Final Appeal No 16-2009 (Civil) Iinternational Cases [2010] Part 10 Case 4 [CFA]



Kasikornbank Public Co Ltd

- vs -

Akai Holdings Ltd

(in liquidation)








Chief Justice Ma

  1. I agree with the judgment of Lord Neuberger of Abbotsbury NPJ.

    Justice Bokhary PJ

  2. I agree with the judgment of Lord Neuberger of Abbotsbury NPJ.

    Justice Chan PJ

  3. I agree with the judgment of Lord Neuberger of Abbotsbury NPJ.

    Justice Ribeiro PJ

  4. I agree with the judgment of Lord Neuberger of Abbotsbury NPJ.

    Lord Neuberger of Abbotsbury NPJ


  5. On 4 December 1998, Mr James Ting (“Mr Ting”), the Chief Executive Officer of a company then called Semi-Tech Global Company Limited, and subsequently renamed Akai Holdings Limited (and to which I shall refer as “Akai”), purportedly acting on behalf of Akai, executed a number of documents giving rise to obligations on the part of Akai in favour of Thai Farmers Bank (“the Bank”). On 7 December 1998, in accordance with the documents executed three days earlier, Mr Ting, again purportedly acting for Akai, borrowed US$30m from the Bank, authorised it being used to pay off the liability to the Bank of The Singer Company NV (“Singer NV”), and lodged certificates in respect of the 56m shares (“the pledged shares”) in a subsidiary of Akai called Akai Electric Co. Ltd (“Akai Electric”) with the Bank, as security for the loan.

  6. This transaction was referred to in argument as “the Switch Transaction”, an expression which I will adopt. The effect of the transaction was to create a liability on Akai to repay the Bank within a year the sum of US$30m plus interest, which had previously been the liability of Singer NV and to give the Bank security in the form of 56m shares owned by Akai in Akai Electric.

  7. A few months thereafter, pursuant to a request from Akai, made on the ground that the market value of the pledged shares had increased, the Bank released some 5.5m of the shares (“the 5.5m shares”), retaining the balance of the pledged shares (“the Shares”). Akai failed to repay the loan when it fell due, and, a short time thereafter, Akai went into liquidation. Thereafter, the Bank sold the Shares for some US$20m, retained that sum to pay off part of the loan, and proved in Akai’s liquidation for the balance.

  8. At the end of 2004, the liquidators of Akai brought the present proceedings against the Bank, essentially based on the proposition that the Bank realised, must have realised, or ought to have realised, that Mr Ting had no power to commit Akai to the Switch Transaction, and that the Bank was accordingly liable to pay compensation to Akai. Stone J dismissed the claim, but the Court of Appeal overturned his decision, and awarded Akai compensation in the sum of some $22m plus interest. The Bank now appeals, contending that Stone J’s order dismissing Akai’s claim should be restored; Akai cross-appeals contending that it is entitled to substantially more compensation than that awarded by the Court of Appeal.

  9. The issues on this appeal, pared to the bone, are two-fold:

    1. Did Mr Ting have apparent authority to commit Akai to the Switch Transaction?

    2. If he did not, then, in the light of Akai’s alternative claim based on knowing receipt, is the compensation payable to Akai limited to the common law measure of damages?

  10. The Bank contends that the answer to both questions is in the affirmative. On the first question, the Bank has realistically abandoned its contention, unsuccessfully maintained below, that Mr Ting had actual authority to enter into the Switch Transaction: hence the issue in this court is limited to apparent authority. On the second question, it is agreed that, in common law, the measure of damages would be based on the value of the Shares at the date they were sold by the Bank, but Akai contends that it has a claim against the Bank for knowing receipt of the pledged shares, which entitles it to claim equitable compensation, which would result in a substantially higher sum.

  11. I turn first to set out the undisputed facts in a little more detail, and will then deal with the two questions identified above, and a small point on interest.

    The facts giving rise to this claim

    The companies involved in this case

  12. During late 1997, many countries in South-East Asia were adversely affected by the so-called Asian Financial Crisis. As the third largest financial institution in Thailand, the Bank was hit by what Stone J referred to as “severe liquidity problems” owing to the Crisis and consequent substantial non-performing loans. The Bank’s entire profit for the year ended 31 December 1997 was only US$17 million, and, in order to survive, it undertook a share issue of US$857 million during April and May of 1998.

  13. One of the Bank’s many non-performing loans had been made to Singer NV, with whom the Bank had often had dealings through Mr Ting. Although it subsequently transpired that he had been dishonest, as at 1998 Mr Ting was a much respected, and apparently very successful, businessman, with substantial interests in, de facto control over, and effective management of, a number of what were then at least substantial companies.

  14. One of those companies was Akai, at that time a public company incorporated in Bermuda and carrying on business from Hong Kong. Akai owned just over 70% of the shares in Akai Electric, a Japanese company, which was Akai’s main operating subsidiary. Mr Ting was its executive chairman and chief executive officer. Its shares were listed on the main board of the Hong Kong Stock Exchange, with the majority of the issued shares being held by members of the public, while 43% were owned by Semi-Tech Corporation (“STC Canada”).

  15. STC Canada was a company listed on the Toronto Stock Exchange, and owned as to 45% by Mr Ting and 55% by the public; Mr Ting was its chairman, president and chief executive officer. He owned a very substantial stake in STC Canada, as did (albeit to a lesser extent) his former University teacher, Dr Holmes, who was also a director of STC Canada and of Akai. Unlike Mr Ting, Dr Holmes gave evidence at the trial.

  16. STC Canada owned 50% of the shares in Singer NV, a company incorporated in the Netherlands Antilles and listed on the New York Stock Exchange. The remaining 50% of Singer NV shares were held by members of the public. Mr Ting was also the chairman and a director of Singer NV.

  17. Singer NV indirectly owned 48% of Singer Thailand Public Company Limited (“Singer Thailand”), a company listed on the Stock Exchange of Thailand. The Bank owned 8.5% of Singer Thailand’s issued shares, and the Bank’s chairman, Mr Banyong Lamsam was on the board of Singer Thailand (as its chairman), together with Mr Ting.

    The background to the Switch Transaction

  18. Since about 1992, Singer NV had enjoyed a revolving credit facility (“the Singer Facility”), with a limit of US$30 million, from the Bank. The Singer Facility was secured by a pledge of 6 million shares in Singer Thailand owned by Singer NV.

  19. By the second half of 1997, the value of shares in Singer Thailand had dropped very substantially, and, as a result, the security provided by Singer NV was no longer anywhere near sufficient to meet the collateral requirements of its facility. On 24 November 1997 Singer NV repaid the Singer Facility in full. Notwithstanding the deficient security, it was allowed to redraw the full US$30 million on 5 January 1998. Stone J held that this was clearly a “mistake” and an embarrassing “error of judgment” on the part of employees of the Bank.

  20. In April 1998, having realised its mistake in permitting the re-drawing of the Singer Facility, the Bank initiated discussions to prompt proposals from Singer NV to address the problem of its severely defective security. The Singer Facility was rolled over for a further six months on 2 April 1998, pending the outcome of those discussions.

  21. During the course of those discussions, one possibility which was mooted was that the loan to Singer NV be redeemed by Akai, an arrangement which would be facilitated by the money for this being provided by the Bank in the form of a loan to Akai. In the context of this possibility, Akai’s memorandum and articles of association (or bye-laws) were provided to the Bank on 25 May 1998.

  22. The discussions eventually resulted in the Bank preparing an internal credit application dated 19 August 1998 in the name of Singer NV (“the Singer Credit Application”). This application included a proposal for a change of borrower from Singer NV to Akai, with Akai providing fresh security, which would fully cover the advance, in the form of Akai Electric shares. The Singer Credit Application was prepared by the Bank’s account officer responsible for the Singer Facility, Ms Tattaya Wattanakul (“Ms Wattanakul”), and approved by her supervisor, Mrs Jutathip Tasma.

  23. After stating that Singer NV needed to renew the US$30m loan facility for three years, the Singer Credit Application continued:

    [The Singer Thailand] shares .... cover only 4.2% of the outstanding. To comply with terms and conditions of the [loan facility] and to reduce the Bank’s risk, we proposed 2 alternatives to [Singer NV] .... :


    1st alternative, the Company shall renew the [US$30 million credit facility] .... for another 3 years by pledging of the remaining [Singer Thailand] shares .... to cover the loan from 4.2% to 9.07%.


    2nd alternative, .... changing of the borrower from [Singer NV] to [Akai] and changing of collateral from .... [Singer Thailand] shares to .... [Akai Electric shares] to reduce the Bank’s risk. .... The 2nd alternative would be beneficial to the Bank namely


    collateral would cover the credit at 100%


    [Akai’s] sales volume has increased every year and sustained profit. The Company has stable financial condition ....


    [Akai Electric] has good operating result ...

    For the above 2 alternatives, we have negotiated with the Company and are of the view that the interest rate should be adjusted from LIBOR + 2.25% to LIBOR + 4.50%.

  24. The Singer Credit Application attached a detailed corporate structure chart for Akai, STC Canada and Singer NV, which is a useful aide-memoire (although it refers to Akai Electric as “Akai Electronic”):

  25. On 24 September 1998, the Singer Credit Application was considered by the Bank’s Credit Committee. Of those attending the Credit Committee meeting of 24 September 1998, only Mr Siripongs Kalayanarooj (“Mr Siripongs”) gave evidence. He could not think of another example of a transaction similar to the Switch Transaction during his thirty years in banking.

  26. The Credit Committee did not accept either of the two options proposed in the application, but instead put forward an alternative proposal, which involved a reduction in the Singer Facility and pledge of additional Singer Thailand shares as security. This proposal was rejected by Singer NV, which was experiencing significant cash-flow problems and was not in a position to repay the US$30 million.

  27. By this stage, as Mr Siripongs explained, the Bank was faced with the choice of continuing to lend to Singer NV, an under-collateralised borrower that would inevitably default in a short time, or proceeding with the Switch Transaction. He also said that, in the event of default by Singer NV under the Singer Facility, which was inevitable without the Switch Transaction, the Bank would have to report such default to the Bank of Thailand and Bank of England, which at the time “would have caused problems for [the Bank]”.

  28. On 2 October 1998, the Bank prepared a second internal credit application, in the name of Akai instead of Singer NV (“the Akai Credit Application”), which was submitted directly to the Bank’s Executive Board. The Board’s members were the Bank’s chairman, Mr Banyong Lamsam, its chief executive officer, Mr Banthoon Lamsam, and two of its senior executives. Of these four people, only Mr Banthoon Lamsam gave evidence.

  29. The Akai Credit Application recommended, and sought approval for, the Switch Transaction. It proposed a short term credit line of US$30 million in favour of Akai secured by the pledge by Akai of Akai Electric Shares, to the value of 143% of the loan. The Akai Credit Application attached the same corporate chart as was attached to the Singer Credit Application, and it stated:


    [Singer NV] was granted 3 years Short Term Revolving Line in the amount of USD30M by the Bank by pledging [Singer Thailand], shares .... as collateral. Due to the declining of the shares value, the collateral value could cover only 4.2% of the outstanding debt.

    To minimize the Bank’s risk, it is proposed to change the Borrower from the [Singer NV] to [Akai], which is in the same group having stronger financial status, and change collateral from .... Singer Thailand shares to .... shares of [Akai Electric], which can fully cover the total credit line provided that [Akai] shall apply the amount of money to repay the [Singer Facility] .... and the interest rate would be adjusted from LIBOR + 2.25% to LIBOR + 4.5%.

    [T]he operating result of [Akai show] .... that the company has continuous growth of sale volume every year and has sustained profit continuously.

    The financial status is also stable .... Also, the value of [Akai Electric] shares to be pledged with the Bank replacing [Singer Thailand], shares can cover 100% of the credit line. As a result, the proposal should be approved as per request.

    The Akai Credit Application also stated that Akai was listed on the Stock Exchange of Hong Kong.

  30. On 6 October 1998, the Switch Transaction, as commended in the Akai Credit Application was considered and approved by the Bank’s Executive Board. The formal record of the meeting largely reflects the wording of the Akai Credit Application. The justification for the Switch Transaction, as recorded in the minutes of the meeting, was “to reduce the Bank’s risk” by effecting a change of borrower from Singer NV to Akai “a company in the same group with more stable financial status .... together with changing collateral .... to [Akai Electric] shares, which could cover the entire credit line”.

