1 This appeal concerns another matter which arose in the aftermath of the collapse of Hin Leong Trading (Pte) Ltd (“HLT”) amidst widespread allegations of fraud by its controllers, comprising Mr Lim Oon Kuin (“OK Lim”) and his family members (collectively, “Lim Family”), in their conduct of HLT’s affairs. The fraudulent conduct included, among other things, material misstatements in HLT’s financial statements that had the effect of concealing HLT’s true financial position, in particular its deepening insolvency, for some years before the truth came to light in April 2020.

2 It is to be expected when a company suffers a collapse in such circumstances that those left to pick up the pieces would begin to investigate how and why things came to be, as well as to identify those persons who ought to bear responsibility for the catastrophe. In many cases, the focus will turn to the acts and omissions of the auditor of the company who, while not a primary wrongdoer, is often sought to be visited with legal responsibility to make good the losses caused by the primary wrongdoers, on the basis that it could and should have discovered the wrongdoing and blown the whistle. This brings to the fore questions as to the scope of an auditor’s duty and the proper extent of its liability.

3 These issues arise before us in the context of proceedings that have been brought by HLT, acting through its liquidators, against HLT’s former auditor for many years, Deloitte & Touche LLP (“Deloitte”). In essence, HLT accuses Deloitte of professional negligence in failing to detect and report the irregularities in HLT’s financial statements and affairs for a period of around six years prior to HLT’s entry into insolvency proceedings in April 2020. In the court below, Deloitte applied to strike out HLT’s claims on a variety of grounds but was largely unsuccessful. Deloitte now appeals to this court.

4 For the purposes of the appeal, the areas of dispute have been narrowed to two of the grounds for striking out HLT’s claims which Deloitte had raised below. The first concerns the significance of the fact that the Lim Family, as HLT’s sole directors and shareholders, had full knowledge of the true state of affairs. According to Deloitte, this is a complete answer to HLT’s claims against it because “no loss could have resulted from any failure by Deloitte to inform HLT of matters that it already knew of”, such that all the losses suffered by HLT are irrecoverable from Deloitte “for want of legal causation and/or for falling outside Deloitte’s scope of duty”. HLT disputes this. It submits that, owing to its insolvency at the time of the audits, Deloitte’s duty was not limited to providing the Lim Family with accurate information, but also extended to “hav[ing] regard to the interests of [HLT’s] creditors”. This has been referred to by the parties as the “Creditor Duty” but, as will be explained below, the usage of this convenient shorthand appears to have obscured from view the true issues that should be asked and answered.

5 The second point in this appeal is whether, assuming Deloitte as having been negligent, Deloitte ought to be liable for the losses that HLT incurred from its trading activities in the period between November 2015 and mid-April 2020 (“Trading Losses”). This head of loss is by far the most significant of the losses that HLT seeks to recover from Deloitte as it has been quantified by HLT in the sum of US$2.6 billion. HLT alleges that, if not for Deloitte’s negligence, it would have been put into liquidation earlier and would not have continued to incur further losses that deepened its insolvency. Deloitte submits that a steady line of authority establishes that negligent auditors cannot be made liable for such losses as they are either not “legally caused” by the auditors’ negligence or fall outside the auditors’ “scope of duty”. Although Deloitte has identified authorities which support the conclusion it urges on the court, its case does not go further than bare analogy in the result and it has not identified a principled justification for its desired outcome.

6 Having considered the parties’ submissions, we allow the appeal in part and hold that HLT’s claim for its Trading Losses should be struck out. However, it would be apparent from the overview of the issues and the parties’ cases above that we consider that the way in which the parties have approached the matter has not been a model of clarity. In particular, the framing of the issues, and in turn, the parties’ approaches to answering them, have not been entirely satisfactory as they appear to have been founded on some fundamental misconceptions. To unravel some of the threads of confusion, it is necessary to return to the basic principles that undergird the law of negligence and we take the opportunity to do so in this judgment.

7 It is common ground that, for this appeal, the parties and the court should proceed on the basis that all the facts pleaded by HLT in its Statement of Claim (Amendment No. 2) are true. A short summary of HLT’s pleaded case so far as it is necessary to understand the parties’ arguments and the issues before us follows.

8 Deloitte was engaged as HLT’s external auditor from at least 2003 until it resigned on 17 September 2020. In performance of its duties, Deloitte audited HLT’s financial statements and issued unqualified opinions for each of the financial years ended (“FY”) 31 October 2014 through 31 October 2019.

9 At all material times, control of HLT was concentrated in the hands of the Lim Family. Sometime in April 2020, Mr OK Lim affirmed an affidavit admitting to an array of irregularities in HLT’s affairs, including material misstatements in HLT’s financial statements that had concealed the fact that HLT had made staggering losses over the years.

