
1106. The Banks' response was, again, one of confession and avoidance. Since Mr Byles (on this occasion free to blame KBC for something which was not part of BTM's responsibility) was trenchantly of the view that the absence of these procedures was the single most important factor in exposing the Banks to the loss which they suffered, and since in cross-examination Mr Lewis accepted, with the benefit of hindsight, that the Banks' focus on 45 days as the relevant period was wrong, they could hardly deny that the failure was blameworthy. Furthermore both Mr Rex and Mr Bryant agreed, in their joint statement, that such monitoring procedures should have been put in place.
1107. The Banks' main points were, first that such monitoring would not have been proof against the Baskans' deliberate fraud, secondly that delays in payment after deliveries would have been explained away by a tissue of initially plausible lies such that, thirdly, the Banks would not have been sufficiently appraised of something having gone wrong to have been able to prevent draw-downs earlier than in February, by which the time the Banks had permitted seven of the eleven draw-downs in the aggregate sum of €13.73 million. The differing dates upon which Ferrero and the Banks submitted that proper monitoring procedures would have enabled the Banks to blow the whistle reflected the opinions of their respective experts.
1108. It is apparent from that summary that the Banks had no answer to blameworthiness or to causal potency, in respect of at least some of their loss, the real issue being how much, and turning on the precise date upon which, during a series of large draw-downs, the Banks would have been in a position to call a halt.
1109. It is of course a matter of speculation what monitoring procedures would have been put in place, and how far they would have afforded reliable protection against a deliberate fraud, and I bear in mind that security structures for business finance are primarily aimed at protecting the lender from the borrower's insolvency, rather than from fraud. Taking a necessarily very broad brush, it seems to me that the absence of proper monitoring procedures may fairly be regarded as having been a cause of approximately half of the Banks' loss.

Failure to Warn Ferrero of the Banks' Rights

1110. Ferrero has throughout the trial made much of the fact that, at no time after Mr Casale informed Mr Byles on 4^th March 2002 that Ferrero was dealing with Aksu Gida, did the Banks warn Ferrero either of their security rights in relation to hazelnuts, or of their belief that the Baskans had been engaged in a fraud. The point is relied upon not merely in relation to contributory negligence, but in relation to conversion and knowing receipt, as well as a response to the Banks' criticism of Ferrero's conduct during the same period.
1111. The relationship of the point to contributory negligence is in my judgment slight, mainly because the Banks had by 4^th March suffered substantially the whole of their loss. The security structure was such that the Banks retained no effective proprietary rights in relation to the hazelnuts released from Warehouse 2 so that the prospect, even if Ferrero had been dissuaded from buying from Aksu Gida after 4^th March, that the Banks rather than some other Turkish creditors or purchasers would have obtained them, seems to me to be verging on the fanciful, as indeed Mr Byles evidently thought at the time. His evidence was that there seemed to him to be no useful point in notifying Ferrero of the Banks' security rights on 4^th March, since in his view, the nuts were either still in Warehouse 2 or had been seized by Yapi Kredi at the end of the previous week. Although he may well have been wrong about that, it is difficult to criticise his judgment, made at a time when the Banks had minimal information about what was going on, as blameworthy. Even if it was, I consider that the improbability that such a warning to Ferrero would have been of any real use to the Banks means that its causative potency was slight. As a matter of analysis, the point might better have been dealt with under the heading failure to mitigate than contributory negligence, but even there, I consider that it would be of minimal effect.

Contributory negligence conclusions

1112. The outcome of the foregoing analysis is that the Banks were in my judgment blameworthy in respect of three out of the four surviving items in Ferrero's list, and that in respect of two of them, namely the audited accounts and the monitoring procedures, each failure was, independently of the other, of real causative potency. Viewed separately, the failure properly to deal with the 30% problem probably was not, but in conjunction with the failure to appreciate the implications of the audited accounts, it adds a little to a case that, in the absence of contributory negligence, the Facility would not have proceeded.
1113. I return to an overall consideration of the Banks' responsibility for their loss, bearing in mind that the primary cause was a premeditated fraud, against which facility structures and due diligence are generally inadequate protection, and bearing in mind that the elements of blameworthy failure which have prevailed stand as isolated exceptions in an otherwise thorough and detailed due diligence carried out over a lengthy period.
1114. Viewed in that way, the Banks' failures come nowhere near to constituting such a predominant cause of their loss as to displace any responsibility which might otherwise have attached to Ferrero. On the other hand, they come much nearer to the heart of the matter than a negligent assertion by Ferrero that Baskan Gida deliveries were timely, which comes nowhere near it at all. Abjuring as I do any deeper comparative analysis, because of the hypothetical difficulties to which it gives rise, I say only that a serious but not overwhelming case of contributory negligence has been established against the Banks.

CONCLUSIONS

1115. The outcome of this very long and no doubt enormously expensive litigation is therefore as follows:

i) The Banks fail in their claims against Ferrero in conspiracy, in deceit, in conversion and in knowing receipt.

ii) The Banks have established that Ferrero Industrial made, through Mr Casale, a negligent mis-statement that Baskan Gida's deliveries were, viewed against their contractual obligations, timely. Nonetheless a claim in negligent mis-statement against Ferrero fails on grounds relating to reliance, the BBL principle and remoteness.

iii) Had I been persuaded that the Banks had established a negligent mis-statement claim against Ferrero Industrial, a substantial reduction of their claim would have been necessary on account of their contributory negligence.

iv) The Banks succeed in their conspiracy claim against Mr Abidali, but only in respect of part of their loss.

v) I direct that there be an inquiry on documents and submissions, but not involving oral evidence, as to the quantification of Mr Abidali's responsibility for the Banks' loss.

vi) I will hear counsel on directions in relation to that inquiry, and upon any other matters arising.