  31. Mr Banthoon Lamsam, who had never previously dealt with Mr Ting, asserted in cross-examination that his sole knowledge of the Switch Transaction had been derived from the Akai Credit Application. Stone J rejected that evidence. He found that the Switch Transaction had already been discussed between Mr Banyong Lamsam and Mr Ting, with a view to Akai taking over the Singer Facility and that it would “beggar belief” if Mr Banthoon Lamsam had not discussed the Switch Transaction with Mr Banyong Lamsam, who was his uncle.

    The Switch Transaction

  32. Following approval of the Switch Transaction, in early December 1998 Ms Wattanakul was, according to the Judge, “instructed by her superior to obtain evidence of the authority of the person who was to sign the agreements on behalf of Akai”. She liaised with Mr Domine Ko (“Mr Ko”) of Akai to fix a time for Akai to execute the Loan Agreement and Share Pledge Agreement. Mr Ko told Ms Wattanakul that Mr Ting would be signing on behalf of Akai, and she asked him to provide evidence at the meeting confirming Mr Ting’s authority. As the signing was to take place in Hong Kong, Ms Wattanakul told Mr Bhadranavik of the Bank’s Hong Kong office that evidence of the authority of the person signing the documentation on behalf of Akai should be provided.

  33. On 4 December 1998, Mr Bhadranavik attended Akai’s office in Hong Kong to collect the documentation for the Switch Transaction, and was met by Mr Ting, Ms Clara Loh and Mr Ko of Akai. Mr Bhadranavik was handed the following documents:

    1. “Minutes of an Executive Committee” meeting of Akai of 4 December 1998 (“the ExCo minutes”), signed by Mr Ting, with Mr Ko as witness;

    2. The Loan Agreement under which Akai agreed to repay the Bank US$30 million within 12 months of drawdown, signed by Mr Ting;

    3. A Drawing Notice, whereby Akai applied to draw down the whole of the US$30m immediately, signed by Mr Ting;

    4. A payment direction, which requested the London Branch of the Bank to “disburse the [US$30m] to yourself for the repayment of the loan in the same amount due by [Singer NV] to yourself ....”, signed by Mr Ting;

    5. A Share Pledge Agreement recording the terms of the pledge of 56 million Akai Electric shares as security for the Loan Agreement, signed by Mr Ting; and

    6. The share certificates, for the pledged 56 million shares in Akai Electric, worth at that time US$56.3 million, to be held as security for the loan.

  34. Mr Bhadranavik took these documents back to his office, and informed his Branch supervisor, Mr Niasinn Lamsam, that the signing of the agreements had been completed and that he had obtained the ExCo minutes. Mr Niasinn Lamsam read the minutes and signed the Loan Agreement as a witness (even though he had not been present at the meeting). The documents were then sent to Ms Wattanakul in Thailand. The loan was thereafter formally drawn down on 7 December 1998, and the pledged shares were lodged with the Bank on the same date. By a promissory note dated 7 December 1998, signed by Mr Ting on behalf of Akai, Akai promised to pay to the Bank on 7 January 1999 the sum of US$30 million plus interest.

  35. It is appropriate to set out the ExCo minutes in full:


    Minutes of a Meeting of the Executive Committee of the Company held at 3001 Two Exchange Square, 8 Connaught Place, Hong Kong on 4 December 1998


    Mr. James H. Ting

    Ms. Clara Loh

    Dr. Frank Edward Holmes (by phone)

    Mr. Chuck Tam (by phone)



    Mr. James H. Ting was elected as Chairman of the Meeting.



    The Chairman declared that a quorum of the Meeting was present and that the Meeting was duly constituted.



    There was tabled at the Meeting a draft Loan Agreement (‘the Loan Agreement’) to be entered between the Bank and the Company in relation to the granting of a loan facility of US$30,000,000 (the ‘Loan’) by the Bank to the Company. It was noted that it is one of the terms of the Loan Agreement that the Company shall pledge 56,000,000 shares of Akai Electric Co., Ltd, a subsidiary of the Company, to the Bank pursuant to a draft Share Pledge Agreement (the ‘Pledge’), as tabled, to be entered between the Bank and the company.

    It was resolved that the Loan Agreement and the Pledge (the ‘Agreement’) be accepted and approved and that any one of the Directors be authorised to approve any changes made to the Agreements prior to their execution and to execute, under hand or seal, and deliver the Agreement, and any other documents which may, from time to time, necessary for the Loan on behalf of the Company.

    It was also resolved that the Loan be operated as follows:

    Mr James H Ting

    )  signing singly

    Ms Clara Loh

    Dr Frank Edward Holmes

    Mr Chuck Tam

    )  two out of
    )  any three
    )  signing jointly



    There being no other business, the Chairman declared the Meeting terminated.

    [Signature of Mr Ting]
    Chairman of the Meeting


  36. In fact, as the Judge found, the ExCo minutes was a false document, which had been dishonestly prepared by Mr Ting, or at least on his instructions. No meeting of Akai’s Executive Committee was held on that day.

    Events subsequent to the Switch Transaction

  37. A few months later, Akai asked the Bank to release a number of the pledged shares, as the market value of Akai Electric had increased significantly. The Bank initially refused, but it then released 5.5m of the pledged shares to Akai, leaving the Bank holding certificates for 50.5m Akai Electric shares, “the Shares”.

  38. In September 1999, Singer NV filed for bankruptcy protection in the USA under Chapter 11. On 6 December 1999, Akai failed to pay US$31,298,906, which had become owing to the Bank under the Loan Agreement. On that date, the Shares were worth US$32,904,372. An event of default was declared by the Bank on 23 December 1999. However, it was not until April and May 2000 that the Bank enforced its security over the Shares, by selling them in the market. In so doing the Bank realised proceeds of US$20,504,295, which was applied in full to discharge partially Akai’s then outstanding debt, and the Bank proved in the winding-up of Akai for the shortfall of over US$13m. Meanwhile, Akai was being wound up in Hong Kong and Bermuda.

  39. On 16 August 2002, Akai’s Hong Kong Liquidators notified the Bank that its claim against Akai had been adjudicated to be some US$13m in the Hong Kong liquidation of Akai. On the same day, the Bermudian Liquidators of Akai notified the Bank that its claim against Akai in the Bermudian liquidation had also been adjudicated to be some US$13m.

  40. These proceedings were issued on 3 December 2004. The trial was held before Stone J over 20 sitting days between 12 February and 18 March 2008, and he gave judgment in favour of the Bank on 26 May 2008. Akai’s appeal was heard from 8 to 12 June 2009 and the Court of Appeal’s unanimous judgment in favour of Akai was delivered on 10 August 2009, and judgment was entered for Akai in the sum of US$22,451,332 plus compound interest at 1% over LIBOR (from time to time prevailing) from 6 December 1999, with quarterly rests.

    Did Mr Ting have authority to bind Akai to the Switch Transaction?

  41. The first of the two main questions to be considered is whether Mr Ting had authority to commit Akai to the Switch Transaction in December 1998. If he did, then the transaction is and was binding on Akai, and, subject to Akai’s claim in knowing receipt (which, for reasons which will be explained below, I consider stands or falls with the authority issue), Stone J would have been right in dismissing Akai’s claim. If, on the other hand, Mr Ting did not have such authority, then Akai was never committed to the Switch Transaction, indeed the whole transaction was void, the Bank never had any rights over the Shares, and it must therefore pay damages, as the Court of Appeal held.


  42. The law relating to apparent authority has been considered in a number of cases, not all of which are easy to reconcile in every respect. There are three issues of principle between the parties, and a number of other disputes as to the consequence of the application of the principles to the facts of the present case.

  43. In an often cited passage in Freeman & Lockyer v Buckhurst Park Properties (Mangal) Ltd [1964] 2 QB 480, 506, Diplock LJ identified four conditions which have to be satisfied before a third party, whom he described as a “contractor”, can enforce a contract against a company entered into by a purported agent with no actual authority. Those conditions are:


    that a representation that the agent had authority to enter on behalf of the company into a contract of the kind sought to be enforced was made to the contractor;


    that such representation was made by a person or persons who had ‘actual’ authority to manage the business of the company either generally or in respect of those matters to which the contract relates;


    that he (the contractor) was induced by such representation to enter into the contract, that is, that he in fact relied upon it; and


    that under its memorandum or articles of association the company was not deprived of the capacity either to enter into a contract of the kind sought to be enforced or to delegate authority to enter into a contract of that kind to the agent.

  44. As Diplock LJ explained at [1964] 2 QB 480, 503, apparent authority:

    .... is a legal relationship between the principal and the contractor created by a representation, made by the principal to the contractor, intended to be and in fact acted upon by the contractor, that the agent has authority to enter on behalf of the principal into a contract of a kind within the scope of the ‘apparent’ authority, so as to render the principal liable [thereunder].
  45. In a decision of the Supreme Court of Western Australia Court of Appeal, Auxil Pty Ltd v Terranova (2009) 260 ALR 164, para.176, Newnes JA said this about representations capable of creating an apparent authority:

    A representation creating an apparent authority of an agent may be made in a number of ways but the most common form of representation by a principal is by conduct, that is, by permitting the agent to act in the management or conduct of the principal’s business. By permitting the agent to act in the management or conduct of the business, the principal thereby represents to anyone dealing with the agent that he or she has authority to do those acts on behalf of the company which an agent authorised to do acts of the kind which he or she is in fact permitted to do normally does in the ordinary course of such business.

    This passage reflects what was said by Diplock LJ in Freeman & Lockyer [1964] 2 QB 480, 503 and 505.

  46. In this case, the Bank’s argument in a nutshell is

    1. that, by holding him out as its executive chairman and chief executive officer, Akai clothed Mr Ting with apparent authority to commit it to the Switch Transaction, and

    2. that, when the Bank entered into the transaction, it justifiably relied on his authority to enter into the transaction on Akai’s behalf. Stone J accepted both those propositions, but they were both rejected by the Court of Appeal.

    In this court, Akai challenges both of those propositions.

  47. Probably the main point of principle which divides the parties centres on what state of mind is required in order to establish that the Bank “justifiably relied” on Mr Ting’s authority. Another issue of principle is the extent, if any, to which a third party who claims to have dealt with an alleged apparent agent can rely on unauthorised statements made by the agent as to his authority to justify the contention that he had authority to commit the principal. The third issue of principle concerns the evidence required from the Bank to show that it relied on the apparent authority of Mr Ting.

  48. I propose to consider those three issues of principle first, and will then turn to the question of how they are to be applied in the present case. In that latter connection, I will first consider whether the Bank can justifiably have believed that Mr Ting had authority to commit Akai to the Switch Transaction, irrespective of the ExCo minutes, and I will then turn to the effect of those minutes.

    The state of mind of the person alleging apparent authority

  49. For the Bank, Mr Jonathan Sumption QC (“Mr Sumption”) (who appeared with Mr Eugene Fung) contended that, unless the Bank had actual knowledge of Mr Ting’s lack of authority or its belief that Mr Ting had authority was dishonest or irrational, then the Bank’s state of mind will suffice for the purpose of establishing apparent authority. Mr Sumption also accepted that, if the Bank was reckless in its belief, or if it was guilty of turning a blind eye, that would not do either, on the basis that recklessness and blind-eye ignorance amount to irrationality or dishonesty in this context. Mr Leslie Kosmin QC (“Mr Kosmin”) (who appeared for Akai with Ms Linda Chan) argued that this set too low a standard on the Bank as a third party seeking to establish apparent authority, and that apparent authority could not be relied on if the Bank had failed to make the inquiries that a reasonable person would have made in the circumstances to verify the authority of Mr Ting.

  50. I have some doubts as to the extent to which there would, in practice, be much difference in outcome between the application of the rival tests. The distinction between Mr Sumption’s “irrationality” and Mr Kosmin’s “unreasonableness” may puzzle anyone who has recourse to the dictionary, although I would accept that the former word tends to carry more pejorative overtones, and therefore would set a higher hurdle for a party in the position of Akai in the present case. Similarly, the distinction between turning a blind eye and being put on enquiry seems fairly slender, but I accept that the former involves a more subjective exercise than the latter, although irrationality ultimately involves an objective assessment.

  51. Approaching the issue by reference to both practicality and principle, I would prefer the Bank’s submission. In terms of practicality, at least when it comes to normal commercial transactions, the application of the concept of constructive notice, which is what Akai’s approach effectively involves, has been deprecated (see for instance per Scrutton LJ in Greer v Downs Supply Co [1927] 2 KB 28, 36). And one can understand why.