10 Shortly after, HLT was placed under judicial management and later into compulsory liquidation. The liquidators’ ensuing investigation revealed fraud and irregularities in HLT’s affairs since 2010, none of which were reflected in HLT’s audited financial statements or highlighted by Deloitte in the audit opinions it had issued. In particular, HLT’s audited financial statements from FY2014 to FY2019 contained material misstatements, such as the inclusion of fictitious profits and overstatement of accounts receivable and inventory, which meant that the value of HLT’s assets was inflated and that the audited financial statements did not present a true and fair view of HLT’s financial position. By the liquidators’ estimate, HLT was insolvent from as early as 2012. However, as this was never reflected in HLT’s audited financial statements, a vastly misleading picture of HLT’s financial health was presented to third parties, which enabled HLT to continue trading and incurring further losses until it entered insolvency proceedings in 2020.

11 In HC/S 237/2021 (“Suit 237”), HLT claims damages from Deloitte for the losses it has suffered. We summarise the case that HLT has advanced in this section. We emphasise that, at this juncture, we confine ourselves to setting out HLT’s claim as it has pleaded it and described it in its submissions. We caveat ourselves in this way because, as we will explain, the way in which HLT’s (and Deloitte’s) case has been presented is rather problematic. Indeed, the difficulties reach down to even the type and number of causes of action which HLT has advanced. This has had implications on the approach that we have taken in the appeal. The summary in this section is thus mainly for the purpose of contextualising the discussion to follow, although we will signpost issues that we will come back to later.

12 According to HLT, it has currently advanced two different causes of action against Deloitte:

(a) First, professional negligence when carrying out the audits of HLT’s financial statements for FY2014 to FY2019, which it has termed the “Negligence Claim”.

(b) Second, a breach of a duty to report to government authorities and/or third parties any fraud and/or irregularities that were likely to result in material loss to HLT (“Reporting Duty”), which it has termed the “Reporting Duty Claim”.

It is useful to note that the Negligence Claim is based on a duty of care arising in contract, as an implied term of the engagement letters between HLT and Deloitte, as well as in tort. Given this, it may be more accurate to say that HLT has advanced three causes of action, as the Negligence Claim consists of both a contractual and tortious variant. We highlight this here because the fact that HLT has advanced the Negligence Claim in both contract and tort is a point of considerable significance which has been overlooked by the parties. We will elaborate on this below.

13 A number of sub-duties which collectively make up Deloitte’s duty of care have also been pleaded. These include: (a) a duty to exercise professional judgment and maintain professional scepticism throughout the audit; (b) a duty to identify risks of material misstatement in HLT’s financial statements; (c) a duty to act in accordance with the relevant auditing standards set out in, among other sources, the Singapore Standards on Auditing issued by the Institute of Singapore Chartered Accountants and the companies legislation; and (d) a duty to report fraud or irregularities that are likely to result in material loss to HLT. At this juncture, we highlight the last of these sub-duties as it appears to be HLT’s pleading of what it has called the “Reporting Duty”, which is the basis of its Reporting Duty Claim that has been presented in the course of arguments as a distinct cause of action from its Negligence Claim. It is odd, to say the least, that the Reporting Duty Claim is advanced as a separate cause of action given that the Reporting Duty is pleaded as a part of Deloitte’s duty of care which is the basis of the Negligence Claim.

14 In Suit 237, HLT charges Deloitte with failing to exercise reasonable skill and care in the conduct of the audit. As a result, Deloitte failed to detect material misstatements in HLT’s financial statements which concealed HLT’s insolvency over many years. This forms the nub of HLT’s case in respect of the Negligence Claim. The Reporting Duty Claim, on the other hand, focuses on Deloitte’s failure to report fraud or irregularities in HLT’s affairs which HLT says Deloitte ought to have detected if it had performed its duties as an auditor with reasonable skill and care. It is apparent from this that the Reporting Duty Claim as pleaded by HLT is intertwined with the Negligence Claim. This is because the Reporting Duty Claim is not confined to Deloitte’s failure to report matters that were in its knowledge but extends to matters that would or should have come to its knowledge if it had not been negligent.

15 In terms of remedies, HLT seeks to recover damages from Deloitte in respect of three heads of loss:

(a) First, the Trading Losses incurred by HLT in the period of November 2015 to mid-April 2020, totalling US$2.6 billion.

(b) Second, dividends that were wrongfully declared by the Lim Family, being the only shareholders of HLT, for FY2017 and FY2018, totalling US$90m.

(c) Third, audit engagement fees paid to Deloitte from FY2015 to FY2019, totalling $612,000, which HLT claims it would not have paid to Deloitte if it had ceased trading earlier, which would have happened if Deloitte had discharged its duties properly.

16 Although HLT’s case currently centres around the Negligence Claim and the Reporting Duty Claim, this was not how its case was always pleaded but was the result of developments in the proceedings below. Moreover, while this appeal centres around the alleged Creditor Duty and the recoverability of HLT’s Trading Losses specifically, this framing of the issues is the result of the narrowing of broader areas of dispute between the parties in the courts below. It is thus useful to situate these issues in the wider context of the proceedings below as it assists in understanding the relevance of the issues and the possible thinking behind the parties’ respective positions.

