Susilawati v American Express Bank Ltd
[2009] SGCA 8

Case Number	:	CA 140/2007
Decision Date	:	27 February 2009
Tribunal/Court	:	Court of Appeal
Coram	:	Chan Sek Keong CJ; Andrew Phang Boon Leong JA; V K Rajah JA
Counsel Name(s)	:	Davinder Singh SC and Bhavish Advani (Drew & Napier LLC) (instructed), Siraj Omar (Premier Law LLC) for the appellant; Francis Xavier, Jerome Robert, Ho Hua Chyi, Dawn Wee (Rajah & Tann LLP) for the respondent
Parties	:	Susilawati — American Express Bank Ltd

27 February 2009

V K Rajah JA (delivering the grounds of decision of the court):

Introduction

1          These proceedings originally centred on allegations of undue influence and breach of fiduciary duties levelled by the appellant, Mdm Susilawati (“the appellant”) against the respondent, American Express Bank Limited (“the respondent”). The appellant attempted to disclaim liability under a third-party charge which expressly made her responsible for losses incurred through her then son-in-law’s personal trading account with the respondent. After a four-day trial, the trial judge dismissed the appellant’s claim with costs.

2          The appellant appealed against the whole of the trial judge’s decision. However, she did not seek to challenge the trial judge’s findings of fact. Instead, the appellant sought this court’s leave to raise and press a new point that had neither been pleaded nor raised earlier. In addition to the substantive appeal, the appellant made three other applications:

(a)        for the court to order a new trial on the issues raised by the appellant in the appeal;

(b)        for leave to adduce further evidence; and

(c)        alternatively, for leave to amend the pleadings.

3          It also bears mention, so as to shed some light on this unusual change of strategic tack, that the appellant was represented by different counsel for the appeal. At the end of the hearing, we dismissed all three applications as well as the appeal. We now set out in full the grounds of our decision.

Facts

4          The essential background facts have been carefully summarised in the High Court’s judgment but are now briefly reprised here for ease of reference (see Susilawati v American Express Bank Ltd [2008] 1 SLR 237 (“the GD”) at [2]–[6]).

5          The appellant is a wealthy Indonesian citizen who was married to a prominent Indonesian businessman until his demise in April 2002. Her late husband controlled a substantial Indonesian conglomerate known as the Gajah Tunggal group. Its business concerns comprise, inter alia, rubber remilling and supply of rubber tyres for vehicles and bicycles, commodities trading, hotels and, at one time, even the largest private commercial bank in Indonesia, known as Bank Dagang Negara Indonesia.

6          The respondent is a limited liability corporation incorporated in the United States of America, providing, inter alia, private banking services to high net-worth individuals in Singapore.

7          The appellant had been a customer of the respondent’s private banking division ever since 27 August 1997, when she opened an account (“the Account”). On or about 11 February 1998, the appellant executed a document entitled “Third Party Liabilities” (“the Charge”), under which she granted a charge in favour of the respondent over all moneys in the Account to secure the due and punctual discharge of all moneys, obligations and liabilities due from her then son-in-law, Tommy, to the respondent.

8          Between 1998 and 2005, Tommy incurred substantial debts to the respondent, which consisted of losses from foreign exchange transactions that he effected through his personal account with the respondent, as well as loans made to him by the respondent. It was only in the later half of 2005 when Tommy had defaulted and ran into problems servicing his loans that the appellant began to challenge the Charge. Pertinently, it was also around this time that the marriage between the appellant’s daughter, Zina, and Tommy ran into difficulty and fell apart.

9          By March 2006, Tommy’s personal liabilities to the respondent, including interest, had ballooned into a staggering sum of about US$17.4 million. As a result of Tommy’s inability to discharge his liabilities, the respondent eventually set off a sum of US$17,560,390.98 from the Account in accordance with the terms of the Charge. Not long after, the appellant commenced this action to recover a sum of US$17,500,605 from the respondent.

10        In the proceedings below, the appellant originally rested its claim on just two central planks:

(a)        that Tommy had a relationship with the appellant of such a dominant nature as to give rise to a presumption of undue influence, and that her signature on the Charge had been procured by this undue influence; the respondent had actual or constructive knowledge of this, rendering the Charge void and unenforceable; and

(b)        that the respondent had breached its fiduciary duties to the appellant, resulting in losses that entitled her to damages.

11        The respondent, on the other hand, maintained that:

(a)        The appellant had executed the Charge freely, with the full knowledge that she was pledging the sums in her Account as security to cover Tommy’s liabilities.

(b)        Alternatively, the appellant had affirmed the Charge by continuing to operate and make investments through the Account. This made it inequitable to set aside the Charge as the respondent, relying on the appellant’s acquiescence, had acted to its detriment by continuing to make available banking facilities to Tommy.

(c)        Finally, it did not owe the appellant any fiduciary duties or, if it did, those duties were not breached.

The decision below

12        The trial judge dismissed the appellant’s claim with costs, holding that:

(a)        The appellant had failed to show that she had reposed such a degree of trust and confidence in Tommy that could justify the inference that the Charge was executed whilst under his undue influence. The execution of the Charge was entirely explicable in the context of the family elements and relationships involved here. In particular, the appellant was family-oriented and close to her daughter. This provided the impetus for her to extend financial help to Tommy. The appellant was also fully aware of the effect and consequences of the Charge and had executed it freely.

(b)        The private banking services afforded by the respondent to the appellant and the communications between them were not characterised by any inequality of bargaining power. The appellant had given very precise instructions to the respondent limiting Tommy’s ability to make investments on her behalf. For instance, her moneys could not be invested in shares, mutual funds or speculative trades. The respondent scrupulously adhered to these instructions. The parties were in essence transacting at arm’s length. Furthermore, there was neither advocacy directed at persuading the appellant to execute the Charge, nor any form of “overreaching” characterised by non-disclosure of material facts. In short, there was no issue of Tommy or the respondent having taken advantage of the appellant by virtue of their relationship. Finally, the standard terms and conditions of the private banking services arrangement between the parties clearly militated against the appellant’s allegations regarding the existence of a fiduciary duty.

13        The trial judge made one further observation, which was to form the crux of the appellant’s new cause of action in this appeal. According to the trial judge (at [72] and [76] of the GD):

72         To begin with, it struck me that the [respondent] was aware of the potential conflict of interest between Tommy (who was a remunerated referral agent for the [respondent]) and the [appellant], from the commencement of its relationship with the [appellant].

…

76         … [I]t cannot be seriously disputed that it was the [respondent’s] lack of vigilance that contributed in no small measure to the present dispute, which could easily have been avoided by closer scrutiny and more comprehensive disclosure on the part of the [respondent].

[emphasis added]

The appeal

14        Before us, the appellant sought to advance a completely new line of argument. She contended that:

(a)        Tommy was an agent of the appellant and owed her fiduciary duties.

(b)        Tommy had breached these fiduciary duties.

(c)        The respondent, knowing that Tommy was the agent of the appellant, entered into a remunerated referral agency agreement with Tommy, or continued with that agreement. This created a conflict of interests and the respondent had failed to ensure that the appellant gave her informed consent to this prevailing conflict of interests.

(d)        Consequently, the law should give precedence to the appellant’s interest over those of the respondent’s.

15        Central to the appellant’s case, that these new points be accorded leave to be canvassed during the appeal, was the manner in which purported new evidence had awkwardly emerged during the trial. It would therefore be helpful to briefly recount the events as they unfolded in the proceedings below, before delving further into the issues on appeal. The following account of events was not in dispute.