  52. In a commercial context, absent dishonesty or irrationality, a person should be entitled to rely on what he is told: this may occasionally produce harsh results, but it enables people engaged in business to know where they stand. As to principle, apparent authority is essentially a species of estoppel by representation (see per Diplock LJ in Freeman & Lockyer [1964] 2 QB 480, 503, cited above, and per Brennan J in the High Court of Australia in Northside Developments Pty Ltd v Registrar-General (1989-1990) 170 CLR 146, 173-4). In the field of misrepresentation, it is clear that “it is no defence to an action for rescission that the representee might have discovered its falsity by the exercise of reasonable care” – per Chitty on Contracts (30th edition) para.6-039 and the cases cited in footnote 190. Even more in point, there is this passage in Halsbury’s Laws of England (4th edition reissue) Vol 16(2), para.1072, dealing with estoppel by representation:

    If .... [the party contending that he relied on the representation] really has relied upon its truth, it is no answer to say that, if he had thought about it, he must have known that it was untrue; the representation itself was what put him off his guard. If the representation is clear and unequivocal .... he is under no obligation to make investigation or inquiry to ascertain whether it is true.
  53. That proposition is supported (albeit in slightly different commercial contexts) by a number of judicial dicta of high authority, including from Lord Halsbury LC and Lord Herschell in Bloomenthal v Ford [1897] AC 156, 161-2 and 166-7 respectively, and Lord Blackburn in Jones v Gordon (1876-7) 2 App Cas 616, 628-9. At the end of that last-cited passage, Lord Blackburn provided a characteristically clear explanation of what constitutes blind eye knowledge, or turning a blind eye when he said this:

    [I]f the facts and circumstances are such that the [judge comes] to the conclusion that he was not honestly blundering and careless, but that he must have had a suspicion that there was something wrong, and that he refrained from asking questions, not because he was an honest blunderer or a stupid man, but because he thought in his own secret mind – I suspect there is something wrong, and if I ask questions and make farther inquiry, it will no longer be my suspecting it, but my knowing it, and then I shall not be able to recover – I think that is dishonesty.
  54. In the context of an agency case, where the issue was whether the defendants should have appreciated that the party with whom they had contracted was an agent for the plaintiff, rather than a principal, Lord Edmund Davies (sitting in the English Court of Appeal with Viscount Dilhorne and Lord Scarman) in By Appointment (Sales) Ltd v Harrods Ltd (1 December 1977, Transcript 465/1977) said at the end of his judgment that “there was possibly constructive notice but not actual notice, and actual notice is required”.

  55. Millett J took the clear view that constructive notice did not involve “the proper approach” in a normal commercial case, on the basis that “[u]nless and until they are alerted to the possibility of wrongdoing, [account officers] proceed, and are entitled to proceed, on the assumption that they are dealing with honest men”. He continued by saying that they were entitled to do so until “the facts .... [make] it imperative for [them] to seek an explanation, because in the absence of an explanation it was obvious that the transaction was probably improper” – Macmillan Inc v Bishopsgate Investment Trust plc [1995] 1 WLR 978, 1014G-H.

  56. It is true that there are judicial dicta which appear at first sight to provide some support for Akai’s case on this issue, but, on closer analysis, I do not consider that they are in point.

  57. Two of the three judgments in the English Court of Appeal decision of AL Underwood Ltd v Bank of Liverpool and Martins [1924] 1 KB 775 contain reference to “negligence” and being put on inquiry - see at [1924] 1 KB 775, 785 and 788-9 (per Bankes LJ) and 793 (per Scrutton LJ). However, that was because the defendant had to rebut an allegation of negligence in order to rely on s.82 of the Bills of Exchange Act 1882. It is fair to say that Atkin LJ’s judgment contains a sentence which, read on its own supports Akai’s argument (see at [1924] 1 KB 775, 797-8), but, if considered in the context of s.82 of the 1882 Act, I do not think it can fairly be regarded as a general statement of the law of apparent authority.

  58. Another English Court of Appeal decision relied on by Akai is Houghton and Co. v Nothard, Lowe and Wills Ltd [1927] 1 KB 246. There is no doubt but that, in that case, the test propounded by the Court was that proposed by Mr Kosmin. However, it was a case involving the so-called indoor management rule (also known as the rule in Turquand’s case ­– see Royal British Bank v Turquand 6 E & B 327), decided at a time when a person dealing with a company was deemed to know of the terms of its memorandum and articles of association which is, of course, no longer the law in Hong Kong, or indeed, England. A passage at [1927] 1 KB 246, 267, in the judgment of Sargant LJ indicates an acceptance of a difference between the indoor management rule and apparent authority.

  59. That difference was, in my opinion, well explained by Dawson J in Northside Developments (1990) 170 CLR 146, 198 in a passage which neatly distinguishes between the rule and apparent (or, as he called it, ostensible) authority:

    The correct view is that the indoor management rule cannot be used to create authority where none otherwise exists; it merely entitles an outsider, in the absence of anything putting him upon inquiry, to presume regularity in the internal affairs of a company when confronted by a person apparently acting with the authority of the company. The existence of an article under which authority might be conferred, if it is known to the outsider, is a circumstance to be taken into account in determining whether that person is being held out as possessing that authority .... In other words, the indoor management rule only has scope for operation if it can be established independently that the person purporting to represent the company had actual or ostensible authority to enter into the transaction. The rule is thus dependent upon the operation of normal agency principles; it operates only where on ordinary principles the person purporting to act on behalf of the company is acting within the scope of his actual or ostensible authority.

    It is also worth referring to what Brennan J (as he then was) said in this connection at (1990) 170 CLR 146, 177-9.

  60. Other cases relied on by the Bank, which similarly concerned the indoor management rule, included Morris v Kanssen [1946] AC 459, 475, and Rolled Steel Ltd v British Steel Corporation [1986] Ch 246, 295 and 304 (per Slade LJ and Browne-Wilkinson LJ respectively). It is fair to say that some of the observations in Rolled Steel [1986] Ch 246, especially at [1986] Ch 246, 296, proposition (6), can be read as supporting Akai’s case. However, it seems to me that the precise nature of the state of mind or knowledge of the person alleging apparent authority was not in issue in that case, and, with all respect to Slade LJ, the cases he had cited did not support the proposition, if it was intended to refer to constructive notice and to apply to apparent authority generally.

  61. Whether or not the indoor management rule is part of the law of agency or companies is a category issue which is of no consequence at least for present purposes. The essential point, to my mind, is that it was developed for good practical reasons to mitigate the harsh consequences of the principle (now thankfully historic) that a person dealing with a company was deemed to know the contents of its memorandum and articles. It is therefore unsurprising that the courts developed the principle that one could only rely on the rule if one took reasonable steps to ascertain the relevant facts. There is no obvious reason why the same principle should apply to cases of apparent authority.

  62. I conclude that it is open to the Bank to rely on Mr Ting’s apparent authority (if he had such authority) unless the Bank’s belief in that connection was dishonest or irrational (which includes turning a blind eye and being reckless).

    Can an agent clothe himself with apparent authority?

  63. A point which was hotly debated was whether a third party, seeking to establish that an alleged agent had apparent authority, can rely on what the apparent agent represented (without the sanction of the principal) about his authority, so as to clothe him with the authority which he would not otherwise have. The point arose out of Mr Sumption’s contention that the argument that Mr Ting had apparent authority to commit Akai to the Switch Transaction was reinforced by the ExCo minutes, which of course had been prepared, signed and handed over by Mr Ting himself. As I am of the view (for reasons explained below) that, properly analysed, those minutes actually harm, rather than assist, the Bank’s case on apparent authority, the point is academic in this case, but, as it has been fully argued, I shall deal with it.

  64. At any rate at first sight, it would appear to require exceptional facts before such an argument could have any chance of success. After all, apparent authority is based on a representation (normally implied) as between the alleged principal and the third party as to the authority of the alleged agent, and if the third party could rely on some statement by the alleged agent, made without the authority of the principal, it would seem precious close to pulling up oneself by one’s own bootstraps.

  65. Support for the notion that an agent who has no apparent authority cannot clothe himself with such authority by his own unauthorised words may be found in Freeman & Lockyer [1964] 2 Q.B. 480, 504, where Diplock LJ summarised the position as follows:

    The second characteristic of a corporation, namely, that unlike a natural person it can only make a representation through an agent, has the consequence that in order to create an estoppel between the corporation and the contractor, the representation as to the authority of the agent which creates his ‘apparent’ authority must be made by some person or persons who have ‘actual’ authority from the corporation to make the representation.
  66. By reference to individuals in the present case, Mr Sumption rhetorically asked why, if Mr Ting could have given apparent authority to Miss Loh to sign on behalf of Akai, should he not be able to give himself apparent authority to sign? Attractive though that argument seems at first sight, it seems to me, on analysis, to take matters little further. If Mr Ting had purported to give Miss Loh authority, the question would still arise as to whether he had Akai’s authority to give her authority to bind Akai. So, by the same token, if he purports to clothe himself with authority, the question would still arise as to whether he had authority to authorise himself to bind Akai, which is, at least normally, the same as asking whether he has apparent authority to bind Akai.

  67. We have been referred to three English cases on this topic. The first is Armagas Ltd v Mundogas SA (“The Ocean Frost”) [1986] AC 717, in the Court of Appeal and House of Lords. The second and third are decisions of the Court of Appeal, Soplex Wholesale Supplies Ltd v Egyptian International Foreign Trade Co. (“The Raffaella”) [1985] 2 Lloyd’s Rep 36 (decided after the Court of Appeal’s decision, but before that of the House of Lords, in Armagas [1986] AC 717) and First Energy (UK) Ltd v Hungarian International Bank Ltd [1993] 2 Lloyd’s Rep 194.

  68. The effect of those decisions appears to be that, at least in the Courts of England and Wales, it is very unlikely that a third party could establish apparent authority on the part of an alleged agent by relying on the agent’s own unauthorised statement to clothe the agent with authority. The judgment of Goff LJ and the speech of Lord Keith of Kinkel in Armagas [1986] AC 717, 730H-732F and 777C-778D indicate great scepticism as to the notion that an agent could clothe himself with authority in this way. Having said at [1986] AC 717, 777D that there might be “very rare and unusual” circumstances in which this might happen, Lord Keith continued at 779D-G:

    Robert Goff L.J. said of the trial judge’s view in this case, ante, pp. 730H-731C:

    the effect of the judge’s conclusion was that, although [the alleged agent] did not have ostensible authority to enter into the contract, he did have ostensible authority to tell [the third party] that he had obtained actual authority to do so. This is, on its face, a most surprising conclusion. It results in an extraordinary distinction between (1) a case where an agent, having no ostensible authority to enter into the relevant contract, wrongly asserts that he is invested with actual authority to do so, in which event the principal is not bound; and (2) a case where an agent, having no ostensible authority, wrongly asserts after negotiations that he has gone back to his principal and obtained actual authority, in which event the principal is bound. As a matter of common sense, this is most unlikely to be the law.

    I respectfully agree. It must be a most unusual and peculiar case where an agent who is known to have no general authority to enter into transactions of a certain type can by reason of circumstances created by the principal reasonably be believed to have specific authority to enter into a particular transaction of that type.

  69. The facts of Soplex Wholesale [1985] 2 Lloyd’s Rep 36 and First Energy [1993] 2 Lloyd’s Rep 194 were both rather unusual. In the former case, the observations of Browne-Wilkinson LJ on the topic at [1985] 2 Lloyd’s Rep 36, 43 were plainly obiter, and not supported in the other two judgments (in particular, Kerr LJ seems to have taken a different view – see at [1985] 2 Lloyd’s Rep 36, 46-7). In any event, the agent’s representation in that case was that one signature would suffice in the London market – not a simple statement that the alleged agent was authorised, more of a simple statement of alleged fact. In First Energy [1993] 2 Lloyd’s Rep 194, the communication was that the principal had agreed, not that the alleged agent had authority to agree on the principal’s behalf.

  70. It seems to me that the reasoning in the judgments of Steyn and Nourse LJJ in First Energy [1993] 2 Lloyd’s Rep 194 illustrate how the law in this field struggles to reconcile principle and predictability with commercial reality and fairness, and this difficulty underlines the inadvisability of seeking to lay down any rigid principles in this area, especially as the point does not arise in this case. In my view, the law is as stated by Lord Keith in the passages I have quoted from his speech in Armagas [1986] AC 717, but it is right to add that I find it very hard indeed to conceive of any circumstances in which an alleged agent, who does not have actual or apparent authority to bind the principal, can nevertheless acquire apparent authority to do so, simply by representing to the third party that he has such authority.