17 In its initial Statement of Claim, HLT had only pleaded the Negligence Claim. No reference was made to the Reporting Duty Claim at that juncture. The Reporting Duty Claim first surfaced in HC/SUM 68/2024 (“SUM 68”), which was an application by HLT for leave to amend its Statement of Claim. Apart from those pleadings relating to the Reporting Duty Claim, HLT also sought to introduce pleadings relating to another cause of action, namely a claim against Deloitte for breach of a statutory duty to report the Lim Family’s breaches of the Companies Act (Cap 50, 2006 Rev Ed) (“CA”) to the Registrar of Companies pursuant to s 207 of the CA (“Breach of Statutory Duty Claim”).

18 Subsequently, Deloitte applied in HC/SUM 474/2024 (“SUM 474”) seeking to strike out HLT’s pleadings in Suit 237 in their entirety. SUM 68 was heard together with SUM 474 by an assistant registrar (“AR”). However, as Deloitte resisted SUM 68 on the ground that the amendments relating to the Reporting Duty Claim and the Breach of Statutory Duty Claim were themselves liable to be struck out, the matter was essentially heard by the AR as a single striking-out application on HLT’s pleadings, inclusive of the amendments sought to be introduced in SUM 68.

19 Deloitte’s primary case was that the Negligence Claim should be struck out as none of the losses which HLT sought to recover were caused by Deloitte’s alleged negligence or were within the scope of Deloitte’s duty of care. The crux of this argument was that, since the Lim Family had been aware of the true state of affairs, they, and by extension HLT itself, had not relied on the audited financial statements which contained the misstatements that Deloitte failed to detect. In the circumstances, any negligence on Deloitte’s part did not cause HLT’s losses. Moreover, as Deloitte’s duty as an auditor was to report to HLT’s shareholders (the Lim Family) on whether HLT’s accounts gave a true and fair view of its financial position, any losses suffered by HLT in circumstances where the Lim Family was aware of the true position did not fall within the scope of Deloitte’s duty of care.

20 In response, HLT submitted that, owing to its insolvency at the time of the audits, Deloitte’s duty was “not restricted to the provision of accurate information to [HLT’s] shareholders” but extended to “protecting the company from undetected errors or wrongdoing”. Given that the creditors were the main economic stakeholders of a company while the company was insolvent, an auditor should be required to have regard to the interests of the creditors where the company was insolvent at the time of the audit (ie, the Creditor Duty). The existence of the Creditor Duty was also consistent with HLT’s case (in SUM 68) on: (a) the Reporting Duty (which formed the basis of the Reporting Duty Claim); and (b) the statutory duty of reporting to the Registrar of Companies under s 207 of the CA (which formed the basis of the Breach of Statutory Duty Claim). We pause here to mention that this pairing by HLT of the Creditor Duty with the Reporting Duty confirms our earlier observations at [13]–[14] above of the two being closely interrelated.

21 In reply, Deloitte argued that there was no basis for the Creditor Duty. There was no authority for such a novel proposition. And, to the extent that the Creditor Duty was based on an analogy to directors’ duties, such an analogy was misplaced as: (a) an auditor was not a fiduciary vis-à-vis the company; and (b) the roles of a director and an auditor were different. Taking a step back, the Creditor Duty was essentially a backdoor attempt at imposing on an auditor a duty to the creditors of the company when it was well-established that no such duty existed at law.

22 Apart from attacking HLT’s claim as a whole above, Deloitte also mounted a specific attack against HLT’s claim for its Trading Losses which, as mentioned earlier (at [5]), is the largest of the heads of loss which HLT seeks to recover in Suit 237. In this regard, Deloitte argued that this head of loss was not caused by any negligence on its part, or within the scope of its duty, because the losses arose from legitimate but loss-making trading activities. According to Deloitte, it was consistent with both principle and precedent that an auditor should not be liable for the losses suffered by a company arising from its legitimate trading activities as trading decisions were not a matter which an auditor had any involvement in.

23 On the other hand, HLT submitted that its Trading Losses were within the scope of Deloitte’s duty as Deloitte’s duty included protecting HLT against the risks of its audited accounts being inaccurate. Since Deloitte’s failure to detect the misstatements in HLT’s audited financial statements created the risk of HLT continuing to trade and incurring further losses, the Trading Losses were the eventuation of a risk that was within the scope of Deloitte’s duty of care. In this connection, HLT emphasised that its accounts had been deliberately prepared by the Lim Family on a “wholly false basis” as they concealed that HLT had been making massive losses and was deeply insolvent, and that its operations were “inherently loss-making” seeing as it had suffered losses in five out of the six years between FY2014 and FY2019. Relatedly, HLT also disputed Deloitte’s characterisation of HLT’s trading activities as “legitimate” on the basis that its continued trading which gave rise to the Trading Losses was conducted in a “fundamentally dishonest manner” due to HLT obtaining credit and financing which the Lim Family knew could not be repaid.

24 Turning to the Reporting Duty Claim and Breach of Statutory Duty Claim, which HLT sought to introduce by amending its pleadings in SUM 68, Deloitte submitted that both claims were unsustainable.