Referral agreement and the appellant’s written acknowledgement

16        Less than an hour before proceedings were adjourned on the second day of the trial, during the cross-examination of the respondent’s key witness, Mr Lim Chee Kong (“Mr Lim”), the relationship manager for the Account, the appellant’s former counsel confirmed that Tommy had received remuneration for recommending the appellant to the respondent. It was revealed that this agreement to remunerate Tommy was embodied in a document (“the Referral Agreement”), which the respondent had apparently failed to disclose prior to the trial. Discovery was immediately sought, and the Referral Agreement was promptly given to the appellant later that evening.

17        Towards the end of the third day of the trial, during the continued cross-examination of Mr Lim, it was revealed that the respondent had failed to obtain the appellant’s written acknowledgement that she had consented to this referral arrangement between the respondent and Tommy. On the morning of the fourth day, the witness phase of the trial ended.

18        Counsel for the appellant candidly acknowledged in their written submissions that the Referral Agreement could quite easily have been the subject of interlocutory applications during the discovery process. Given the concatenation of circumstances and the conduct of the cross-examination, it could not be gainsaid that the original counsel must have been aware of the existence of such an agreement (see also [33] and [34] below). Nevertheless, they contended that it was the respondent’s obligation to give discovery of the Referral Agreement and, notwithstanding the failure of the appellant to ask for the Referral Agreement, the respondent could not be absolved from its original lapse to abide by its fundamental obligation to give proper discovery. Counsel for the appellant also forthrightly conceded that when the new facts emerged during the proceedings, the appellant’s counsel in the proceedings below ought to have appreciated that these facts could provide the foundation for the arguments canvassed in the appeal or, alternatively, sought an adjournment to carefully reflect on the significance and implications of the new evidence. In short, this argument could have been presented earlier.

19        Mr Siraj Omar, one of the appellant’s solicitors in the proceedings below, filed an affidavit to explain the apparent failure to take either step. He contritely claimed to have been solely focused on his preparation of Mr Lim’s cross-examination when the existence of the Referral Agreement was revealed. As he had a lot of ground to cover with Mr Lim on the third day of trial, he single-mindedly concentrated on preparing for that. This resulted in his unfortunate oversight in not appraising and following up on the Referral Agreement.

20        The appellant's counsel, Mr Davinder Singh SC, presented a forceful argument, with his customary clarity and eloquence, that grave injustice had been caused to the appellant as a result of this non-disclosure. He contended that, although there is a public interest in the finality of court proceedings, justice should prevail in the present case because if the Referral Agreement had been disclosed earlier, the appellant would have mounted a different case against the respondent with some chance of success. We now explain why, despite the late emergence of this purported new evidence in the earlier proceedings, we were not at all persuaded that any notions of fairness or justice required us to allow the appeal or any of the appellant’s applications.

New trial

21        Mr Singh urged us to order a new trial, and submitted that this was an exceptional case in which a new trial would serve the interests of justice. The appellant complained that the respondent had not informed her of the details of the remuneration of Tommy as a referral agent and that the non-disclosure prevented her from asserting a claim to void the liabilities of Tommy to the respondent which liabilities were secured by the Charge with the respondent over her assets. The appellant’s new case was that, if the respondent had been informed of the Referral Agreement, she might not even have executed the Charge to secure Tommy’s borrowings from the respondent as the referral fees payable to Tommy constituted secret bribes or commissions to him.

22        Neither O 57 r 14 of the Rules of Court (Cap 322, R 5, 2006 Rev Ed) (“ROC”) nor the Supreme Court of Judicature Act (Cap 322, 2007 Rev Ed) (“SCJA”) flesh out the grounds on which a new trial may be ordered. Section 39(2) of the SCJA merely states that this court is not bound to grant a new trial on the ground of misdirection, or on the improper admission or rejection of evidence unless “some substantial wrong or miscarriage of justice has been thereby occasioned” [emphasis added]. It is therefore axiomatic that a new trial will only be ordered when it serves the interests of justice, and the circumstances in which this is so cannot be exhaustively stated nor classified: see G P Selvam ed, Singapore Civil Procedure 2007 (Sweet & Maxwell Asia, 2007) (“Singapore Procedure Code 2007”) at para 57/14 for a non-exhaustive list of the grounds on which a new trial may be ordered. A few examples will suffice for the present. A retrial may sometimes be necessary if an adjournment has been improperly rejected by the trial court and there is good reason to believe that the absence of material evidence has prejudiced a fair outcome of the trial process. That said, a new trial will not be directed if because of the inadvertence of counsel, evidence was not adduced during the trial. Another scenario where a new trial might be ordered is when there is irrefutable proof at the appellate stage that material evidence relied upon by the trial judge is false.

23        Though the appellant did not express her written case as such, it was quite evident that she was relying on the alleged discovery of fresh evidence as the main ground on which a new trial should be ordered. This issue can be looked at from two points of view. First, whether there was good reason for the appellant not to have adduced the evidence, if it was available before or during the trial. Second, whether the respondent’s conduct in failing to give timeous discovery of material evidence had resulted in a miscarriage of justice such that it warranted a new trial. For the present purposes, it can be said that the principles upon which a new trial may be ordered on the ground that evidence, not available at the time of the trial, has since been discovered, are not unlike those applicable where the appellate court is asked to hear the new evidence itself (see Singapore Procedure Code 2007 at para 57/14/10, and also Ladd v Marshall [1954] 1 WLR 1489 (“Ladd v Marshall”) at 1491). As the principles underpinning the power to order a new trial are also conceptually similar to those applicable when an appellate court is asked to hear new evidence (see also Chong Joon Wah v Tan Lye Thiang [1991] SLR 225 at 227, [6]), we propose to consider the two applications together.

Applying the relevant principles

24        The relevant principles applicable when an appellate court is asked to hear new evidence, and also upon which a new trial may be ordered, was elucidated by Denning LJ in Ladd v Marshall at 1491:

[F]irst, it must be shown that the evidence could not have been obtained with reasonable diligence for use in the trial; secondly, the evidence must be such that, if given, it would probably have an important influence on the result of the case, although it need not be decisive; thirdly, the evidence must be such as is presumably to be believed, or in other words, it must be apparently credible, although it need not be incontrovertible.

The Ladd v Marshall test has been cited and followed by this court on a number of occasions: Cheong Kim Hock v Lin Securities (Pte) [1992] 2 SLR 349 (“Cheong Kim Hock”) at 355, [21]; Cheng-Wong Mei Ling Theresa v Oei Hong Leong [2006] 2 SLR 637 (“Theresa Cheng-Wong”) at [39]; Sim Cheng Soon v BT Engineering Pte Ltd [2006] 3 SLR 551 (“Sim Cheng Soon”) at [7]. The three conditions must be cumulatively satisfied: Su Sh-Hsyu v Wee Yue Chew [2007] 3 SLR 673 (“Su Sh-Hsyu”) at [15].

25        The respondent conceded that the third condition of the Ladd v Marshall test, ie, that the evidence must be apparently credible though it need not be incontrovertible, was satisfied here. We agreed that since the nature of the evidence took the form of a document, and there was no dispute as to its authenticity, the requirement was indeed fulfilled in this case.

26        We were also in no doubt that the second condition was satisfied here. The Referral Agreement suggested a potential conflict of interests, and the appellant’s case rested heavily on that point. In our view, since the requirement only demanded that the evidence have an important influence on the result of the case, but need not be determinative, this was a low threshold to mount, and on the face of the facts here, we accepted that the Referral Agreement would most likely have a significant influence on the result of the case.

27        In the circumstances, it was the first condition that posed a dilemma. The question we had to resolve was whether the new evidence could have been obtained for use in the trial with reasonable diligence.