  71. It is worth adding that, as was explained in the two cases just referred to, before any representation by the agent could be relied on to assist the contention that he had apparent authority, the court would have to be satisfied that the principal had given the alleged agent apparent authority to make the representation in question. Furthermore, any such representation would have to be “clear and unequivocal”, as in any case of estoppel by representation (unlike, arguably, proprietary estoppel – see per Lord Walker of Gestingthorpe in Thorner v Major [2009] 1 WLR 776, paras 54-56), and what is clear and unequivocal must be judged by reference to the practical realities of the particular case (as I suggested in Thorner [2009] 1 WLR 776, paras 84-86). Otherwise, apparent authority could not be established on the back of the statement as a matter of principle.

    Evidence of reliance

  72. Given that apparent authority is a species of estoppel by representation, it follows that, as Diplock LJ said in his third proposition in Freeman & Lockyer [1964] 2 QB 480, 506, the third party must establish that it relied on the apparent authority of the alleged agent before it can succeed in establishing its case. The issue which divides the parties is the quality of the evidence which the third party has to lead to establish reliance. Inevitably, that is a fact-sensitive issue, but I detected a suggestion in Akai’s case that it is positively necessary for a third party to establish that he positively relied on the apparent authority in every case.

  73. In Nationwide Building Society v Lewis [1998] Ch 482, a case concerning the alleged holding out of a person as a partner in a firm, Peter Gibson LJ said at 491:

    It does not seem to me to be impractical or unjust for the law to require a person claiming an estoppel to have to prove in a partnership context what he would have to prove in other contexts. Given that reliance is a necessary requirement, it is not obvious that there should be a presumption in favour of the person who claims reliance and is in a better position to know whether he did rely on the holding-out and who should thereby be able to prove it .... Of course, there may be circumstances from which it would be appropriate for the court to infer that there was reliance on a holding-out. As is stated in Spencer Bower and Turner on Estoppel by Representation, 3rd ed. (1977) pp. 114 – 115:

    Though on questions of fact the onus will be upon the representee, it may happen that the probability of inducement from a given set of facts is so great, or in other words the materiality is so plain and palpable, as to justify a finding of the inducement itself merely from the circumstantial context .... but it must be remembered that the inference so made is one of fact and not of law.

  74. While there is nothing with which I would fundamentally disagree in those observations, I consider that their thrust could risk placing too high a hurdle for third parties seeking to establish a claim in apparent authority. In Silver v Ocean Steamship Co Ltd [1930] 1 KB 416, it was held that the taking up of a bill of lading was sufficient evidence of reliance on it to raise an estoppel. The point was dealt with very shortly in the judgments of Scrutton, Greer and Slesser LJJ at [1930] 1 KB 416, 428, 434, and 441 effectively on the basis that it was obvious. Thus, Slesser LJ said that “[t]he appellant took the bill and may be assumed to have relied upon it to his detriment in the absence of any evidence to the contrary”.

  75. In my view, once a third party has established that the alleged agent had apparent authority, i.e. that the principal held out the alleged agent as having authority to bind the principal, and that the third party has entered into a contract with the alleged agent on behalf of the principal, then, in the absence of any evidence or indication to the contrary, it would be an unusual case where reliance was not presumed.

    Did Mr Ting have apparent authority (disregarding the ExCo minutes)?

  76. Having resolved the issues of principle between the parties, I now turn to consider whether in this case, the Bank has established that Mr Ting had apparent authority to commit Akai to the Switch Transaction. In considering that issue, it is important to bear in mind that the primary fact-finder in this case is Stone J, who had the advantage of seeing live witnesses, although a number of the centrally important figures (notably Mr Ting and Mr Banyong Lamsam) were either unable or unwilling to give evidence, a feature from which he seems to have drawn no adverse inferences.

  77. As I have explained, it is no longer contended by the Bank that Mr Ting had actual authority to commit Akai to the Switch Transaction: the Bank relies solely on the contention that he had apparent authority. However, it is important to emphasise that the effect of the Bank’s concession that Mr Ting had no actual authority is that, when purporting to commit Akai to the Switch Transaction, he was acting outside the scope of what he was actually authorised to do. That concession, which was plainly rightly made, must carry with it the implication that Mr Ting was acting dishonestly, especially in the light of the forged ExCo minutes. The Court of Appeal plainly concluded that Mr Ting acted dishonestly in connection with the transaction, and Stone J appears to have thought so too, albeit that he expressed himself rather elliptically: at para.331 of his judgment, having said that he was not “disposed to assume the honesty of Mr Ting”, he added “the overwhelming probability, I should have thought, is to the contrary”.

  78. As already indicated, I propose to consider the issue in two stages, first ignoring the ExCo minutes, and then considering its effect. I appreciate that this may seem a slightly surprising approach, but I believe that it is helpful in the light of the unusual facts of this case.

  79. Absent the ExCo minutes, the only basis upon which Mr Sumption put his case that Akai represented that Mr Ting had authority to commit it to the Switch Transaction was that he was authorised by Akai to manage its affairs, by virtue of being its executive chairman and chief executive officer.

  80. As far as I can see, that was indeed the only basis upon which the Bank could put its case. It is clear that, by December 1998, the Bank had had dealings with Mr Ting, and with some companies in which he had a substantial interest and which he managed (most obviously, Singer NV), for many years, and that at least one senior member of the Bank, Mr Banyong Lamsam, the chairman, must have had a fairly close relationship with Mr Ting. However, it also apparent that there had been no history of any dealings whatever between the Bank and Akai until the Switch Transaction.

  81. It is clear that, as executive chairman and chief executive officer of Akai, Mr Ting would have had a large measure of apparent authority - indeed, no doubt he would have had a large measure of actual authority. That authority would, no doubt, extend to entering into many types of contract, including, I would have thought, contracts which might involve Akai incurring a US$30m liability. However, the question which has to be faced is whether the nature and circumstances of the Switch Transaction were so peculiar as to preclude the Bank contending that his apparent authority extended as far as committing Akai to it.

  82. In that connection, in support of his contention that the Switch Transaction was so extraordinary as to take outside the ambit of Mr Ting’s apparent authority, Mr Kosmin essentially relied on the fact that the transaction was not merely very unusual in nature and very substantial in the sum involved, but, above all, that it was, quite plainly, for the benefit of Singer NV and the Bank, but not in any way for the benefit of the party allegedly represented by Mr Ting, Akai.

  83. The effect of the Switch Transaction was, at least on the face of it, as follows. Akai took on a liability to the Bank for US$30m, which it had to repay with interest in a year’s time, which liability was secured by the pledging of a large proportion of the shares in Akai’s chief operating subsidiary. In return, Akai got nothing: the US$30m was immediately used to pay off the liabilities of another company, Singer NV, which (whether directly or indirectly) neither owned nor was owned by Akai, albeit that they shared the same parent: the two companies were neither subsidiaries nor affiliates. The transaction was of obvious and very substantial benefit to Singer NV (even on the assumption that Akai became subrogated to Singer NV’s liability to the Bank). It was also of obvious and very substantial benefit to the Bank: for no cost to itself, it gained an apparently financially strong borrower and excellent security in place of a weak borrower, already in default, with security of almost negligible value.

  84. This in turn gives rise to a number of other points of concern. Although the largest shareholder in both Akai and Singer NV was STC Canada, the majority of the shareholders in Akai (and, if it is relevant, which I doubt, half of the shareholders in Singer NV) were members of the public, who had no apparent interest in using a substantial amount of Akai’s money to prop up the ailing Singer NV. That is not only significant in itself, but (as was stated in the Akai Credit Application) Akai was listed on the Hong Kong Stock Exchange, and its Rules required the Switch Transaction to be disclosed to the Stock Exchange authorities, as it was an arrangement between connected persons, and the Stock Exchange would probably have required shareholder approval, as well as main Board approval, so far as Akai was concerned.

  85. In addition, the very person who was purporting to represent Akai, Mr Ting, was in an obvious position of conflict, given his very substantial interest in STC Canada and his management responsibilities for both Singer NV and Akai. To appreciate this conflict did not require any nuanced understanding of Akai’s bye-laws or of company law – Mr Ting’s conflict must have been obvious to any banker reviewing the Akai Credit Application. But, over and above this, arts 103 and 104 of Akai’s bye-laws, which the Bank had obtained in connection with the transaction, made it clear that he should not have been involved with the decision to enter into the Switch Transaction, owing to his conflicted position.

  86. All these facts were well known to the Bank as was clear from the evidence, and indeed as is clear from the contents of the Singer and Akai Credit Applications. It does not appear that the Bank at any time asked for, or was given, any commercial justification for Akai entering into the Switch Transaction. The two Credit Applications (in particular the Akai Credit Application) made much of the benefit to the Bank, but nothing was said about Akai’s reasons for entering into the transaction. Because neither Mr Banyong Lamsam (who produced a medical certificate to explain his absence) nor Mr Ting (whose absence is unsurprising) gave evidence, and because Mr Banthoon Lamsam gave unreliable evidence about it (and even if accurate, his evidence would have been second-hand), one simply does not know what passed between the two most senior representatives of the principal participants at their meeting before the crucial Bank executive committee meeting which approved the Switch Transaction.

  87. It appears to me that the various features of the Switch Transaction that I have described, especially when taken together, render it, at least in the absence of significant factors pointing the other way, very difficult for the Bank to contend that, merely because he was the executive chairman and chief executive officer of Akai, Mr Ting was, without more, held out as having authority to commit Akai to the transaction. As Brennan J said in Northside Developments (1990) 1 CLR 146, 183, “the execution of guarantees and supporting securities for another’s liabilities, not being for the purposes of a company’s business nor otherwise for its benefit, is not ordinarily within the authority of the officers or agents of a company” (and see to similar effect, the observations of Mason CJ at (1990) 1 CLR 146, 160-1). In such a case, one must scrutinise with particular care any allegation of apparent authority.

  88. All the more so, where, as here, the transaction is one from which the third party seeking to assert the apparent agency benefits very substantially. Indeed, in that sort of case, the court’s state of mind may almost verge on the sceptical. As Lord Cairns LC said in Gray v Johnston (1868) LR 3 HL 1, 11 “if it be shewn that any personal benefit to the bankers themselves is designed or stipulated for, that circumstance, above all others, will most readily establish the fact that the bankers are in privity with the breach of trust which is about to be committed.”.

  89. When one examines the Bank’s conduct against the standards which, on the basis of the evidence (which included testimony as to the Bank’s own practice and that of Thai banks generally) and commercial common sense, one would have expected, its case on apparent authority gets worse. So far as standards are concerned, the only witness to whose evidence I need to refer is Mr Banthoon Lamsam, the sole person present at the Executive Board meeting to give evidence, and who was described by Stone J as “an experienced and highly intelligent international banker”.

  90. Mr Banthoon Lamsam’s evidence displayed extensive knowledge of proper practices of corporate governance and banking practice. His evidence was that:

    • The “normal course” in the case of “a large transaction” would be for the Bank to obtain a Board resolution from the corporate borrower authorising the arrangement, and this was “standard practice” for a Thai bank, and the resolution normally would state that the arrangement was in the best interests of the company; and

    • When dealing with a new foreign borrower, the Bank would normally obtain legal advice from a lawyer qualified in the borrower’s country of incorporation, to ensure that the borrower was bound by the arrangement.

  91. Mr Banthoon Lamsam accepted that the Switch Transaction was “important” and a “major shift in the structure of the relationship”, and that the descriptions “routine” and “unremarkable”, contained in his witness statement to describe the transaction, were not appropriate. He accepted that Mr Banyong Lamsam (as chairman and a director of the Bank and of Singer Thailand) and Mr Ting (as chief executive and a director of both Akai and Singer NV) were each in positions of conflict in respect of the transaction. He also knew that Akai was quoted on the Hong Kong Stock Exchange and that over half of its shares were in the hands of the public. He was unaware of any benefit to Akai from the Switch Transaction, but appreciated the benefit to the Bank; he said that the only information which the Executive Board had was in the Akai Credit Application.