25 First, on the Reporting Duty Claim, Deloitte argued that it would be contrary to the purpose of its duty as an auditor – to report information to HLT’s shareholders – for it to be under the Reporting Duty, which contemplated a wider obligation to report information to persons other than the Lim Family such as government authorities. In any event, even if there was a Reporting Duty, it could only have been breached if the irregularities in HLT’s affairs had in fact been detected by Deloitte and were known to it. Since it was HLT’s own case that Deloitte had failed to detect the irregularities, Deloitte could not have breached the Reporting Duty by failing to report matters which it did not know of.

26 Second, on the Breach of Statutory Duty Claim, Deloitte submitted that: (a) the statutory duty of reporting breaches of the provisions of the CA to the Registrar of Companies under s 207(9) of the CA was not actionable by HLT as s 207(9) did not give rise to any private right of action; and (b) in any event, Deloitte did not breach its duty under s 207(9) for the same reason as it did not breach the alleged Reporting Duty, ie, Deloitte could not have failed to report the Lim Family’s breaches of the CA as it did not know of them.

27 HLT argued that it had a viable case on the Reporting Duty Claim. Similar to the Creditor Duty, HLT submitted that it was incorrect to limit the scope of Deloitte’s duty to the reporting of accurate information to HLT’s shareholders (see [20] above). In some cases, an auditor’s duty would extend to reporting any fraud or irregularity that was likely to result in material loss to the company to either the directors of the company or to a third party directly, especially where the auditor suspected that the management of the company was involved in the fraud or irregularity. The existence of the Reporting Duty was also supported by the relevant statutory framework which contained provisions that imposed on auditors an obligation in certain circumstances to report fraud or irregularities in the company’s affairs to persons outside of the company. These included: (a) s 207(9) of the CA; and (b) s 39(1) of the Corruption, Drug Trafficking and Other Serious Crimes (Confiscation of Benefits) Act (Cap 65A, 2000 Rev Ed) (which made it mandatory for a person to lodge a suspicious transaction report if information or matters came to his attention in the course of his trade, profession, business or employment gave him reason to suspect that certain property was linked to criminal conduct).

28 Moreover, the content of the Reporting Duty was not limited to matters within Deloitte’s actual knowledge but also included matters which would have come to its knowledge had it acted with reasonable skill and competence. This is because, HLT argued, the Reporting Duty was “founded on a claim for breach of [a] duty of care”. We pause here to make the observation again that, seeing as HLT’s own submission located the Reporting Duty within Deloitte’s “duty of care”, it is far from clear how the Reporting Duty Claim is distinct from the Negligence Claim and why the parties have continued to proceed on this basis.

29 Finally, in relation to the Breach of Statutory Duty Claim, HLT argued that s 207 of the CA did confer on HLT a private right of action in respect of Deloitte’s breach of its duty under s 207(9) of the CA. Moreover, as the issue of whether an auditor had breached s 207(9) should be assessed objectively based on the standard of a reasonably competent auditor rather than based on the auditor’s own subjective perception, the fact that Deloitte was not aware of HLT and the Lim Family’s breaches of the CA did not mean that Deloitte could not have breached its duty under s 207(9).

30 At first instance, the AR allowed SUM 68 in part in relation to the Reporting Duty Claim and dismissed SUM 474 in its entirety. Save for the Breach of Statutory Duty Claim, the AR did not think that any of HLT’s causes of action should be struck out or disallowed from being introduced into HLT’s pleadings by amendment.

31 First, the AR considered that the Creditor Duty was not foreclosed by the statutory framework or by the case law. In this regard, he saw force in an analogy to directors’ duties where it had been held, most prominently in this court’s decision in Foo Kian Beng v OP3 International Pte Ltd [2024] 1 SLR 361 (“Foo Kian Beng”), that the content of a director’s duty to act in the best interests of the company should be sensitive to the company’s financial health. We observe that the usage of “Creditor Duty” as a reference to HLT’s submission that Deloitte’s duty was not limited to providing HLT’s shareholders with accurate information appears to have originated here in the AR’s decision, as it does not appear to have been used by the parties in their submissions (at least not in their written submissions).

32 Second, the AR considered that the existence of the Reporting Duty was “not plainly unarguable”. The AR also did not agree with Deloitte that the Reporting Duty would be limited to matters that were in fact known to Deloitte, as he considered that the protection that the Reporting Duty afforded would be hollow if it was so confined.

33 Third, the AR rejected the Breach of Statutory Duty Claim as “plainly and obviously unsustainable” on the basis that nothing in s 207(9) of the CA conferred on HLT a private right of action for any breach of it by Deloitte. Accordingly, SUM 68 was only allowed in part in relation to the Reporting Duty Claim and not the Breach of Statutory Duty Claim.

34 Fourth, the AR held that HLT’s claim for its Trading Losses was not legally unsustainable. The AR considered the issue to be one of causation. He observed that neither Deloitte’s submission – which focused on how the Trading Losses had stemmed from legitimate but loss-making trades – nor HLT’s submission – that HLT’s trading was “inherently loss-making” – was attractive because the focal point of each of their arguments rested on the “protean” terminology of whether HLT’s continued trading should be seen as “legitimate” or “inherently loss-making”. However, the AR decided not to strike out HLT’s claim for its Trading Losses as causation was a “complex, fact-sensitive, policy-sensitive inquiry, on which reasonable persons can reasonably disagree”.