28        Counsel for the appellant and the respondent cited different authorities for the interpretation of the term “reasonable diligence”. Counsel for the appellant relied on the case of Dickson v Telstra Corporation Ltd [2005] ACTCA 36, where the Supreme Court of the Australian Capital Territory held:

7          As to the question of diligence, it is to be noted that what is required is not perfection but reasonable diligence and “reasonable” means to take account of all the circumstances of the litigation in question and of the accident in question and of the events that presumably preceded it.

…

13          I also note that section 48A of the Australian Capital Territory Self-Government Act 1998 (Cth) places an overriding duty on this court to ensure that justice is done and that when the Court is looking at a concept such as reasonableness diligence the question of reasonableness demands some flexibility in balancing the need for the finality of litigation with the public interest in ensuring that justice is done.

…

19          … [T]he rule for receipt of fresh evidence is far from absolutely rigid. A discretionary judgment must be made not only as to the question of the cogency of the evidence but also the question of whether there was reasonable diligence or not. That is, diligence being reasonable in all the circumstances. The rule is also subject to the overriding consideration that the interests of justice are always of paramount importance.

[emphasis added]

29        Counsel for the respondent, meanwhile, cited Sim Cheng Soon ([24] supra), where this court referred (at [10]) to the Malaysian High Court decision of Re Lim Hong Kee David [1995] 4 MLJ 564, which had explained “diligence” in the following manner at 572:

Diligence means industry. It is a steady, earnest and meticulous pursuit towards the attainment of one’s goal. In the context of obtaining the fresh evidence that is intended to be adduced, the party seeking to make such an introduction ought to satisfy the court that he has made all reasonable cogent and positive efforts in the pursuit of obtaining the best evidence to prove his case. It is his duty to show to the court that neither indolence nor a lackadaisical attitude predominated in the preparation of his case nor that insufficient preparation at the pre-trial stage led to the party being unable to adduce the evidence he now seeks to introduce as fresh evidence. If all reasonable efforts had been exhausted and fresh evidence that is sought to be introduced could not have been thus obtained the court ought then to allow for the admission of such fresh evidence subject of course to any existing statutory provisions. Positive, tangible or clear explanation ought to be given as to why such evidence could not have been obtained with reasonable diligence for use at the trial.

The requisite standard, as noted by Andrew Phang JA (at [10] of Sim Cheng Soon), is one of reasonable diligence, “not Herculean or extraordinary effort”.

30        There can be no question here that the appellant’s solicitors had every opportunity to adduce the evidence before the end of the trial, and the appellant could not argue otherwise. Their only excuse was that they had received the Referral Agreement at a very late stage in the proceedings. Counsel for the respondent submitted that with reasonable diligence, the appellant’s solicitors would have inquired and discovered the existence of the Referral Agreement prior to the trial, or would at least have realised its significance before the end of the proceedings. We agreed with counsel for the respondent.

31        During the trial, the appellant’s solicitor had indeed cross-examined Mr Lim on whether any written acknowledgement to the referral arrangement had been obtained by the respondent. This was the precise exchange that took place[note: 1]:

Q:	Yes. And [the email] says:
	[Reads] “I note that the referring agent to the Bank …”.
	And that’s a reference to Tommy?
A:	Yes.
Q:	[Reads] “…will also be operating the account. To avoid any further misunderstandings or potential problems with the actual client. I would require a written acknowledgement from the client to the effect that she is aware of Tommy Lim’s position as both a compensated referral agent and operator of the account.”
	You see that?
A:	Yes.
…
Q:	And if you look at page 23, the email right at the top---
A:	Yes.
Q:	---you will see:
	[Reads] “PB-Credit concurs on the CR with the condition that the written acknowledgment as discussed is recvd on the next client visit.”
	Correct?
A:	Yes.
Q:	Was such a written acknowledgement obtained?
A:	Your Honour, I can’t recall but I would probably say no, it was not obtained. But we have told Mdm Susilawati that Tommy is a referral agent.
Q:	Mr Lim, can you answer the question?
A:	Okay, sorry.
Q:	So you would say that no written confirmation was received, correct?
A:	I said I can’t recall but I think it would be fair to say that no written acknowledgement was received.
Q:	Right. Did you obtain such a written confirmation from Mdm Susilawati?
A:	I don’t think so.

[emphasis added]

32        On re-examination, Mr Lim was again asked about the written acknowledgement[note: 2]:

Q:	Now, my question is, are you aware or not whether at that time, from the time the account was opened or shortly thereafter, whether Mdm Susilawati was or was not aware of Tommy being a referral agent?
A:	Mdm Susilawati knows Tommy is the referral agent, yes.
Court:	How?
Q:	How did she know that?
A:	No, I have to tell her. For the referral agent agreement, when---when a referral agent is signed up, when a referral agent brings in a new pros---brings in a new prospect, we have to make it known to the prospect that this is the referral agent. As far as the bank is concerned, we just have to let them know, that’s it. The internal arrangement between the bank and the referral agent need not be discussed with the client or the prospect. So for---for the clients that, er, Mr Tommy refer to us, I do tell them that he is the referral agent and he is paid a certain amount of money by the bank.
Q:	Now did you tell---your evidence is that---maybe let me ask you this so that we are clear. Did you tell Mdm Susilawati or did you not that Tommy was the referral agent in this case?
A:	I think I would have.
…
Court:	Did you?
A:	Yes, I think I would have. But Your Honour, I would say I think I would have, I can’t confirm, you know, it’s yes or no.
Q:	To the best of your recollection?
A:	I think I would have.

[emphasis added]

33        Mr Lim’s testimony on the stand was clear and unequivocal. He admitted that no written acknowledgement had been obtained by the respondent from the appellant, but maintained that he had in all likelihood informed the appellant of the referral arrangement between Tommy and the respondent. It is not insignificant that the appellant’s solicitor did not query, let alone dispute Mr Lim’s claim that he had informed the appellant of the referral arrangement. Instead, the cross-examination concentrated solely on whether any written acknowledgement had been obtained. This subtleness was not lost on us. It was plain to us that as the appellant’s counsel had actually referred to an internal e-mail of the respondent adverting to this very issue, the appellant was familiar with the existence of the referral arrangement, well before the trial. This e-mail was discovered to the appellant’s solicitors on 18 September 2006, some ten months before the trial. The email dated 6 January 1998 from John Hughes, a senior risk assessor, to Mr Lim stated[note: 3]:

I note that the referring agent to the Bank will also be operating the account. To avoid any future misunderstandings or potential problems with the actual client, I would require a written acknowledgement from the client to the effect that she is aware of [Tommy’s] position as both a compensated referral agent and operator of the account. If he is not being compensated for the referral of this business, then I have no problem in concurring in the CR as presented.

The fact that this email was discovered at a relatively early stage also made it plain to us that there was no premeditated attempt by the respondent to suppress information about the existence of the referral arrangement. Indeed, the respondent had also discovered two other documents, namely the “Personal Client Profile”[note: 4] and “Write up of the family of Mr Gustimego, Mdm Susilawati, Ms Farida Gustimego and Ms Zina Gustimego”[note: 5], which clearly adverted to Tommy’s role as referror. We should also express our profound disquiet with the appellant’s failure to file any personal affidavit in support of the applications, attesting to her knowledge (or lack thereof) of the referral arrangement. This would have undoubtedly helped shine some further light on this issue. Thus, in the absence of evidence to the contrary, we were minded to conclude that Mr Lim had indeed informed the appellant of the relationship between Tommy and the respondent. For good measure, we should also mention that the trial judge did not find the appellant to be a credible witness (see the GD at [11], [31], [43], [45], [46], [55], [56] and [58]). Mr Lim, on the other hand, was deemed to be a credible witness (see the GD at [50]–[52]). In the final analysis, the fact that no written acknowledgement had been obtained was, to our minds, quite irrelevant to the appellant’s real state of knowledge as to the existence of such an arrangement.