  92. Mr Banthoon Lamsam’s evidence was to the effect that none of this was his concern, and he was not required to pay any regard to Akai’s interests. In a sense, of course, that is true. However, given that the Bank’s normal practice was to obtain a Board Resolution, and to get appropriate legal advice, it appears to me to have been quite extraordinary for the Bank to have proceeded with the Switch Transaction, with all its remarkable and questionable features, simply on the basis of Mr Ting’s say-so. All the more so given that there was plenty of time, nearly two months, to do this between the Executive Board’s approval of the Akai Credit Application and the completion of the transaction.

  93. There is some support for the notion that a formal resolution by the board of directors is appropriate for a large transaction, and that such a resolution should fully describe the transaction, state that it is in the interests of the company, and record any conflicts of interests. That evidence is in the form of a full and detailed board resolution of Akai in December 1997 relating to the so-called “Pfaff transaction”, involving the sale of just over 80% of the shares in GM Pfaff AG by Akai to Singer NV. The nature of the transaction, its benefit for Akai, the conflicted position of Mr Ting and Dr Holmes, and the need to comply with the regulatory requirements in the light of the existence of public shareholders are all recorded in the board resolution approving the transaction, signed by all the directors. This was not a document seen by the Bank (as far as one can tell), but its existence and terms rather underline the oddity of the absence any such document in relation to the Switch Transaction, which cried out for justification.

  94. Subject to any good reason to the contrary, for the reasons I have attempted to give, I would accordingly conclude that:

    1. Despite his wide powers of management, Mr Ting did not have apparent authority to commit Akai to the Switch Transaction, in the sense that he was not clothed with the authority to commit Akai to the transaction; and

    2. Even if he had otherwise been clothed with such authority, the Bank was simply irrational in its belief that it relied on that authority.

  95. I am aware that there is a significant degree of overlap between these two preliminary conclusions, but they do involve separate questions, namely

    1. whether Mr Ting was held out as having authority at all, and, if he was,

    2. whether the Bank could rely on such holding out.

    I am also aware that in holding (b), I would be making a serious finding against the Bank. However, subject again to any arguments to the contrary, the facts speak for themselves. Furthermore, it is worth remembering that the Executive Board’s approval of the Switch Transaction was at a time when the Bank appears to have been in dire financial straits, and the enormous attraction of the transaction to the Bank (the loan represented nearly twice its annual profits for the previous year) could easily have caused a normally responsible banker to depart from his or her normally rational approach.

  96. Although there may well have been evidence on which a judge could have concluded that the Bank had turned a blind eye to the question of whether Mr Ting had authority to commit Akai to the Switch Transaction, Stone J expressly rejected such a contention, and the Court of Appeal upheld on him on that point. In those circumstances, I accept Mr Sumption’s submission that this Court should not revisit that conclusion. First, there are concurrent findings in the lower courts; secondly, it is a very strong thing for an appeal court to reverse a first instance rejection of an allegation of dishonesty (which is what blind eye ignorance probably involves – see per Lord Blackburn in Jones (1876-7) 2 App Cas 616, 629, cited above). These reasons, even taken together, do not constitute an absolute bar to reconsidering the issue, but in this case I am satisfied that such reconsideration would be inappropriate.

  97. As I have emphasised, the views I have expressed in para.95 above are provisional, albeit that they coincide with the views of the Court of Appeal. Before adopting them, it is necessary to consider the factors relied on by Mr Sumption on behalf of the Bank. They are really two-fold. First, he said that a reasonable banker could have concluded that the Switch Transaction was in the interest of Akai, or at least that the directors of Akai, and in particular Mr Ting, could have taken that view. Secondly, he relied on the fact, which is very telling, at least in principle, that Stone J had concluded that Mr Ting had apparent authority, which was properly relied on by the Bank.

  98. As to the first argument, Mr Sumption relied on the contention that Akai had a strong commercial interest in the continuing survival and success of Singer NV and indeed of what may be called the whole STC group: he invoked the following points in that connection:

    1. The companies in which STC Canada had a substantial interest were operated “very much as a commercial group”;

    2. Singer NV’s business operation represented a very important distribution network for Akai’s products; and

    3. Akai had a substantial financial interest in Singer NV’s survival, holding US$75m convertible loan stock, being a guarantor of its subsidiary, Pfaff, to the tune of just under US$15m, and because there was intra-group liability, which would have resulted in Akai having to pay up a substantial sum if Singer NV became insolvent.

  99. So far as the first instance judgment is concerned, Stone J set out his conclusions on apparent authority in some detail in paras 392-426 of his judgment. He made the point that the liquidators had accepted the Bank’s proof of debt, and warned himself against using hindsight. He then said that he was “unable to discern anything of substance occurring at the time which should have placed the Bank ‘on inquiry’, far less is there any evidence of actual knowledge on the part of the Bank either as to any breach of fiduciary duty on the part of Mr Ting, or in terms of any want of authority on his part” (para.401). He went on in para.413 to describe the Switch Transaction as “a hard-nosed commercial transaction”, even though it may “have been .... regarded by some within the Bank .... as a ‘small miracle’”. In para.417, he said that he found “neither dishonesty nor disingenuousness on the part of the Bank”, and in the next paragraph, he made the points that Akai “relied upon Singer[ NV]’s distribution network” and “had a major financial investment in Singer[ NV]”.

  100. These findings, from made by the trial judge, who is the primary fact-finder and who heard relevant oral testimony, give one considerable pause for thought, particularly when taken with Mr Sumption’s identification of the factors supporting the view that Akai had an interest in the survival of Singer NV. However, like the Court of Appeal, I have come to the clear conclusion that the Judge’s conclusion, carefully considered as it was, simply cannot stand.

  101. The best starting point for the explanation for this conclusion is to be found in paras 329-343 of Stone J’s judgment, when he was dealing with the argument that Mr Ting had actual authority to commit Akai to the Switch Transaction (which was then in issue, as it was in the Court of Appeal). Having said in para.329 that “it is unarguable but that at face value Mr Ting’s actions .... can only be reflective of a failure to consider the interests of Akai”, the Judge said at para.334 that the Switch Transaction was “[p]lainly to the financial detriment of Akai; this can permit of no contrary argument”. He then stated that “it was not in the ordinary course of business for Akai to assume the liability of another company”. In the same paragraph, he emphasised the oddity of the Switch Transaction in the light of the fact that Akai “had no equity interest” in Singer NV, and then remarked on the unusualness of the Switch Transaction, contrasting it with “the more usual commercial approach of a guarantee of Singer[ NV]’s indebtedness”.

  102. In para.335, the Judge made the points that “no account apparently was taken .... of the remaining 57% of the shareholders of Akai”, and that the Hong Kong Stock Exchange had not been informed, as it should have been, of the transaction, and, had that been done, “at the very least a public announcement would have been required”. Paragraph 336 contains a rejection of the Bank’s reliance on the alleged financial and commercial interest which Akai had in Singer NV’s survival. There was no evidence that Mr Ting had considered these features when entering into the Switch Transaction, and in any event “the matters now relied on as post-facto rationalisation .... did not on any view amount to justification for the assumption, for no consideration, of [Akai’s obligations under the Switch Transaction]”.

  103. At para.339 of his judgment, the Judge described it as “objectively .... ‘absurd’” to suggest that “it was in the interests of Akai to take up another debt of [$30m]” because of the debt already owing from Singer NV to Akai. In conclusion on the issue of actual authority, in para.342, the Judge concluded that it was “difficult to avoid the conclusion that this was a transaction adopted by Mr Ting .... in blatant disregard for the interests of Akai”. Drawing on a judgment of Mason J in Walker v Wimborne (1975-6) 137 CLR 1, he described the Switch Transaction as “offering no prospect of advantage” to Akai, “seriously prejudice[ing its] unsecured creditors”, “more than improvident”, and “adopted .... in total disregard of the interests of the company and its creditors”.

  104. It is not realistically possible to reconcile the Judge’s analysis and assessment of the justifiability of the Switch Transaction in paras 329-343 of his judgment with his analysis and assessment in the subsequent paras 392-426. It is not enough to say that the earlier passages were concerned with what Mr Ting must have appreciated, whereas the latter passages were directed to the perception of the Bank’s employees. Reading the two passages together, such an explanation plainly will not do, and closer analysis confirms that conclusion.

  105. First, in the earlier passages, the Judge relied on facts which were known to the Bank to justify his strongly expressed view that the transaction was plainly not to Akai’s advantage, and then seems to have ignored those facts in the later passages. Examples include the public shareholding in Akai, the requirement of Stock Exchange involvement, the very unusual nature of the transaction, and the “objective .... absurdity” of, in effect, pouring good money after bad. Secondly, if, as the Judge said, “at face value Mr Ting’s actions” could “only be reflective of a failure to consider the interests of Akai” and it was “difficult to avoid the conclusion that this was a transaction adopted by Mr Ting .... in blatant disregard for the interests of Akai”, then it is hard to see why that should not have been appreciated by the Bank, unless it was shutting its eyes to the point. Thirdly, if the transaction “was not in the ordinary course of business for Akai”, that should, at the very least, give rise to serious concerns about the conclusion that Mr Ting had apparent authority to commit Akai to it. Fourthly, if the Judge was right in rejecting so robustly what he called “the post-facto rationalisation” of the transaction in relation to the actual authority issue, it is hard to understand why he did not take the same line when it came to the issue of apparent authority.

  106. Over and above these points, when considering the issue of apparent authority, I consider that there are a number of questionable aspects in the Judge’s analysis. He appears to have overlooked the fact that the Bank had never had dealings with Akai before the Switch Transaction was proposed. Thus, in para.384, he said that Mr Ting’s apparent authority to commit Akai to the Switch “was confirmed by the fact that Akai invariably had permitted Mr Ting to enter into all manner of financing agreements on its behalf”. Yet the Bank had no knowledge or experience of this. Anyway, this observation wholly overlooks the extraordinary nature of the transaction.

  107. The Judge also gave quite insufficient weight in para.413 of his judgment to the effective windfall the Bank was enjoying as a result of the Switch Transaction – see the remarks of Lord Cairns in Gray (1868) LR 3 HL 1, 11, cited above. It was also inappropriate to give weight to the Akai liquidators’ acceptance of the Bank’s proof of debt: at the relevant time, they had limited resources, considerable burdens, incomplete records and little assistance from Akai, they had no evidence of Mr Ting’s misdeeds, and they did not have the Bank’s relevant internal documents which formed an important part of Akai’s case.

  108. Mr Sumption contended that the Bank was entitled to rely on the more favourable findings in paras 392-426 of the judgment, presumably because they related to the point at issue, namely apparent authority. I do not agree. Where a Judge makes inconsistent findings, an appellate court must either choose between the two findings on some principled basis, or try and make its own findings, or, if neither of these alternatives is possible, send the case back for further findings or a retrial. The Court of Appeal rightly did not send the case back, but effectively concluded that the Judge’s reasoning and conclusion in paras 392-426 could not stand. For the reasons I have given, which are much the same as their reasons, I agree.

  109. I have not so far dealt specifically with the points made by Mr Sumption as to why the Bank could have believed that the Switch Transaction was for the benefit of Akai. The argument that the companies in which STC Canada had a substantial interest were run as a single entity is scarcely helpful to the Bank: the whole point of Akai’s case is that that patently should not have happened. In any event, it must be irrelevant to the question of apparent authority, as the Bank had had no prior dealings with Akai. The points based on Akai’s financial interest in saving Singer NV carried no weight with the Judge when considering actual authority: as I have mentioned, he clearly thought that this involved saying that Akai was justified in pouring in good money after bad.

  110. As the Judge said, these points are in any event speculative: there is no reason to think that they would have been advanced by Mr Ting if he had been challenged about the transaction. He was not so challenged, because, as stated above, the Bank was only concerned about its own position, and did not at any stage consider the question of whether there could have been any commercial benefit to Akai in entering to the Switch Transaction. In any event, these points do not begin to meet many of the arguments which justify the conclusion that the Bank was irrational in relying on Mr Ting’s alleged authority (e.g. the unusual and large nature of the transaction, the 57% public shareholders, the Hong Kong Stock Exchange factor, and Mr Ting’s patently conflicted position).

  111. In these circumstances, subject to any point which can be made by the Bank in relation to the ExCo minutes, to which I now turn, I would reject the contention that it was open to the Bank to hold Akai to the Switch Transaction on the ground that Mr Ting had apparent authority to commit Akai to it.