35 The parties filed three cross-appeals against the AR’s decision, which were heard by a Judge of the High Court (“Judge”): (a) in HC/RA 141/2024 (“RA 141”), Deloitte appealed against the AR’s decision to dismiss its striking-out application in SUM 474; (b) in HC/RA 142/2024, HLT appealed against the AR’s decision to refuse leave to amend its pleadings to include the Breach of Statutory Duty Claim in SUM 68; and (c) in HC/RA 143/2024, Deloitte appealed against the AR’s decision to allow HLT to amend its pleadings to include the Reporting Duty Claim in SUM 68. The Judge dismissed all three appeals for reasons which were generally similar to those given by the AR.

36 Following the Judge’s dismissal of its appeals against the AR’s decision, Deloitte brought two applications for permission to appeal against the Judge’s decision: (a) in AD/OA 17/2025 (“AD 17”), Deloitte applied to the Appellate Division of the High Court (“Appellate Division”) for permission to appeal against the Judge’s decision affirming the grant of leave to HLT to amend its pleadings to include the Reporting Duty Claim; and (b) in CA/OA 17/2025 (“CA 17”), Deloitte applied to this court for permission to appeal against the Judge’s decision refusing to strike out the Negligence Claim due to the plausible existence of the Creditor Duty, as well as HLT’s claim for its Trading Losses.

37 AD 17 was dismissed on the ground that a decision of a Judge of the High Court giving a party permission to amend its pleadings was a decision that could not be subject to a further appeal to either the Appellate Division or this court pursuant to s 29(b) read with para 1(i) of the Fourth Schedule to the Supreme Court of Judicature Act 1969 (2020 Rev Ed), save for two exceptions which were not applicable to the amendments pertaining to the Reporting Duty Claim. On the other hand, CA 17 was allowed based on Deloitte’s agreement to have all the facts alleged by HLT in its pleadings assumed true for the appeal, as well as HLT’s confirmation that it would not seek to file any further amendments or further particulars relating to the Creditor Duty and its claim for its Trading Losses after the appeal. In granting Deloitte permission to appeal in CA 17, this court also directed that the appeal should proceed as a summary determination of two questions of law pertaining to the existence of the Creditor Duty and the recoverability of HLT’s Trading Losses.

38 The parties have agreed that this appeal raises two issues:

(a) First, on the basis of the facts pleaded, whether Deloitte’s duty of care includes a duty to have regard to the interests of HLT’s creditors assuming HLT was insolvent at the time of the audits (“Creditor Duty Question”).

(b) Second, whether HLT’s Trading Losses are recoverable in law from Deloitte in the event that breach of duty is established (“Trading Losses Question”).

39 The parties’ cases on the Creditor Duty Question and the Trading Losses Question are essentially the same as the arguments made at first instance before the AR and also before the Judge. We thus do not propose to repeat them here and refer to our summary of their submissions at [19]–[23] above.

40 We begin with examining the general structure of liability for negligence. We do so because, as already mentioned above, and which will become clearer as we develop our analysis below, we consider that the way in which the parties have approached the matter has been plagued with some conceptual problems that have obscured what the true issues in dispute are.

41 It is perhaps surprising, given how ubiquitous tortious claims of negligence are, that the issue of its constituent building blocks is not a straightforward matter. As James Goudkamp & Donal Nolan, Winfield & Jolowicz on Tort (Sweet & Maxwell, 21st Ed, 2025) (“Winfield & Jolowicz”) comments, “[j]udges and commentators describe these in a surprising diversity of ways” (at para 3-004). A striking example of this can be seen in the decision of the UK Supreme Court in Meadows v Khan [2022] AC 852 (“Meadows”) where much, if not most, of the discussion across the three different judgments delivered did not concern the specific issue arising on the facts of that case, but concerned the conceptual structure of the tort of negligence. Indeed, the uncertainty on this front is apparent from how no consensus was reached: a majority of the court, comprising Lord Hodge DPSC and Lord Sales JSC (with whom Lord Reed PSC, Lord Kitchin JSC and Lady Black agreed), came up with a six-question framework (at [28]), one judge (Lord Burrows JSC) preferred what he considered was a more “conventional approach” made up of a series of seven questions (at [79]), while the last member of the court (Lord Leggatt JSC) opined that it was “undesirable as well as unnecessary to engage in such an exercise” (at [96]).