34        This actual knowledge was of real significance because the appellant’s solicitors had sought to paint the discovery of the Referral Agreement at trial as an event that had taken them by surprise. It appears to us, however, that the fact that the acknowledgement of the referral arrangement by the appellant was not embodied in the form of a document did not materially alter or redefine the pertinent circumstances. Armed with prior knowledge of the referral arrangement between Tommy and the respondent, the appellant’s solicitors, by exercising reasonable diligence, could have quite easily discovered the existence of the Referral Agreement prior to trial; alternatively, interrogatories could have been administered. In any event, the appellant’s awareness of the referral arrangement should have been more than sufficient to alert her solicitors to any possible cause of action grounded on that basis, and having chosen not to pursue that lead, it did not then lie in their mouths to argue that it was the discovery of the Referral Agreement that apprised them of the new cause of action which they belatedly wanted to advance at the appeal stage. The cross-examination of Mr Lim (see [31] above) quite clearly also conveys the unmistakeable impression that the appellant’s original counsel had indeed received prior instructions on this issue and had decided to focus his forensic efforts to undermine the respondent’s case solely on the absence of written acknowledgment rather than on the actual absence of knowledge of the referral arrangement on the part of the appellant.

35        We hasten to add that our conclusion on this point should not be construed as our condoning of the failure by the respondent to give discovery of the Referral Agreement prior to the trial. In this context, we were helpfully referred by appellant’s counsel to the Australian case of Commonwealth Bank of Australia v Quade (1991) 178 CLR 134 (“Commonwealth Bank of Australia”). In that case, the applicants had sustained financial losses by reason of unfavourable exchange rate fluctuations in relation to a Swiss franc loan which had been made to them by the Commonwealth Bank of Australia (“Commonwealth Bank”). The applicants then sued the Commonwealth Bank in the Federal Court of Australia for damages, but the proceedings were dismissed. After judgment the Commonwealth Bank made available to the applicants some documents which had been wrongly omitted from an affidavit of documents that had been filed and served in purported compliance with an order for discovery. The High Court of Australia considered the appropriate approach to be adopted by an appellate court in determining whether a new trial should be ordered when documents which should have been discovered were not disclosed by the successful party. Its incisive treatment of the innate procedural tensions in play when there is a prior breach of a discovery obligation (at 142–143) merit repetition in their entirety here:

The position is, however, different in a case such as the present where the unavailability of the evidence at the trial resulted from a significant failure by the successful party to comply with an order for the discovery of relevant documents in his possession or under his control. The application to that category of case of the general rule that a new trial should only be ordered on the ground of fresh evidence if it is “almost certain” or “reasonably clear” that the opposite result would have been produced if the evidence had been available at the first trial would, particularly where the failure was deliberate or remains unexplained, serve neither the demands of justice in the individual case nor the public interest in the administration of justice generally. In so far as the demands of justice in the individual case are concerned, it would cast upon the innocent party an unfairly onerous burden of demonstrating to virtual certainty what would have happened in the hypothetical situation which would have existed but for the other party's misconduct. In so far as the public interest in the administration of justice generally is concerned, it would be likely to ensure to the successful party the spoils of his own default and thereby encourage, rather than to penalize, failure to comply with pre-trial orders and procedural requirements.

It is neither practicable nor desirable to seek to enunciate a general rule which can be mechanically applied by an appellate court to determine whether a new trial should be ordered in a case where misconduct on the part of the successful party has had the result that relevant evidence in his possession has remained undisclosed until after the verdict. The most that can be said is that the answer to that question in such a case must depend upon the appellate court's assessment of what will best serve the interests of justice, “either particularly in relation to the parties or generally in relation to the administration of justice”. In determining whether the matter should be tried afresh, it will be necessary for the appellate court to take account of a variety of possibly competing factors, including, in addition to general considerations relating to the administration of justice, the degree of culpability of the successful party, any lack of diligence on the part of the unsuccessful party and the extent of any likelihood that the result would have been different if the order had been complied with and the non-disclosed material had been made available. While it is not necessary that the appellate court be persuaded in such a case that it is “almost certain” or “reasonably clear” that an opposite result would have been produced, the question whether the verdict should be set aside will almost inevitably be answered in the negative if it does not appear that there is at least a real possibility that that would have been so.

[emphasis added]

36        In our opinion, the court in Commonwealth Bank of Australia summed up the balancing exercise rather aptly. We agree that the factors suggested by the court, in addition to general considerations relating to the administration of justice – the degree of culpability of the successful party, any lack of diligence on the part of the unsuccessful party and the extent of any likelihood that the result would have been different if the order had been complied with and the non-disclosed material had been made available – are useful guiding principles as to whether a retrial might be necessary in any future cases relating to the failure to give proper discovery. The courts also have to remain constantly vigilant so that parties are not encouraged to think that the spoils of victory can be invariably retained regardless of the extent of their failure to observe pre-trial obligations of discovery. Litigation in Singapore is now a ‘cards-face-up-on-the-table’ process and parties must appreciate that a lack of honesty in the pre-trial proceedings can, in egregious cases, have exceptionally adverse consequences. Legal advisors have an abiding and grave responsibility, as officers of the court, to ensure that their clients are fully apprised of their serious, and often extensive, discovery obligations to provide all relevant material that may either be supportive or destructive of their case theories. Clients ought to be informed in writing at the earliest appropriate opportunity of this unwavering obligation under our adversarial system.

37        We should, however, emphasise that the present case did not involve a failure to comply with a specific order for discovery, nor were there any allegations that the respondent had deliberately suppressed the Referral Agreement or behaved in any way mala fide, and rightly so. We do not, for one moment, think that this omission was anything other than an inadvertent oversight; there was no question of any deception or impropriety having taken place in relation to the discovery process whether prior to the trial or during the trial itself. In the light of these factors, coupled with our finding that the appellant appears to have been well aware of the referral arrangement between Tommy and the respondent (see [33] and [34] above), we were not at all persuaded that the delay in discovery had resulted in any injustice whatsoever to the appellant.

38        Further, we also agreed with counsel for the respondent that with reasonable diligence, the appellant’s solicitors could have inquired and discovered the existence of the Referral Agreement prior to the trial, or at the very least ought to have realised its significance before the end of the trial. Therefore, as the first requirement of the Ladd v Marshall test had not been satisfied, we were unable to allow either the appellant’s application for leave to adduce further evidence, or the application for the court to order a new trial.

39        In the final analysis, this was certainly not one of those exceptional cases where, by reason of the discovery of fresh evidence and/or the appearance of a patent miscarriage of justice, leave to adduce further evidence or a new trial was warranted. In the normal course of events, this determination would have been sufficient to dispose of the matter. However, for the sake of completeness, we should add further that we were satisfied, in any event, that the appellant’s new point was plainly unsustainable in the face of all the adduced evidence on record and it is to this issue that we next turn.

Duty of full and frank disclosure by a fiduciary

40        One of the key questions raised in the appellant’s submissions was whether Tommy owed the appellant any fiduciary duties as an agent. The appellant submitted that it is settled law that an agent owes his principal fiduciary duties and that these duties arise whether or not the agency is gratuitous. However, we should point out that an agency relationship does not always give rise to a fiduciary duty. Moreover, the extent of any existing fiduciary duty may vary from situation to situation. In this context, Lord Upjohn’s observation in Phipps v Boardman [1967] 2 AC 46 is often relied upon (at 127):

The facts and circumstances must be carefully examined to see whether in fact a purported agent and even a confidential agent is in a fiduciary relationship to his principal. It does not necessarily follow that he is in such a position … [emphasis added]

Hence, while it is accepted that a fiduciary relationship is often presumed to exist in many types of agency relationships, this presumption can be rebutted by the relevant facts.