    The effect of the ExCo minutes

  112. Both Akai and the Bank contended that their respective cases on apparent authority was reinforced by the production of the ExCo minutes at the meeting on the 4 December 1998, when almost all the documents relating to the Switch Transaction were signed and handed over to the Bank’s representatives by Mr Ting.

  113. The Judge effectively decided that the minutes were of little, if any, relevance to the question of apparent authority, because they were merely required by the Bank for formal reasons. In effect, he said, the ExCo minutes did not detract from the Bank’s contention that it relied on Mr Ting’s apparent authority to commit Akai to the Switch Transaction, by virtue of his role as executive chairman and chief executive officer of Akai. The Court of Appeal took a very different view. They held that the Bank’s requirement for a document such as the ExCo minutes was fatal to the Bank’s case on apparent authority: the Bank could not say that it relied on his position in Akai to clothe him with authority to commit Akai if, at the same time, it had required documentary evidence of his authority to do so.

  114. Subject to my views as to the effect of the contents of those minutes, I am in agreement with Stone J, rather than the Court of Appeal, albeit for slightly different reasons. I do not think the Court of Appeal would have been entitled to interfere with the Judge’s conclusion that the Bank regarded the provision of the ExCo minutes as effectively only a formality.

  115. It does not seem to me that the provision of the minutes undermined the Judge’s conclusion on reliance. It was Mr Ting who produced those minutes, Mr Ting who signed the minutes, Mr Ting who was recorded as presiding at the meeting referred to in the minutes, and Mr Ting whose authority and reputation founded the whole of the Bank’s case on apparent authority. In those circumstances, the production of the minutes at the 4 December 1998 meeting to satisfy the Bank’s formal requirements seems to me to have been simply part and parcel of Mr Ting’s alleged authority: it was because he produced them and was clearly so closely involved in the alleged decision that the minutes were accepted as effective.

  116. The Bank sought to rely on the minutes to bolster its case on apparent authority. As just mentioned, the minutes were produced and signed by Mr Ting, and to invoke them to support the contention that he had apparent authority might well fall foul of the general rule laid down in Armagas [1986] AC 717, 777C-778D, cited above. It is unnecessary to decide that point, however, because examination of the ExCo minutes appears to me to justify the contention that, rather than reinforcing Mr Ting’s apparent authority, they actually weaken the Bank’s case on reliance, as they should have added to the Bank’s concerns as to whether Mr Ting really could have had authority to bind Akai to the Switch Transaction.

  117. Before summarising the features of the minutes which I consider justify this conclusion, it is right to refer to Akai’s bye-laws, which were in the possession of the Bank in anticipation of the Switch Transaction. Although one must be careful of attributing knowledge of such bye-laws to a third party, it appears to me right to do so in this case, given that the Bank had been specifically provided, at its request, for the very purpose of the transaction. The bye-laws prohibited Mr Ting from any involvement in the approval of the Switch Transaction. In particular, given his interest in Singer NV, arts 103 and 104 required him both to disclose his interest and to refrain from voting on the transaction. Not only do these bye-laws thus cast a shadow over the Bank’s primary case on Mr Ting’s authority to commit Akai to the Switch Transaction, as already mentioned: they impinge on anybody reading the ExCo minutes in the context of the transaction.

  118. In particular, the minutes:

    • Are not, and do not purport to be, minutes of a meeting of Akai’s board of directors, which according to the Bank’s own evidence, is what one would expect even in a “normal” transaction – a view supported by the documents relating to the Pfaff transaction;

    • Contain no disclosure of interests, notwithstanding the position of patent conflict of Mr Ting and, indeed, Dr Holmes;

    • Contain no suggestion that Mr Ting or Dr Holmes did not vote at the alleged meeting;

    • Do not describe the nature or purpose of the Switch Transaction let alone its true purpose, which was to cause Akai to assume Singer NV’s liability to the Bank;

    • Do not identify any benefit to Akai from the Switch Transaction or, indeed, any purpose for the loan;

    • Were signed only by Mr Ting, who was in the most obvious position of conflict;

    • Disclose no involvement in the purported approval process by any of the independent, non-conflicted directors of Akai; and

    • Make no reference to the public shareholders of Akai, much less to any consideration of any regulatory requirements.

  119. These features were either not noted or ignored by the Bank. Although the Bank had Akai’s bye-laws in its possession, there is no suggestion that the minutes were checked against the bye-laws, as was the Bank’s normal practice, and the practice of a reasonable Thai bank. Moreover, also contrary to its usual practice the Bank did not obtain any legal advice from a lawyer qualified to assess whether the minute was in fact proper evidence of Mr Ting’s authority.

  120. It is worth emphasizing that the ExCo minutes did not purport to authorize Mr Ting to commit Akai to the very aspect of the Switch Transaction which was so extraordinary, namely using the US$30m facility to pay off the liability of Singer NV to the Bank. What the minutes purported to authorize Mr Ting to do was agree the loan facility, and to draw on it, as well as to pledge Akai Electric shares as security. All those aspects of the Switch Transaction were unexceptionable. The fact that there was no mention of the extraordinary aspect of the transaction both undermines the Bank’s argument that it reasonably relied on the minutes and strengthens Akai’s case that the Bank’s belief in Mr Ting’s authority was irrational. Particularly as a representation has to be “clear and unequivocal”, the absence of any mention of discussion, let alone approval, of the US$30m being used to pay off Singer NV’s liability must destroy the Bank’s case that the minutes support the rationality of its belief that Mr Ting was authorized to commit Akai to that course. Indeed, given my view of the position in the absence of the ExCo minutes, I consider that the absence in the minutes of any reference to the use to which the US$30m was to be put actually renders the Bank’s case on rationality even weaker.

  121. In my opinion, these failings on the part of the Bank in relation to the ExCo minutes provide significant reinforcement for the view that the Bank was irrational in its belief that Mr Ting had authority to commit Akai to the Switch Transaction, with all its remarkable features as discussed above. I am therefore of the opinion that the ExCo minutes assist the conclusion that the Bank’s case on apparent authority must fail.

    What is the correct amount to be awarded to Akai?


  122. On the basis that Akai is entitled to succeed in its claim that Mr Ting had no authority to enter into the Switch Transaction on its behalf, the Court of Appeal awarded Akai the sum of US$22,541,332 plus compound interest, which was also the sum which the Judge would have awarded if he had found for Akai on that issue. This sum is calculated on the basis that the Bank received a total of US$20,504,295 for the Shares when it sold them, with the balance being accounted for by interest and fees.

  123. This is clearly the correct sum to have awarded Akai by way of common law damages. As Mr Ting had no authority to enter into the Switch Transaction on behalf of Akai, then, as between the Bank and Akai, the Switch Transaction was void, and the Bank had no right to retain the share certificates or to sell the Shares. Accordingly, Akai had a claim for conversion of the Shares. On that basis, it is clear that Akai’s damages should be measured by the value of those shares at the date they were sold: see BBMB Finance (Hong Kong) Ltd v Eda Holdings Ltd [1990] 1 WLR 409. The Privy Council’s decision in that case is accurately summarised in the headnote, [1990] 1 WLR 409, 409H-410A:

    [T]he general rule [is] that where property [is] irreversibly converted the measure of damages for conversion [is] the value of the property at the date of conversion, and that, .... since the plaintiffs’ shares had been irrevocably sold by the defendant, the plaintiffs were entitled to damages for that conversion .... at the date of conversion ....

    As the Shares were sold on the open market by the Bank, the value of those shares must be measured by reference to what the Bank obtained for them.

  124. However, by its cross-appeal, Akai contends that it is entitled to a greater sum than $22,541,332, because, over and above its claim based on want of authority and conversion, it has a claim against the Bank based on the proposition that, when they accepted the Electric shares in December 1998, or in the alternative in December 1999, the Bank at once became liable to Akai for knowing receipt. Hence Akai’s cross-appeal.

  125. Akai’s case in this connection may be summarised as follows:

    1. The classic test for a claim in knowing receipt is whether trust property was received in circumstances where there is unconscionability on the part of the recipient;

    2. The facts which justify Akai’s case against the Bank on apparent agency also justify its case on unconscionability;

    3. Accordingly, Akai has a claim against the Bank for knowing receipt in respect of the Shares;

    4. Given that Akai has claims based on conversion and knowing receipt, it can elect between the two claims when it comes to assessing compensation;

    5. While damages based on the conversion claim are assessed according to common law principles, equitable compensation for knowing receipt are assessed on a different basis;

    6. In this case, equitable compensation should be assessed by the value of the Shares, as at either

      1. when they were wrongly received by the Bank, or

      2. when the Bank wrongly refused to hand them back, namely by reference to their value in (i) December 1998 or (ii) December 1999, as either basis would produce a higher level of compensation than assessment by reference to the proceeds of sale; and

    7. Accordingly, Akai is entitled to (i) US$50,775,000 being the value of the retained shares at the time they were wrongly received by the Bank, or (ii) US$32,904,372, the value of the shares on 6 December 1999, rather than the US$20,504,295, which the Bank received for those shares when it sold them (plus, in all cases, plus interest and fees).

  126. Proposition (a) is not in contention, and, provided Akai establish propositions (b) and (c), propositions (d) and (e) are not in contention either. However, propositions (b), (c) and (f) are challenged by the Bank. Proposition (g) is not precisely agreed, but it raises no issue of significance, if propositions (a) to (f) are made out.

    Akai’s propositions which are not in dispute

  127. So far as Akai’s proposition (a) is concerned, the characterisation of the precise state of mind on the part of a defendant needed to render him liable for knowing receipt has been the subject of much debate in judgments and academic articles. In Bank of Credit and Commerce International (Overseas) Ltd v Akindele [2001] 1 Ch 437, the law on the topic was reviewed by Nourse LJ in a full and careful judgment (with which Ward and Sedley LJJ agreed). At [2001] 1 Ch 437, 455, Nourse LJ concluded that:

    [J]ust as there is now a single test of dishonesty for knowing assistance, so ought there to be a single test of knowledge for knowing receipt. The recipient’s state of knowledge must be such as to make it unconscionable for him to retain the benefit of the receipt. A test in that form, though it cannot, any more than any other, avoid difficulties of application, ought to avoid those of definition and allocation to which the previous categorisations have led. Moreover, it should better enable the courts to give commonsense decisions in the commercial context in which claims in knowing receipt are now frequently made ....
  128. This approach was subsequently followed by the English Court of Appeal in Criterion Properties plc v Stratford UK Properties LLC [2003] 1 WLR 2108, although a little doubt may be cast on it by an uncharacteristically obscure remark of Lord Nicholls of Birkenhead in Criterion in the House of Lords – [2004] 1 WLR 1846, para.4, first sentence. Nonetheless, in Charter v City Index Ltd [2008] Ch 313, para.8, the English Court of Appeal reaffirmed that the conclusion reached in Akindele [2001] 1 Ch 437 represented the law. In the light of this, without deciding the point, I am prepared to proceed on the basis of Mr Sumption’s realistic concession, and to assume that proposition (a) is correct.

  129. Before turning to the contentious propositions, it is right briefly to explain why I am also prepared to proceed on the basis that propositions (d) and (e) are correct, and what is at stake in relation to proposition (g).

  130. Proposition (d) rests on the general principle that, where a person can bring a claim on two alternative bases, he can elect to proceed on the basis which is more beneficial to him. This proposition is not challenged - unsurprisingly, as it receives support from three decisions of the House of Lords, Henderson v Merrett Syndicates Ltd [1995] 2 AC 145, 193, Kleinwort Benson Limited v Lincoln City Council [1999] 2 AC 349, 387 (in each case per Lord Goff of Chieveley), and Deutsche Morgan Grenfell Group Plc v Inland Revenue Commissioners [2007] 1 AC 558, para.51 (per Lord Hope of Craighead).

  131. As for proposition (e), the notion that equitable compensation is assessed on a somewhat different basis from common law damages is clearly right (albeit that the difference can be overstated). That is clear from another decision of the House of Lords (to which it will be necessary to return). In Target Holdings Limited v Redferns [1996] 1 AC 421, 434D-G, Lord Browne-Wilkinson said this:

    .... If specific restitution of the trust property is not possible, then the liability of the trustee is to pay sufficient compensation to the trust estate to put it back to what it would have been had the breach not been committed .... Even if the immediate cause of the loss is the dishonesty or failure of a third party, the trustee is liable to make good that loss to the trust estate if, but for the breach, such loss would not have occurred .... Thus the common law rules of remoteness of damage and causation do not apply. However there does have to be some causal connection between the breach of trust and the loss to the trust estate for which compensation is recoverable, viz. the fact that the loss would not have occurred but for the breach ....