42 While we understand Lord Leggatt JSC’s reluctance to impose a structure for the tort of negligence by judicial pronouncement rather than developing the law incrementally, we are cognisant of what the majority of the court sought to do in terms of promoting analytical coherence. The risk of incoherence is especially acute in the context of liability for negligence. It is, in the first place, liability of an entirely judge-made origin, even if modifications have been made by statute: see, for example, the Contributory Negligence and Personal Injuries Act 1953 (2020 Rev Ed). This means that, unlike statutory causes of action where the relevant statutory language provides an anchor from which the ingredients of liability can be discerned through a process of exegesis, the conditions for and limits of liability for negligence are left entirely to the courts to develop. The difficulty in this area is that liability for negligence may potentially expand unchecked over time because any unexpected damage suffered by persons in the course of their lives can often be traced back to some form of negligent conduct by other persons. Thus, in Spandeck Engineering (S) Pte Ltd v Defence Science & Technology Agency [2007] 4 SLR(R) 100 (“Spandeck”), this court noted that the perennial challenge for the courts in the tort of negligence was to identify “legal control mechanisms” for liability in a way that struck an appropriate balance between “a fair and just result for the parties and the imposition of indeterminate liability on an indeterminate class of tortfeasors” (at [29]–[30]). This echoes Professor John Fleming’s observation that “[t]he basic problem in connection with the ‘tort’ of negligence is … that of limitation of liability”, and “[t]he mechanisms associated with liability for negligence, such as the duty and causation concepts, are nothing more or less than the control devices fashioned by the courts to achieve that purpose”: see J G Fleming, “Remoteness and Duty: The Control Devices in Liability for Negligence” (1953) 31 Canadian Bar Review 471 at 474.

43 In view of this, we consider that there is no objection to a court setting out what the elements of the tort of negligence are, especially if this is done at a relatively general and uncontroversial level. Frameworks like those which some of the members of the court in Meadows sought to devise provide broad headers that assist in identifying the relevant issues and imposing some sort of order and structure in analysis. This conduces to clarity of thought and transparency in the reasoning for and against imposing liability on a given set of facts. In contrast, as Professor James Goudkamp explains, “[d]ispensing with formal prerequisites to liability risks creating a legal vacuum in which judges would impose or withhold liability based on an instinctive sense of whether a wrong had been committed”: James Goudkamp, “Elements of Torts” in Taking Law Seriously: Essays in Honour of Peter Cane (James Goudkamp, Mark Lunney & Leighton McDonald eds) (Hart Publishing, 2022) at p 44. We agree. There are of course grey areas that involve finer and more technical distinctions and which are thus prone to more pronounced differences in opinion. But, in our view, these do not detract from the core utility of identifying, in broad terms, what the elements of a claim in negligence are, and in turn, what issues the parties should direct their minds to. As will be seen below, the present case demonstrates in more than one way how this essential step can sometimes be overlooked.

44 In this regard, we find the following five broad requirements that Professor Jane Stapleton has identified to be quite useful; her approach is generally consistent with the way courts in this jurisdiction have conceived of the elements of liability for negligence (see Jane Stapleton, “Conceptual Interplay between Elements of the Tort of Negligence” in Jane Stapleton, Three Essays on Torts (Oxford University Press, 2021) at pp 65–66):

To these five headers, we may add a sixth requirement, which is the absence of any operative defences, for instance, illegality or contributory negligence: see Winfield & Jolowicz at para 3-004.

45 The first element is that the claimant must have suffered some sort of damage which is legally recognised in the tort of negligence. It is, after all, trite that damage is the “gist of the action” of negligence in that negligence is only actionable upon proof of damage: see ACB v Thomson Medical Pte Ltd [2017] 1 SLR 918 (“ACB”) at [47]–[51]. In this regard, “damage” is generally understood as “an abstract concept of being worse off, physically or economically, so that compensation is an appropriate remedy” [emphasis added]: Rothwell v Chemical & Insulating Co Ltd [2008] 1 AC 281 at [7]. The qualifier of “actionable” damage, as opposed to damage simpliciter, contemplates that there must be an “interference with a right or interest recognised as capable of protection by law” [emphasis added]: Cattanach v Melchior (2003) 215 CLR 1 at [23].

46 The present case concerns a straightforward claim by HLT for economic loss. That is a well-established form of actionable damage.

47 The second element is that the defendant must have owed the claimant a duty of care. The modern root of the juristic device of a “duty of care” is the decision of the House of Lords in M’alister (Or Donoghue) (Pauper) v Stevenson [1932] AC 562, in which Lord Atkin famously articulated the “neighbour principle” (at 580), but in Singapore, the framework applied by our courts to determine the existence of a duty of care is that set out in this court’s decision in Spandeck. The Spandeck framework, in broad terms, contemplates the court considering: (a) first, as a threshold matter, whether it was factually foreseeable that the claimant would suffer damage from the defendant’s carelessness (at [75]–[76]); (b) second, whether there is sufficient legal proximity between the parties for a prima facie duty of care to arise (at [77]–[82]); and (c) third, whether there are any policy considerations which negate the imposition of a duty of care (at [83]–[85]).

48 In most cases, however, the existence of a duty of care would not be disputed or open to any real dispute. Indeed, the present case, involving an auditor and the audited company, is a category of case where a duty of care is essentially beyond question: JSI Shipping (S) Pte Ltd v Teofoongwonglcloong [2007] 4 SLR(R) 460 (“JSI Shipping”) at [30]. In such cases, it is in our view preferable for the court to resolve the issue of duty of care in the affirmative and move on to other stages of the analysis. Questions as to what the defendant should or should not have done, or whether the defendant should be responsible for a certain loss suffered by the claimant, should not be subsumed into the duty of care inquiry.