41        In view of our decision not to allow the appellant leave to raise and argue the new point, we did not make a finding on whether Tommy, in fact, owed the appellant any fiduciary duties and, if so, the extent of these duties. We need only mention that we consider the analysis made in Frame v Smith [1987] 2 SCR 99, by Wilson J (dissenting) in identifying the common features of a fiduciary relationship (at [60]), helpful in such an exercise:

Relationships in which fiduciary obligations have been imposed seem to possess three general characteristics:

(1)        The fiduciary has scope for the exercise of some discretion or power.

(2)        The fiduciary can unilaterally exercise that power or discretion so as to affect the beneficiary’s legal or practical interests.

(3)        The beneficiary is peculiarly vulnerable to or at the mercy of the fiduciary holding the discretion or power.

Wilson J’s summary was accepted in Hodgkinson v Simms [1994] 3 SCR 377 (at 408), and a similar formulation was adopted by the High Court of Australia in Hospital Products Ltd v United States Surgical Corporation (1984) 156 CLR 41 (“Hospital Products”), where Mason J said (at 96–97):

The accepted fiduciary relationships are sometimes referred to as relationships of trust and confidence or confidential relations … viz., trustee and beneficiary, agent and principle, solicitor and client, employee and employer, director and company, and partners. The critical feature of these relationships is that the fiduciary undertakes or agrees to act for or on behalf of or in the interests of another person in the exercise of a power or discretion which will affect the interests of that other person in a legal or practical sense. The relationship between the parties is therefore one which gives the fiduciary a special opportunity to exercise the power or discretion to the detriment of that other person who is accordingly vulnerable to abuse by the fiduciary of his position. [emphasis added]

Mason J’s analysis in Hospital Products was accepted and followed by G P Selvam J in Kumagai-Zenecon Construction Pte Ltd v Low Hua Kin [2000] 2 SLR 501 at [15], and this court also found it of assistance in Friis v Casetech Trading Pte Ltd [2000] 3 SLR 590 at [29].

42        It is trite law that if a person occupying a fiduciary position wishes to enter into a transaction which would otherwise amount to a breach of duty, he must, if he is to avoid liability, make full disclosure, to the person to whom the duty is owed, of all relevant facts known to the fiduciary: see R P Meagher, J D Heydon and M J Leeming, Meagher, Gummow and Lehane’s Equity – Doctrines and Remedies, (Butterworths LexisNexis, 4th Ed, 2002) at para 5-115.

43        The appellant alleged that Tommy had breached this duty of full and frank disclosure by failing to disclose the existence of the referral arrangement between the respondent and himself, and further that the respondent had compounded this breach by failing to ensure that the appellant gave her informed consent to the conflict of interests which the referral arrangement allegedly created. In this regard, the appellant’s submissions leaned heavily on the fact that her written acknowledgement of the referral arrangement had not been obtained.

44        As we explained earlier (at [33] and [34]), we were not impressed by this rather strained reasoning. First, Mr Lim’s evidence that he had informed the appellant of the referral arrangement was not even cursorily contested by the appellant. There was hence no doubt in our minds that the appellant had been informed of the material facts. Our view was further fortified by the failure of the appellant to file an affidavit attesting to her knowledge (or lack thereof) of the referral arrangement. Given our conclusion that the appellant was aware of the existence of the referral arrangement, the absence of a written acknowledgement was really quite irrelevant. We were not pointed to any authority which stated that informed consent could only be obtained vide a written acknowledgement and, on our part, we do not think such a prerequisite is necessary. While it is of course always preferable to have concrete documentary evidence that informed consent has indeed been obtained, it seems axiomatic to us that the factum of consent can also be proved through oral evidence and/or inferences from established facts. In the final analysis, the appellant’s very narrow focus and unflagging emphasis on the lack of a written acknowledgement was merely a red herring floating in a sea of legal flotsam.

45        We would also add that the appellant’s awareness of the referral arrangement, inter alia, meant that this was plainly not a case where the alleged fiduciary had obtained a bribe or a secret commission to favour the payor’s interest. It was clear to us that the respondent did not provide referral fees to Tommy for the purpose of inducing him to favour the respondent’s interests in preference to that of the appellant’s. There are two further points worthy of mention. First, that the appellant made an apparent paper profit of about US$1.8m during the operation of the Account. Secondly, there was no suggestion, let alone evidence, that Tommy and/or the respondent had manipulated the appellant’s Account in any manner whatsoever with the intention of deriving some benefit through the referral arrangement or otherwise. In the final analysis, we were also not convinced that the referral arrangement in this matter engendered a conflict of interests, and that Tommy had breached his fiduciary duties, if any.

New point on appeal

46        Order 57 r 13(4) of the ROC permits the admission of a new point not taken at trial:

The powers of the Court of Appeal under paragraphs (1), (2) and (3) may be exercised notwithstanding that —

…

(b)        any ground of allowing the appeal or for affirming or varying the decision of that Court is not specified in any of the Cases filed pursuant to Rule 9A or 10,

and the Court of Appeal may make any order, on such terms as the Court thinks just, to ensure the determination on the merits of the real question in controversy between the parties.

[emphasis added]

47        Arguably the most frequently cited (and authoritative) rendition of the principle governing the introduction of a new point of law on appeal is that expressed by Lord Herschell in the House of Lords decision of The Owners of the Ship “Tasmania” and the Owners of the Freight v Smith and others, The Owners of the Ship “City of Corinth”, The Tasmania (1890) 15 LR App Cas 223 (at 225):

My Lords, I think that a point such as this, not taken at the trial, and presented for the first time in the Court of Appeal, ought to be most jealously scrutinised. The conduct of a cause at the trial is governed by, and the questions asked of the witnesses are directed to, the points then suggested. And it is obvious that no care is exercised in the elucidation of facts not material to them.

It appears to me that under these circumstances a Court of Appeal ought only to decide in favour of an appellant on a ground there put forward for the first time, if it is satisfied beyond doubt, first, that it has before it all the facts bearing upon the new contention, as completely as would have been the case if the controversy had arisen at the trial; and next that no satisfactory explanation could have been offered by those whose conduct is impugned if an opportunity for explanation had been afforded them when in the witness box.

[emphasis added]

Over the years, this principle has been cited and applied in Singapore on a number of occasions: Attorney-General for the Straits Settlement v Pang Ah Yew [1934] MLJ 184; Cheong Kim Hock ([24] supra); MCST No 473 v De Beers Jewellery Pte Ltd [2002] 2 SLR 1; Riduan bin Yusof v Khng Thian Huat (No 2) [2005] 4 SLR 234; Panwah Steel Pte Ltd v Koh Brothers Building & Civil Engineering Contractor (Pte) Ltd [2006] 4 SLR 571 (“Panwah Steel”).

48        In our view, however, Lord Birkenhead LC’s carefully measured caveat in North Staffordshire Railway Company v Edge [1920] AC 254 at 263–264, also merits very close attention and adherence:

[T]here are very few cases of which it can be confidently stated that a failure to raise a relevant contention at the appropriate stage will not prejudice the other litigant. It is, for instance, argued in this case that a proper order for costs would compensate the respondent for the expenses of the litigation unnecessarily incurred, on the hypothesis that, being informed at first instance of the true contention, he might have been advised to abandon his claim. I do not think it necessary to point out that this suggestion may possibly in another case require qualification or consideration, having regard to the notorious incompleteness of the indemnity furnished by our present system of costs, because in the present case counsel for the respondent satisfied me that there was reasonable ground for supposing that he might have been able to strengthen his case upon this branch of it by calling parol evidence.