    The same view was taken by the High Court of Australia in Maguire v Makaronis (1997) 188 CLR 449 – see especially at 188 CLR 449, 469-470 per Brennan CJ, citing Target [1996] 1 AC 421 with approval.

  132. Proposition (g) concerns the precise valuation of the Shares in December 1998 or December 1999, and is only relevant if Akai succeeds in establishing its claim for more compensation than the Judge would have awarded, and the Court of Appeal did award. Although I have little hesitation in rejecting the argument (for reasons given below) I can see some logical basis for arguing that compensation for knowing receipt should be assessed by reference to the value of the Shares in December 1998, i.e. by reference to what might be said to be the date of receipt. However, I cannot see any basis for the December 1999 date: that was the date on which the US$30m became repayable under the Loan Agreement. If that agreement was valid, then the share certificates were lawfully retained, as they were security for repayment of the loan, and it was not repaid. If, on the other hand, that agreement was void (as was the case for treasons already given), then ex hypothesi the date has no relevance.

  133. At the hearing, we were told that there were small disagreements between the parties as to the valuations of the Shares as at those dates. If any of those disagreements had been relevant to the assessment compensation, I would have thought it right to refer them to the High Court, in the event of Akai succeeding on its propositions (b), (c) and (e), the three propositions which I now turn to consider.

    Proposition (b): the requisite state of mind of a recipient in knowing receipt

  134. Proposition (b) raises the issue of the Bank’s state of mind when it received the share certificates in December 1998. Applying the test laid down in Akindele [2001] 1 Ch 437, 455, cited above, the question is whether its state of mind when it accepted the share certificates was such that it would be unconscionable for it to have retained the Shares. I have already explained that the Bank was irrational, but not dishonest, in believing, or more accurately in assuming, that Mr Ting had authority to commit Akai to the Switch Transaction. In my view, subject to considering proposition (c), the facts of this case provide a good illustration of how it can be unconscionable for a party to retain property in circumstances which do not involve dishonesty, but which do involve irrationality.

  135. Indeed, particularly in a commercial context, the test which equity applies to a claim in knowing receipt of an asset is effectively identical to the test which the common law would apply to determine whether there was what I have called reasonable reliance on the apparent authority of an alleged agent to commit the principal to handing over the asset. It makes little commercial sense, and would provide a fertile source of confusion and inconsistency, if the tests were different. It is true that common law talks about actual knowledge and presumed knowledge whereas equity sometimes refers to actual notice and constructive notice. But, above all in a case involving knowing receipt in an arm’s length commercial context, I consider that, at any rate absent very special facts, equity would follow the law.

  136. This conclusion receives support from the penetrating analysis in the judgment of Sir Robert Megarry, V-C in In re Montagu’s Settlement Trusts [1987] 1 Ch 264, 277D-F, 278B-D, and 285B-F (especially propositions (4) and (5)). This judgment was described as “seminal” in Akindele [2001] 1 Ch 437, 452C, by Nourse LJ, who summarised its effect at [2001] 1 Ch 437, 453D as being “that, in order to establish liability in knowing receipt, the recipient must have actual knowledge (or the equivalent) .... and that constructive knowledge is not enough”.

  137. If the recipient’s reliance on the alleged agent’s apparent authority, when accepting the asset from the alleged agent on behalf of the principal, was dishonest or irrational, it seems to me that it would be unconscionable for the recipient to retain the asset against the wishes of the principal, or, to put it another way, the recipient would have the relevant “actual knowledge (or the equivalent)”. On the other hand, if the reliance was merely negligent, then I doubt that the unconscionability test would, at least normally, be satisfied – at best it would amount to “constructive knowledge”.

    Proposition (c): is there a claim for knowing receipt in this case?

  138. That brings me to Akai’s proposition (c), which raises the question of whether there is in fact a claim for knowing receipt. Mr Sumption’s argument that there is no such claim, is based essentially on two reasons. The first is that no legal or equitable title passed in the share certificates or the Shares, when the certificates were received by the Bank, because the Switch Transaction was void: accordingly, there is no “receipt” in respect of which a knowing receipt claim can be mounted. The second reason is that, especially in the commercial context, where the claimant has a perfectly good common law claim, it is both unnecessary and undesirable to permit him to invoke equity: his claim should be satisfied by his common law claim, and an equitable claim would create a proprietary interest which would rank ahead of subsequent equitable interests, or legal interests acquired with notice of the proprietary interest. Both these reasons are said to be supported by the reasoning of Lord Browne-Wilkinson in Westdeutsche Landesbank Girozentrale v Islington London Borough Council [1996] AC 669.

  139. So far as the first reason is concerned, at [1996] AC 669, 706E-F, Lord Browne-Wilkinson explained that “[u]nless and until there is a separation of the legal and equitable estates, there is no separate equitable title”. In this case, says Mr Sumption, legal and equitable title in the Shares remained with Akai until the shares were sold, and therefore there can be no claim against the Bank for knowing receipt of the Shares.

  140. The arguments on this point were not developed in any detail by either party, and, in the light of my conclusion on Akai’s proposition (f), it is unnecessary to resolve it. Because it may give rise to a rather controversial and difficult issue of some significance in other cases, I am, in these circumstances, rather reluctant to express a firm conclusion on the point.

  141. It may well be that the correctness of Mr Sumption’s contention stands or falls on determining whether the Bank had an equitable interest in the Shares at the moment that the share certificates were pledged. That in turn may be resolved by construing the pledge agreement (and possibly other documents signed on 4 December 1998). If the Bank merely had a contractual right to sell the Shares, which right was protected by the pledging of the certificates, then, at least until the Shares were sold and the proceeds of sale retained by the Bank, there may have been no asset or money in the possession of the Bank on which a claim for knowing receipt could bite. However, even if that is right, when the Shares were sold, the proceeds of sale were retained by the Bank as its own money, so there would, I think, be a claim for knowing receipt of the proceeds of sale of the Shares.

  142. This is not quite the point which Lord Browne-Wilkinson had in mind in the Westdeutsche case [1996] AC 669, 706E-F, as he was addressing an issue relating to an alleged resulting trust. More in point is another passage relied on by Mr Sumption, in the opening two sentences of the judgment of Hoffmann LJ in El Ajou v Dollar Land Holdings plc [1994] 2 All ER 685, 700:

    This is a claim to enforce a constructive trust on the basis of knowing receipt. For this purpose the plaintiff must show, first, a disposal of his assets in breach of fiduciary duty; secondly, the beneficial receipt by the defendant of assets which are traceable as representing the assets of the plaintiff; and thirdly, knowledge on the part of the defendant that the assets he received are traceable to a breach of fiduciary duty.

    Even though it emanates from Lord Hoffmann, this should not, I acknowledge, be treated as a comprehensive definition of the essential ingredients of a knowing receipt claim in every case.

  143. Nonetheless, it may usefully highlight the need for “beneficial receipt” of an asset or money by a defendant in a knowing receipt claim. It is true that, by holding the certificates, the Bank did receive a benefit, but it was merely a protection of its alleged right to sell the Shares, and Akai’s claim relates to the Shares. (There is, I suppose, an argument, which, even for the purpose of this hypothesis, need not be considered, that when the Bank actually purported to exercise its right of sale of the Shares, there was a scintilla temporis before sale when it was exercising sufficient degree of control over the Shares to have beneficial ownership of them.)

  144. I propose to take this point no further. Even if it is correct to says that no claim in knowing receipt could be made in respect of any time before the Bank acquired an equitable interest in the Shares or proceeds of sale, it would be necessary to construe the pledge agreement (and, possibly, other documentation) in order to be able to identify and characterise the Bank’s rights in the Shares, and we heard no submissions on that issue.

  145. I would, however, reject the Bank’s alternative contention that, as a matter of principle, there should be no equitable claim in this area. It is true that in Westdeutsche [1996] AC 669, 704G-705B, Lord Browne-Wilkinson warned against “the wholesale importation into commercial law of equitable principles inconsistent with the certainty and speed which are essential requirements for the orderly conduct of business affairs”. The question when it is right to invoke an equitable right in a commercial context is not always easy to resolve. There are dicta which, if read in isolation, appear to provide some support for the Bank’s case on the point – see e.g. per Millett LJ in Paragon Finance Ltd v D B Thakerar & Co. [1999] 1 All ER 400, 409e-g, and, as Lord Millett, in Dubai Aluminium Co Ltd v Salaam [2003] 2 AC 366, 404D-H.

  146. However, in Bristol and West Building Society v Mothew [1998] 1 Ch 1, 18A-F, Millett LJ, after pointing out that the mere fact that a person with fiduciary duties to another had been negligent would not of itself give rise to an equitable remedy, referred to:

    .... those duties which are special to fiduciaries and which attract those remedies which are peculiar to the equitable jurisdiction and are primarily restitutionary or restorative rather than compensatory. A fiduciary is someone who has undertaken to act for or on behalf of another in a particular matter in circumstances which give rise to a relationship of trust and confidence .... A fiduciary must act in good faith; he must not make a profit out of his trust; he must not place himself in a position where his duty and his interest may conflict; he may not act for his own benefit or the benefit of a third person without the informed consent of his principal.


    .... Breach of fiduciary obligation, therefore, connotes disloyalty or infidelity. Mere incompetence is not enough.

    This seems to me to apply here: Mr Ting clearly owed a fiduciary duty to Akai, which he equally plainly breached when he intentionally, indeed dishonestly, purported to commit Akai to the Switch Transaction, without its authority or knowledge, for the benefit of Singer NV.

  147. Mr Sumption also relied on Criterion [2004] 1 WLR 1846, where the House of Lords held that a claim for knowing receipt was effectively irrelevant in a case where the claimant’s case was based on want of authority of an alleged agent. I do not consider that that case assists. It involved an executory contract, so a claim for knowing receipt could not arise as the contract had not been completed, and no property had changed hands. If the claim based on want of authority succeeded, the contract would have been void and unenforceable, so there could be no receipt. If, on the other hand, the claim based on want of authority failed, the House of Lords took the view that the claim for knowing receipt would also fail. (This appears to me to imply fairly strong support for my view that what is required of the recipient’s state of mind in a knowing recipient claim in a case such as this is the same as the test of want of honest and rational belief for the purpose of defeating apparent authority.) Mr Kosmin suggested that a different result might obtain if the potential recipient acquired knowledge between contracting with the agent and receiving the property; that could be right, but that takes matters no further here.

    Proposition (f): the measure of equitable compensation

  148. I turn to Akai’s proposition (f), namely that the measure of equitable compensation should be the value of the Shares at the date they were first received by the Bank, in December 1998, or the date the Bank refused to hand them back, in December 1999. In a nutshell, assuming in Akai’s favour that a claim for knowing receipt arose in December 1998, its primary case is that it is entitled, as a matter of principle, to be compensated, if it so elects, by reference to the value of those shares as at that date. (Alternatively, the Bank should have handed the share certificates back in December 1999, which is the alternative valuation date alleged by Akai.)

  149. In support of the contention that, if the claim for knowing receipt arose in December 1998, equitable compensation should be assessed by reference to the value of the Shares as at that date, Mr Kosmin relied on dicta in a number of cases. Some of those cases were concerned with the duty of a person in a fiduciary position to account for a profit or a bribe; others arose out of a trustee’s duty to make good a loss made on an unauthorised investment. In my view, none of those circumstances are in point here, even as useful analogies.

  150. Thus, I do not see how Akai gets help from the proposition which was expressed by Nourse LJ in these terms in In re Duckwari Plc [1999] Ch 253, 262H:

    If a trustee applies trust moneys in the acquisition of an unauthorised investment, he is liable to restore to the trust the amount of the loss incurred on its realisation: see Knott v Cottee (1852)16 Beav. 77.

    In a case where a trustee acquires an unauthorised investment or wrongly retains an investment (as in Fry v Fry (1859) 27 Beav 144), one can well see the logic of holding that the beneficiaries should have the option of keeping the investment or requiring the trustee to reimburse the trust the cost of the investment: if the investment has gone down in value, the trust will have suffered a loss through the wrongful act of the trustee, which he ought to be required to make good.