49 But, as we expand on at [70]–[89] below, it appears that there is a common tendency for parties and courts to frame issues which are not properly issues of a duty of care in these terms. To use the present case as an example, Deloitte has framed both of its grounds for striking out HLT’s claim on the basis that HLT’s losses, or its Trading Losses specifically, are not within the scope of its duty of care (see [19] and [22] above). As we explain below, however, Deloitte appears not to appreciate that it has used the phrase “scope of duty” in no less than three different senses and across three different elements of the tort of negligence, none of which is properly a question of whether it owes HLT a duty of care (which it quite obviously does) (see [102] below). Likewise, in its submission that Deloitte’s duty of care includes a duty to “have regard to the interests of [its] creditors”, HLT’s focus is on what Deloitte should have done in the discharge of its duty of care, and not whether Deloitte owed a duty of care to it.

50 The third element is that the defendant must have acted in breach of his duty of care. This is invariably a question of fact. It is worth emphasising, as a point of some importance in the present case, that the content of a “duty of care” is exactly what is described by that phrase. It is a duty to exercise reasonable skill and care so that injury or loss is not caused to the claimant, and not a duty to do any specific act or acts. Reasonable skill and care may require a defendant to take a certain course of action in the circumstances at hand, but the need to take such action is incidental to the duty of care rather than a duty in and of itself. Thus, when a claimant frames an allegation of breach of duty, the ultimate question that has to be decided is not whether the defendant failed to do any specific thing but whether the failure to do a certain thing constituted a failure to act with due skill and care: Go Dante Yap v Bank Austria Creditanstalt AG [2011] 4 SLR 559 (“Go Dante Yap”) at [19].

51 The decision of the English Court of Appeal in N v Agrawal [1999] PNLR 939 is instructive. In that case, Stuart-Smith LJ rejected the lower court’s finding that a prosecution expert had breached a duty to attend court to give evidence and was liable for psychiatric injury suffered by the alleged victim of the crime as follows (at 943):

As Clerk & Lindsell on Torts (Andrew Tettenborn gen ed) (Sweet & Maxwell, 24th Ed, 2023) (“Clerk & Lindsell on Torts”) explains with reference to this dictum, expressing a standard of care in terms of a duty “confuse[s] two distinct questions”, one as to whether there was a duty of care (“whether the nature of the relationship [between the parties] reasonably requires that care be taken”) and another as to what the duty of care required the defendant to do (“what conduct is reasonably required in the particular circumstances”) (at para 7-162).

52 It is useful to take the fourth and fifth elements (factual causation and the extent of legal responsibility) together because they are not always clearly defined. We thus begin with disentangling the two before elaborating on them individually.

53 On a somewhat reductionist view, any question of whether a given defendant who has acted in breach of duty should be liable to compensate the claimant for certain damage suffered by the latter can be put as an overarching question of “causation”. In JSI Shipping, we noted the multi-faceted usage of “causation” in the law with an observation that “[c]ausation is often a thicket of complex factual, legal and policy issues” (at [140]). A survey of case law across multiple jurisdictions will bear out that courts do not always use “causation” in a uniform sense, as Professor Stapleton has noted (see Jane Stapleton, “Occam’s Razor Reveals An Orthodox Basis for Chester v Afshar” (2006) 122 LQR 426 at 426):

54 However, it would be preferable, in our view, if care was taken in the use of the language of causation, even if it may not be possible to clearly demarcate the boundaries of where it is appropriate. In this regard, a useful starting point is the exposition of Lord Nicholls of Birkenhead in the decision of the House of Lords in Kuwait Airways Corpn v Iraqi Airways Co (Nos 4 and 5) [2002] 2 AC 883 (“Kuwait Airways”), which this court cited with approval in JSI Shipping (at [142]). Drawing on the writings of Professor Stapleton, Lord Nicholls opined that the courts should approach the causation inquiry in two stages (at [69]–[70]):

55 A number of important points are made in this extract. They can be unpacked as follows:

(a) The identification of the defendant’s liability for the claimant’s loss comprises two stages: (i) the causal relevance of the defendant’s wrongful conduct to the claimant’s loss; and (ii) the extent of the loss for which the defendant ought to be liable.

(b) The first stage is a question of fact that is typically expressed in terms of a test of “but for” causation.

(c) Mere factual causation between the defendant’s breach and the claimant’s loss is a necessary but insufficient condition for the defendant to bear legal responsibility for it. The second stage limits the over-inclusiveness of the first stage by requiring the court to make a “value judgment” as to the appropriate scope of the defendant’s legal responsibility.

(d) The second stage may require consideration of different legal rules and doctrines. These include remoteness, mitigation, and novus actus interveniens (intervening cause). But the common thread running through them is that they are devices developed by the common law to limit the defendant’s responsibility for the consequences of his conduct.