But I desire to draw attention to a consideration which in my view is both more general and more important. The appellate system in this country is conducted in relation to certain well-known principles and by familiar methods. The issues of fact and law are orally presented by counsel. In the course of the argument it is the invariable practice of appellate tribunals to require that the judgments of the judges in the Courts below shall be read. The efficiency and the authority of a Court of Appeal, and especially of a final Court of Appeal, are increased and strengthened by the opinions of learned judges who have considered these matters below. To acquiesce in such an attempt as the appellants have made in this case is in effect to undertake decisions which may be of the highest importance without having received any assistance at all from the judges in the Courts below. Decisions of this House have laid it down that in very exceptional cases, and in spite of the considerations above referred to, new matters may be considered by your Lordships: see the judgment of Lord Halsbury in Sutherland v. Thomson and the judgment of Lord Watson in Connecticut Fire Insurance Co. v. Kavanagh. I have carefully examined the cases upon the subject which have been decided in this House, and my examination of them has led me more and more to the conclusion that such attempts must be vigilantly examined and seldom indulged.

[emphasis added]

Part of the preceding passage has been cited with approval by this court in Feoso (Singapore) Pte Ltd v Faith Maritime Company Limited [2003] 3 SLR 556 at [32].

49        The appellant vigorously submitted that it would be in the interests of justice for the new point to be heard. However, this argument can only be taken so far. As McHugh JA aptly noted in Holcombe v Coulton (1988) 17 NSWLR 71 at 77–78:

Certainly, I cannot accept the notion that the interests of justice require that cases should be heard and re-heard until every conceivable factual pattern or every conceivable legal principle of relevance that finally occurs to the parties have been litigated. The cost and strain of litigation and the limits of curial resources have to be weighed against the demand of the appellant for justice according to the set of rules which represent the "law" which should have governed the case.

Under the adversary system of justice, the function of the trial court is to determine disputes in respect of issues formulated by the parties, and the function of an appellate court is to correct any error of the trial court in making its determination. Moreover, the policy of the law is that, when a matter becomes the subject of litigation and adjudication, both parties are forever precluded from litigating any issue which might have been brought forward as part of the matter in dispute: Port of Melbourne Authority v Anshun Pty Ltd (1981) 147 CLR 589.

To allow a party to raise in an appellate court a matter which was not litigated in the trial court not only undermines the respective functions of the trial and appellate courts and the policy of law but perhaps more importantly it deprives the appellate court of the benefit of the views of the trial court.

[emphasis added]

50        It is axiomatic that the “interests of justice” must be afforded a wider scope than to just merely encompass one party’s (ie, the appellant’s) expectation of “justice”. In Geelong Building Society v Encel [1996] 1 VR 594, Tadgell J of the Supreme Court of Victoria made the following highly pertinent observation at 605:

An abstract notion that an appellant should receive justice, though undoubtedly one of them, will not be accorded undue weight at the expense of other considerations which are entitled to be accorded their own due weight, including a respondent's expectation also to receive justice.

We agree. It has to be emphasised that finality in litigation is usually as much in the interests of justice as a party’s interest in succeeding in its claim. In a case where neither life nor limb is involved, “justice” must be understood (and applied) in the sense that all legitimate expectations are satisfied; the legitimate expectations of the respondent here are that the case should be resolved on the basis of arguments presented and evidence adduced in the court below, and not on newly minted arguments that require further testimony and forensic examination. Plainly there will be a grave and altogether understandable sense of palpable injustice felt by a respondent if a verdict is reversed on grounds that ought to have been raised earlier and could perhaps have been addressed by contrary evidence at the trial.

51        The unequivocal weight of case law and policy firmly stood against allowing the new point to be raised and argued in this appeal. While the respondent ought not to be lightly excused for its tardiness in discovery, which unhappily resulted in the late disclosure of the Referral Agreement only at the trial, all said and done, it is evident that the appellant did have more than ample time, both before and during the course of the trial, to raise the new point. While one may also have some modicum of sympathy for the appellant’s counsel, arising from the respondent’s failure to produce the Referral Agreement (assuming arguendo that they were unaware of the existence of this document), the courts, on the other hand, have to be vigilant against excessively indulging counsel, who have an abiding professional responsibility to examine and to cover all necessary aspects of a matter prior to the commencement of trial. Suffice it to say that the appellant’s situation plainly did not qualify as one of the “very exceptional cases” described by Lord Diplock in the Privy Council case of Rengasamy Pillai v Comptroller of Income Tax [1970] 1 MLJ 233, and applied in Hoong Chin Wah v Cheah Kum Swee [1967] 1 MLJ 163. The courts will not and should not, for obvious reasons, allow the main arena for the resolution of evidential disputes to be moved from the trial court to the appellate court; to do so would fundamentally alter the limited role of appellate review.

52        Even if the appellant’s circumstances could be considered exceptional, such as to justify allowing the new point to be raised, we would still have to be satisfied beyond doubt that we had all the material facts bearing upon the new contention before us, and that further, no satisfactory explanation could be offered by the respondent if its witnesses had been accorded the opportunity to clarify matters in the witness box (see [47] above). This, unfortunately for the appellant, was far from being the case here.

53        Whether it can be said that a court has all the necessary facts before it must of course, in every case, ultimately tilt on the precise point being raised. The conundrum here is that the inquiry at this preliminary stage must not focus on the merits of the substantive case, ie, the court must ordinarily decide whether to allow the new point to be raised without, at the same time, deciding whether the claim will ultimately succeed. It is therefore not surprising that it is easiest to allow a new point to be raised when the new point relates merely to the construction of a document or resolution of a matter that requires no adduction of fresh evidence or review of the factual matrix (see Panwah Steel at [17]). In scenarios such as these, the appellate court can just as easily resolve the controversy without further evidence and/or the need to assess the credibility of the relevant witnesses. Unfortunately for the appellant, this was again not the type of case before us.

54        Here, the appellant had just belatedly fashioned an entirely new cause of action on appeal, alleging a breach of fiduciary duties owed by Tommy to the appellant. Several inter-connected facts would necessarily be relevant to the evaluation of this issue. In particular, whether the appellant had given her informed consent to the purported conflict of interests would have been a key question; in this context, it was essential that the respondent be accorded the opportunity to cross-examine the appellant on this matter. There had been no proper cross-examination on this point at trial simply because this new cause of action had not been pleaded. Allowing the new point to be raised and argued now without giving the respondent an opportunity to adequately cross-examine the appellant would, without doubt, severely, and perhaps irretrievably, prejudice the respondent. We therefore had little hesitation in refusing the appellant leave to raise and argue this new point before us.

Application for leave to amend pleadings

55        The appellant’s final application was for leave to amend the pleadings. Counsel for the appellant submitted that this was an exceptional case where the court should exercise its discretion to allow an amendment of the pleadings on appeal so that the new point could be argued. In light of our determination of the procedural issue on leave to raise a new point, this matter became moot. Nevertheless, out of deference to counsel’s industry in addressing this issue, we shall also explicate our views on it. In addition, it may be helpful to the wider legal community if we reiterate our views on late amendments to pleadings.

56        It is settled law that this court has the power to grant leave to a party to amend his pleadings at any time under s 37(2) of the SCJA and O 57 r 13(1) of the ROC: Asia Business Forum Pte Ltd v Long Ai Sin [2004] 2 SLR 173 (“Asia Business Forum”) at [9].