  151. However, it is important to recall that a basic principle applicable to equitable compensation is that, as already quoted, “there does have to be some causal connection between the breach of trust and the loss to the trust estate for which compensation is recoverable viz. the fact that the loss would not have occurred but for the breach” – per Lord Browne-Wilkinson in Target Holdings [1996] 1 AC 421, 434G. That chimes with what he said at [1996] 1 AC 421, 432E-G, 437C-E, 439A-B and 440E-F. In the third of those passages, after referring with approval to certain passages in the judgment of McLachlin J in the Canadian case of Canson Enterprises Ltd v Boughton & Co. (1991) 85 DLR 129, 160-163, he said this:

    Equitable compensation for breach of trust is designed to achieve exactly what the word compensation suggests: to make good a loss in fact suffered by the beneficiaries and which, using hindsight and common sense, can be seen to have been caused by the breach.
  152. The same point underlies the decision of the Court of Appeal in Mothew [1998] 1 Ch 1. The same approach, as I have indicated, has been adopted in Australia and Canada. As McLachlin J expressed it in Canson Enterprises 85 DLR 129, 163, “the losses made good are only those which, on a common sense view of causation, were caused by the breach”. Kirby J put it this way in Maguire 188 CLR 449, 496:

    [Equitable] remedies will be fashioned according to the exigencies of the particular case so as to do what is ‘practically just’ as between the parties. The fiduciary must not be ‘robbed’; nor must the beneficiary be unjustly enriched.
  153. In this case, it appears to me clear that equitable compensation should be assessed by reference to the value of the Shares at the date when they were sold by the Bank, even if the claim for knowing receipt arose some 18 months earlier. As Mr Sumption pointed out, until the Shares were actually sold, it was always open to Akai to recover them from the Bank. Even more significantly, it seems to me clear that, if the Bank had not accepted the share certificates and retained them until selling the shares, Akai would have retained those shares until Akai Electric was placed into [administration] by the Japanese courts. The uncontested evidence very strongly suggests that this is so: Akai Electric was Akai’s effective operating subsidiary; there was no history of the STC group transferring its assets, let alone its shareholdings in subsidiaries, out of its ultimate control; after the Bank returned the 5.5m Akai Electric shares to Akai, they were retained by Akai until they became valueless.

  154. In those circumstances, it is clear on the balance of probabilities, indeed it is, in truth, clear beyond any real doubt, that, if the Bank had returned the share certificates to Akai, far from being sold earlier than the date upon which the Bank sold them, Akai would have kept the Shares until they had become worthless. The ironic fact is that Akai is substantially better off as a result of the Bank having received and sold the Shares. Of course, that does not mean that Akai has no right to equitable compensation, and the Bank is entitled to keep the proceeds of sale of the Shares: normal equitable principles entitle Akai to elect between receiving a sum equal to the proceeds of sale of the Shares or, unless it is impossible to obtain them, an equivalent number of Akai Electric shares.

  155. Accordingly, even though I consider that, although Akai may well have a claim in knowing receipt, it takes matters no further. I should add that, even if the principles so far discussed had suggested that Akai was entitled to equitable compensation in an amount greater than it should recover by way of common law damages, I would have been very sympathetic to the notion that the equitable remedy would have to be refashioned so as to equate the amount of such compensation with the common law damages. In that connection, I draw support from what was said by Lord Browne-Wilkinson in Target Holdings [1996] AC 421, 432F-433A (already referred to) and 435F-G, and by Millett LJ in Mothew [1998] Ch 1, 17G-H.

    Issues on interest

  156. Towards the very end of the oral argument, the question was raised as to whether the Judge was right

    1. to have awarded Akai compound interest on the damages or compensation, and

    2. to have ordered that such interest should start to run from 6 December 1999.

    Neither point had been raised before the Court of Appeal or in the Bank’s notice of appeal or written case, but Mr Sumption indicated that he wanted leave to amend the bank’s notice of appeal, at least in relation to issue (b).

  157. I would not allow the Bank to raise the issue of compound interest. The circumstances in which interest should be awarded on a compound basis are not always entirely clear and it seems to me that it would be unfair on Akai to allow it to be raised at the last minute, and determined without full argument. Further, there is no commercial injustice in requiring a bank to pay compound interest on money it has wrongly been holding for a few years, at least in circumstances such as the present.

  158. The issue of the date from which interest is to run stands on a different footing. If, as the Court of Appeal rightly held, damages or compensation should be based on the proceeds of sale received by the Bank for the Shares, then I can see no possible justification for ordering that interest should run from any date earlier (or indeed later) than the date (or dates) on which the proceeds of sale were received by the Bank (unless there was some unjustifiable delay in getting in the proceeds). In particular, there is no possible justification for choosing the date on which the US$30m became repayable under the Loan Agreement, not least because the reason the claim has succeeded is that that agreement was void. The point is really the same as that already discussed in relation to Akai’s proposition (g) on the measure of the compensation payable.

  159. There was an argument that the Bank delayed unnecessarily in selling the Shares. In my view, in agreement with the Court of Appeal, that argument plainly fails on the facts: the Bank understandably considered that it should get certain regulatory clearances before it sold the Shares, which delayed the sale for a few months. Although it is unnecessary to decide the point, I strongly suspect that that argument fails on the law as well: the Bank was not under a duty to sell the Shares with any particular despatch, or at all, so far as I can see.

  160. In order to avoid the risk of any argument that the Bank took an unreasonable time to get in the proceeds of sale, and in order to ensure that Akai is not prejudiced in that connection, I think it would be fair if interest ran from the date (or, more accurately, the dates) of sale of the Shares. After all, the Bank is very much in mercy by having raised this point so very late.


  161. It would be wrong to end this judgment without referring to the quality of the judgments below, from all of which I have derived considerable assistance. I have inevitably referred to aspects of the judgments in the courts below which appear questionable. However, this should not detract from the fact that Stone J delivered an admirably full and carefully considered judgment, and, although I have not made much specific mention of the judgments in the Court of Appeal, perusal of those judgments will reveal that I have applied and followed most of the points made in the incisive and well reasoned judgments of Tang VP, Le Pichon and Cheung JJA.

  162. In the event, for the reasons given above, my conclusions are that

    1. Mr Ting did not have authority to commit Akai to the Switch Transaction, and

    2. the Bank is limited to the common law measure of damages.

    Although I do not agree with all their reasoning on issue (a), I consider that the Court of Appeal reached the right conclusion on both those two issues. Accordingly, I would dismiss both the Bank’s appeal and Akai’s cross-appeal, but I would vary the award of interest so that it runs from the dates on which the Bank sold the Shares.

  163. So far as costs are concerned, I would order that they be dealt with on written submissions as to which the parties should seek written directions from the Registrar.

    Chief Justice Ma

  164. For the above reasons, the Bank’s appeal and Akai’s cross-appeal are dismissed. The Order of the Court of Appeal dated 10 August 2009 is varied only to the extent that the award of compound interest (on the damages totalling US$22,541,332) will begin to run as from the dates on which the Bank sold the Shares. Costs will be dealt with as indicated in para.163 above.

    21 JANUARY 2011

    Lord Neuberger of Abbotsbury NPJ

  165. On 8 November 2010, the court dismissed the Bank’s appeal on liability (FACV 16 of 2009), dismissed Akai’s cross-appeal on quantum (FACV 9 of 2010), varied the Order of the Court of Appeal to the minor extent of adjusting the date from which interest on the damages awarded should begin to run, and ordered that costs should be dealt with by way of written submissions.

  166. The written submissions (and counter-submissions) on costs have now been received, and they reveal a relatively small degree of dispute between the parties. They agree that

    1. the Bank should pay at least a large proportion of Akai’s costs of the Bank’s unsuccessful appeal, and that

    2. Akai should pay all the Bank’s costs of Akai’s unsuccessful cross-appeal, such costs in each case to be taxed if not agreed.

  167. The difference between the parties relates to the costs of the Bank’s unsuccessful appeal. Akai contends that it should receive 100% of its costs of that appeal, whereas the Bank contends that Akai should receive only 70% of its costs.

  168. It is common ground that the prima facie position is that the Bank should pay 100% of Akai’s costs of the Bank’s appeal, on the simple ground that the Bank lost, and Akai won, the appeal. However, the Bank argues that, given that Akai succeeded in having the appeal dismissed on the ground that Mr Ting did not have authority to bind it, the costs of the appeal were inflated because Akai ran an unnecessary additional argument based on knowing receipt. Accordingly, argues the Bank, Akai should not have its costs in so far as they were attributable to that unnecessary argument.

  169. In this case, at least in the absence of Akai’s cross-appeal, the position is as follows. Akai relied both on want of authority and on knowing receipt to make out its case for defeating the Bank’s appeal, and it succeeded on both arguments, although the outcome would have been the same if it had run, and succeeded, on only one of those arguments. Where a party advances its case on two different grounds and it succeeds on both grounds, then, unless the court considers that it was unreasonable for that party to have relied on both grounds, or there is some other special reason, it would be wrong to deprive that party of its costs in connection with maintaining both grounds.

  170. That seems to be right as a matter of principle, and it is consistent with Order 62 rule 7(1) of RHC, which refers to things done “improperly or unnecessarily”: in this context, the word “unnecessarily” should not be construed as applying to anything done which was not, with wisdom of hindsight, strictly necessary. Where a party has two alternative ways of putting his case, it is normally not unreasonable to run them both, especially in a case, such as this, where a reasonably substantial amount of money is at stake, and it is therefore hard to argue that it was disproportionate to raise two arguments.

  171. In this case, again at least if one ignores the cross-appeal for the moment, it was proper for Akai to have run its case on the Bank’s appeal by reference to knowing receipt as well as want of authority. In the event, the two lines of argument involved the same considerations: if there had been no want of authority, there would have been no knowing receipt; and given that there was want of authority, there was also knowing receipt. However, the fact that that was the outcome of a week’s argument cannot possibly of itself mean that it was unreasonable of Akai to have run both points.

  172. Further, as is rightly emphasised in Akai’s submissions, the Court of Appeal decided the issue raised on the Bank’s appeal in Akai’s favour on the basis of knowing receipt rather than want of authority. That renders it particularly hard to criticise Akai for running the knowing receipt argument on the appeal in this court.

  173. That analysis suggests that Akai should have 100% of its costs of the Bank’s appeal. The only point which gives one pause for thought is that, quite apart from running the knowing receipt argument to support its case on the Bank’s appeal, Akai had an additional reason for running the knowing receipt argument, namely in order to support its case on its unsuccessful cross-appeal: unless it succeeded on knowing receipt, Akai would have accepted that its case on the cross-appeal would have been bound to fail.

  174. On closer consideration, this argument does not assist the Bank. As a matter of principle, if the knowing receipt argument was, in reality, part of Akai’s case on its cross-appeal, rather than part of its case on the Bank’s appeal, then the costs attributable to the argument would be borne by Akai as part of the costs of the cross-appeal, which it is common ground that it should bear. However, it is noteworthy that the Bank does not put its case in that way – and rightly so. Akai ran the knowing receipt argument in connection with both the Bank’s appeal and its own cross-appeal.

  175. It is ultimately a matter for the taxation process as to whether the costs incurred in connection with that argument are to be treated as

    1. attributable only to the appeal (on the basis that the costs of the cross-appeal are only to be treated as incurred in so far as they would not have been incurred on the appeal in any event), or

    2. apportioned (possibly but not necessarily equally) between appeal and cross-appeal.

  176. If option (a) is adopted on taxation, then, given that the costs of arguing the knowing receipt point were, for the reasons discussed above, reasonably incurred by Akai in connection with the appeal, such costs should, in principle, be recovered from the Bank, and, in so far as there were any extra costs incurred in relation to the argument on the cross-appeal, they would be borne by Akai. If option (b) is adopted, the costs of the argument will be borne in part by each party. On either approach, there would be no reason to depart from the prima facie view that the Bank should pay 100% of Akai’s costs of the appeal.

  177. Accordingly, the order as to costs we think it right to make is that the Bank pay 100% of Akai’s costs of the Bank’s appeal, and Akai pay 100% of the Bank’s costs of Akai’s cross-appeal. As for the costs of the written argument on costs, they must be paid by the Bank, as Akai has succeeded in obtaining the order as to costs which it seeks, over the Bank’s arguments to the contrary.


Jonathan Sumption, QC and Eugene Fung (instructed by Messrs Baker & McKenzie) for the appellant in FACV 16/2009 and respondent in FACV 9/2010

Leslie Kosmin, QC and Linda Chan (instructed by Messrs Hogan Lovells) for the respondent in FACV 16/2009 and appellant in FACV 9/2010.

all rights reserved