57        In Ketteman v Hansel Properties Ltd [1987] AC 189 (“Ketteman”), Lord Brandon of Oakbrook, after reviewing the authorities, summarised the applicable principles (at 212) as follows:

First, all such amendments should be made as are necessary to enable the real questions in controversy between the parties to be decided. Secondly, amendments should not be refused solely because they have been made necessary by the honest fault or mistake of the party applying for leave to make them: it is not the function of the court to punish parties for mistakes which they have made in the conduct of their cases by deciding otherwise than in accordance with their rights. Thirdly, however blameworthy (short of bad faith) may have been a party’s failure to plead the subject matter of a proposed amendment earlier, and however late the application for leave to make such amendment may have been, the application should, in general, be allowed, provided that allowing it will not prejudice the other party. Fourthly, there is no injustice to the other party if he can be compensated by appropriate orders as to costs.

Lord Griffiths then elaborated (at 220):

Whether an amendment should be granted is a matter for the discretion of the trial judge and he should be guided in the exercise of the discretion by his assessment of where justice lies. Many and diverse factors will bear upon the exercise of this discretion. I do not think it possible to enumerate them all or wise to attempt to do so. But justice cannot always be measured in terms of money and in my view a judge is entitled to weigh in the balance the strain the litigation imposes on litigants, particularly if they are personal litigants rather than business corporations, the anxieties occasioned by facing new issues, the raising of false hopes, and the legitimate expectation that the trial will determine the issues one way or the other. Furthermore to allow an amendment before a trial begins isquite different from allowing it at the end of the trial to give an apparentlyunsuccessful defendant an opportunity to renew the fight on an entirelydifferent defence. 

Another factor that a judge must weigh in the balance is the pressure onthe courts caused by the great increase in litigation and the consequentnecessity that, in the interests of the whole community, legal businessshould be conducted efficiently. We can no longer afford to show the sameindulgence towards the negligent conduct of litigation as was perhapspossible in a more leisured age.

[emphasis added]

58        In Asia Business Forum, this court cited Ketteman with approval, but added (at [12]):

Generally, it is true that an amendment may be allowed at any stage of the proceedings, including post-judgment. Clearly, the later an application is made, the stronger would be the grounds required to justify it. As it is a matter of discretion, no hard and fast rules may be laid down. But at the end of the day, the court must balance it against the justice of the case. [emphasis added]

The court went on to explain (at [18]):

Without intending to be exhaustive, and in the context of the present case, it was our view that two key factors should be borne in mind. Firstly, the amendments should not cause any prejudice to the other party. Secondly, the amendments should not give rise to a situation whereby the applicant was effectively asking for a second bite at the cherry.

59        It is therefore plain that the considerations which permeate the exercise of discretion to grant leave to amend pleadings are very similar to those underpinning the discretion to allow a new point to be raised and argued on appeal. In Chwee Kin Keong v Digilandmall.com Pte Ltd [2004] 2 SLR 594 (“Chwee Kin Keong”), the High Court summed up the procedural equation thus at [85]:

Having stated the general rule, it is imperative that the rationale underlying this approach be understood. Rules of court which are meant to facilitate the conduct of proceedings invariably encapsulate concepts of procedural fairplay. They are not mechanical rules to be applied in a vacuum, devoid of a contextual setting. Nor should parties regard pleadings as assuming an amoeba-like nature, susceptible to constant reshaping. Rules and case law pertaining to amendments are premised upon achieving even-handedness in the context of an adversarial system by:

(a)        ensuring that the parties apprise each other and the court of the essential facts that they intend to rely on in addressing the issues in controversy or dispute;

(b)        requiring that an amendment should be attended to in the usual course of events, at an early stage of the proceedings, to ensure that no surprise or prejudice is inflicted on or caused to opposing parties;

(c)         requiring careful consideration whether any amendments sought ata late stage of the proceedings will cause any prejudice to the opposingparty. Prejudice is to be viewed broadly to encompass any injustice andembraces both procedural and substantive notions;

(d)         recognising that while a costs award against the party seeking lateamendments can frequently alleviate any inconvenience caused, this maynot always be appropriate;

(e)         taking into account policy considerations that require finality inproceedings and proper time management of the courts’ resources andscheduling. From time to time there will be cases where this is anoverriding consideration.

In short, where does the justice reside? There is constant tension in our legal system to accommodate the Janus-like considerations of fairness and finality.

[emphasis added]

60        The appellant’s application had to fail. As alluded to earlier, an amendment at this stage would severely prejudice the respondent for which a costs order could not sufficiently compensate. In any event, a point often emphasised in the authorities considered is that justice cannot always be measured in terms of money.

61        In Chwee Kin Keong, the High Court observed (at [86]):

As a matter of fairness, allowing amendments at a late stage should usually go hand in hand with granting leave to the other party to adduce further evidence, if necessary.

Thus, if the appellant was to be allowed leave to amend its pleadings and assert a new cause of action, the rules of fairness and procedure would dictate that the respondent also be granted leave to adduce further evidence to defend against the new claim. In short, a re-trial would be necessary.

62        Moreover, where as in this case, the application to amend pleadings comes so late in the proceedings, the court must necessarily be slow to grant leave. This was also made clear in Chwee Kin Keong, where the court held (at [84]):

It is axiomatic that a court will generally be cautious if not reluctant to effect any amendments once the hearing has commenced; even more so once the evidential phase of the proceedings has been completed. Lord Griffiths in Ketteman v Hansel Properties Ltd [1987] AC 189 at 220 stated:

[T]o allow an amendment before a trial begins is quite different from allowing it at the end of the trial to give an apparently unsuccessful defendant an opportunity to renew the fight on an entirely different defence.

In Asia Business Forum, Chao Hick Tin JA similarly noted at [16]–[17]:

It seemed to us that with the proposed amendments, while the nature of the claim remained substantially the same, the premises upon which the claim was based would be altered. If the position were truly as ABF had asserted, there would have been no need to amend. The fact that the amendments were needed necessarily suggested that there were some material differences and, in our view, there were. Evidence was adduced and the parties cross-examined at the trial on the basis that what was in the training manual were trade secrets and what was in the database, confidential information. The trial judge had reached his decision on that basis. If we had allowed the amendments, what would be before the Court of Appeal would not be the same as what was before the trial judge. The appeal court would not be deciding whether the trial judge was correct because the premises would have changed. Unlike the trial judge, we would have to decide if the contents of the training manual were confidential information and the database, trade secrets. Thus, the grounds of decision of the judge would become immaterial and this court would be deprived of the benefits of the opinion of the court below. We would have had to examine the claim almost from scratch.

We have mentioned above that the court has the power to grant an amendment even after final judgment and pending appeal. Clearly, leave toamend in such circumstances would be rare. Generally, very good andcompelling grounds must be shown, unless the proposed amendments aretechnical and of no real consequence.

[emphasis added]

We reiterate that the present case was certainly not one in which the proposed amendments could be said to be technical or where the profound consequential evidential implications could be lightly put aside.

63        Finally, we would endorse the concluding observations of Chao JA in Asia Business Forum (at [19]), which we found especially apposite here:

We would hasten to add that we fully appreciated the position of the present solicitors for ABF. They stepped into the picture after judgment had been delivered by the court below. They studied the case and felt that the amendments were necessary at the appeal stage. Perhaps, the case of LansingLinde Ltd had a bearing on the proposed amendments. But, as explained before, this presented the respondents with a somewhat different “battle”. Therein was the problem. It was not a problem which could be taken care ofjust by way of an order for costs. Solicitors do differ as to how a case shouldbe presented. But the other party should not be required to fight a different“battle” at the appeal stage. [emphasis added]

In a case like the present one, where the appellant’s solicitors could have, with reasonable diligence, addressed all these evidential and pleading shortcomings before or during the trial phase, allowing such amendments, that would necessitate a retrial, would be plainly inappropriate and unjust. We were therefore constrained to dismiss the appellant’s application for leave to amend the pleadings, not least because it had been made at such a late stage in the proceedings, long after the decision of the trial judge had been delivered.

Banking secrecy

64        In her judgment, the trial judge made two particular observations on which we find necessary to comment.

Relationship between section 47 of the Banking Act and the common law duty of confidentiality

65        First, the trial judge said (at [83]–[86] of the GD):

83         Section 47 of the [Banking] Act provides for a general prohibition against disclosure of customer information followed by a limited number of exceptions embodied in the Third Schedule under which disclosure is permitted. The relevant sub-sections provide that:

(1)        Customer information shall not, in any way, be disclosed by a bank in Singapore or any of its officers to any other person except as expressly provided in this Act.

(2)        A bank in Singapore or any of its officers may, for such purpose as may be specified in the first column of the Third Schedule, disclose customer information to such persons or class of persons as may be specified in the second column of that Schedule, and in compliance with such conditions as may be specified in the third column of that Schedule.

84         Unsurprisingly, a brief perusal of these statutory exceptions reveals a close parallel between the statutory duty imposed by the [Banking] Act and a banker’s common law duty of confidentiality.

85         In the seminal case of Tournier v National Provincial Bank and Union Bank of England [1924] 1 KB 461, the English Court of Appeal held that a banker came under an implied duty to keep the affairs of a customer confidential, subject to four general exceptions (at 473) under which disclosure could be made by the bank. These have been classified into four categories: (a) where disclosure is under compulsion by law; (b) where there is a duty to the public to disclose; (c) where the interests of the bank require disclosure; (d) where the disclosure is made by the express or implied consent of the customer.

86         We are, for our purposes, primarily interested in the fourth exception, which interestingly finds no exact parallel under our (stricter) statutory regime. Under s 47, nothing less than a written consent by the customer or his personal representatives would suffice to lift the prohibition against disclosure (see the [Banking] Act, Third Schedule, Pt I, cl 1).

[emphasis in original]

66        We note that the trial judge’s overview may give the impression that the case of Tournier v National Provincial and Union Bank of England [1924] 1 KB 461 (“Tournier”) remains applicable in Singapore, and that the four general exceptions to a banker’s implied duty to keep the affairs of a customer confidential stated therein run parallel to the statutory exceptions provided for under s 47 of our Banking Act (Cap 19, 2003 Rev Ed) (“Banking Act”). We therefore feel compelled to address this issue to ensure that the position is free from doubt.

67        Section 47(1) of the Banking Act states categorically that:

Customer information shall not, in any way, be disclosed by a bank in Singapore or any of its officers to any other person except as expressly provided in this Act. [emphasis added]

In light of the plain wording of s 47, our current statutory regime on banking secrecy leaves no room for the four general common law exceptions expounded in Tournier to co-exist. They have been embraced within the framework of s 47 of the Banking Act, which is now the exclusive regime governing banking secrecy in Singapore. Section 47 makes it plain that no customer information shall be disclosed by a bank in Singapore or any of its officers except as expressly provided for in the Banking Act. A breach of any of the prescribed statutory obligation amounts to a criminal offence. The Third Schedule to the Banking Act sets out, in illuminating detail, the circumstances, conditions, and details of permissible disclosure. It is axiomatic that, in terms of details and scope, this is a more comprehensive regime than that articulated in Tournier. There is simply no room, in Singapore, for the less sophisticated and more general common law rules articulated in Tournier to have any further relevance save for the perspective of historical evolution and context it provides.

Suggestion by trial judge that changes be made to the current statutory regime on banking secrecy

68        The trial judge also opined (at [96] of the GD):

I have at [82], [90], [91] and [95] above referred to standard textbooks commonly referred to by banking law practitioners but not relied on by the parties in their submissions. At [88], [89], [92] and [93] above I have referred to cases cited by Paget as well as to useful articles in [93] and [95] above to highlight the continuing difficulties faced by banking institutions in an area fraught with uncertainty, akin to the banks treading through minefields. It is hoped that in time to come, there would be amendments to the [Banking] Act or new legislation altogether to strike an appropriate and fair balance between the interests of confidentiality for banks and the protection of guarantors of banks’ customers. At the very least, it would provide a modicum of guidance to banks who find themselves in a similar position to the defendant, to avoid being the recipient of writs from disgruntled guarantors. [emphasis added]

69        We wish to clarify that we do not currently see, on our part, any problems with our current statutory banking regime in relation to the particular issues identified by the trial judge. These are all matters that banks can regulate quite easily by way of contractual arrangements with their customers and/or guarantors. Indeed, as a matter of practice, most prudent banks currently do contractually regulate the parties’ rights in this area of recurring conflicting tensions. There is, with respect, no peculiar lacuna in the law which presently necessitates the immediate attention of and intervention by the legislature.

70        Finally, for completeness, we think that it is also useful to draw attention to the helpful and authoritative summary of the current legal position in relation to the proper management of conflicts of interests usually subsisting in the banker/customer relationship as stated in E P Ellinger, Eva Lomnicka and Richard Hooley, Ellinger’s Modern Banking Law (Oxford University Press, 4th Ed, 2006) at pp 134–135:

In principle, it always remains open for a bank to avoid liability for a conflict of interest by making full disclosure to the customer that there is an actual or potential conflict and obtaining the customer’s fully informed consent to the bank to continuing to act. General advanced disclosure may be enough for some purposes, but express disclosure of specific circumstances is more reliable. What is required for the customer to give his or her fully informed consent is a question of fact in all the circumstances. Nevertheless, the bank’s duty of confidentiality to one customer may prevent another customer giving fully informed consent. It may not always be practical to seek the customer’s consent to disclosure of information. Moreover, advanced consent to disclosure of information may not be specific enough to cover every situation. If the bank cannot get a customer’s fully informed consent to the bank continuing to act, it must desist from acting.

Alternatively, the bank may seek to exclude or modify the fiduciary obligations that it would otherwise owe to its customer through the terms of their underlying contract. The use of contractual techniques to exclude or modify the scope of fiduciary duties has been endorsed by both the Privy Council and House of Lords.  For example, a multifunctional bank, providing a range of financial services to its customers, may seek to rely on a contractual term to permit it to use Chinese walls to restrict the flow of information within the bank and to restrict the information to be provided to particular customers. There are, however, limits to this approach. First, contractual terms excluding or modifying fiduciary obligations run the risk of being held ‘unreasonable’ under the Unfair Contract Terms Act 1977 and/or ‘unfair’ under the Unfair Terms in Customer Contracts Regulations 1999. The chances of the courts upholding such terms under the relevant legislation seem to be greater when they are used against a financially sophisticated customer than against a relatively unsophisticated consumer. Secondly, such terms will be construed contra proferentem. Thirdly, where there is a pre-existing fiduciary relationship between the bank and its customer, the bank would not be able to exclude or modify its fiduciary obligations without the fully informed consent of the customer. Fourthly, it is unlikely that contractual terms could give the bank an unrestricted general right to act for opposing parties in the same transaction where the customers’ respective interests conflict or to exclude liability where the deliberate suppression of information has deceived the customer.

[emphasis in original, emphasis added in bold italics]

Banks and their advisors should take careful note of these sage observations, particularly in relation to the potential limits of the actual protection that can be conferred by overreaching contractual provisions and/or exclusion clauses.

Conclusion

71        We accordingly dismissed the appellant’s three applications before us as well as the appeal against the decision of the court below. We awarded costs of the appeal and the three applications to the respondent on a contractual basis, with the usual consequential orders.