56 We note that Lord Nicholls’ approach finds resonance with the current approach in Australia, where legislation has been enacted in terms that expressly draw a distinction between the question of “factual causation” – which asks if “the negligence was a necessary condition of the occurrence of the harm” – and the question of “scope of liability” – which asks if “it is appropriate for the scope of the negligent person’s liability to extend to the harm so caused”: see, for example, s 5D(1) of the Civil Liability Act 2002 (NSW). In the unanimous decision of the High Court of Australia in Wallace v Kam (2013) 250 CLR 375, the court began by observing that the “familiar elements” of the cause of action of negligence were “duty, breach and causation of damage” (at [7]), but went on to unpack the element of “causation” as consisting of two questions which have often been conflated: (a) first, “a question of historical fact as to how a particular harm occurred”; and (b) second, “a normative question as to whether legal responsibility for that particular harm occurring in that way should be attributed to a particular person” (at [11]). We agree with this dichotomy and elaborate on each question below.

57 The element of factual causation, as defined by this court in Sunny Metal & Engineering Pte Ltd v Ng Khim Ming Eric [2007] 3 SLR(R) 782 (“Sunny Metal (CA)”), is “concerned with the question of whether the relation between the defendant’s breach of duty and the claimant’s damage is one of cause and effect in accordance with scientific or objective notions of physical sequence” (at [52]). In short, the defendant’s breach of duty must be historically involved in producing the outcome of the claimant’s damage.

58 In law, the ordinary standard of causal relevance that is required is that of necessity. This finds expression in what has been termed the “but for” test: but for the defendant’s wrongdoing, would the claimant have suffered the damage that he complains of? There might be circumstances in which the court will apply a different test of factual causation, but it is unnecessary for us to consider those exceptional cases for the purposes of the general overview in this judgment.

59 It would, however, be overinclusive for the defendant to be legally responsible for all the consequences created by his conduct as “the consequences of an act theoretically … stretch into infinity”: Sunny Metal (CA) at [56]. In this regard, the question of legal responsibility, as alluded to by Lord Nicholls in Kuwait Airways above, and as observed by Lord Hodge DPSC and Lord Sales JSC in Meadows, is not a monolithic one but “in reality a number of separate questions which must be addressed because the law does not impose responsibility on a defendant for everything that follows from his or her act or omission, even if it is wrongful” (at [55]). In our view, the doctrines which come under the collective umbrella of the legal responsibility question can broadly be placed into two categories which are distinguished by what they focus on:

(a) The first concerns the attribution of responsibility as between two or more causes of the claimant’s damage (including the defendant’s wrongdoing), each of which are “but for” causes in the sense of being necessary conditions for creating that damage. This has been referred to variously in terms of whether the defendant’s wrongdoing is a “legal cause” or whether there is some other factor which constitutes a “novus actus interveniens”. The principles of mitigation, as explained by this court in POP Holdings Pte Ltd v Teo Ban Lim [2025] 2 SLR 90 and Kupetz, Jonathan v Terraform Labs Pte Ltd [2026] 1 SLR 355, also fall within this category in so far as avoidable loss that is incurred due to the claimant’s failure to mitigate is attributable to that failure rather than the defendant’s wrongdoing.

(b) The second category concerns the foreseeability of the damage suffered by the claimant. Different approaches have been applied in this category based on whether the cause of action rests on contract or tort (see [67]–[68] below).

60 Before turning to outline the two categories above, we make three preliminary comments. First, as a matter of design, it is possible, if one seeks to formulate the conceptual structure of negligence in finer terms, to use each of the doctrines that attenuate legal responsibility as individual elements of the cause of action. Thus, one could, for example, as is not infrequently done, refer to “legal causation” and “remoteness” as separate constituents of the tort of negligence. But for present purposes, we have considered it useful to classify them under the broad header of “legal responsibility” as that is really the common thread running through them even if they each have a different focus. Our main concern in the present case is drawing up a broad structure that allows parties to identify and focus on the real issue based on the relevant stage or element of the negligence framework that is in play.

61 The second point is that, in line with our earlier observation that some care ought to be exercised in the use of the language of causation, it would be preferable to confine the use of causal language (if it is to be used at all) to cases in the first category identified at [59] above, that is, where the parties have identified more than one cause for the claimant’s damage and seek to attribute legal responsibility as between them. In that scenario, it is meaningful, to some extent, to speak of causation in relative terms as between the multiple causes, even if what is really being considered is not so much a question of breaking the chain of causation but the chain of legal responsibility. This is because, if the defendant’s wrongdoing is a necessary condition for the claimant’s damage, there is ex hypothesi an unbroken chain of causation.

62 The final point is that, while we accept that the two categories we have set out at [59] above may not be a closed list of the legal concepts that regulate the extent of legal responsibility, the courts should be careful to avoid a proliferation of novel doctrines and principles beyond what is necessary. Such proliferation will create uncertainty in the law and undermine consistency in reasoning and outcomes. What this means in practice is that, if it is possible for an outcome to be rationalised in terms of an established doctrine, that explanation ought to be preferred.

63 The present case is a good example of this. In support of its claim that it should not be made liable for HLT’s Trading Losses, Deloitte cites case law that has variously reasoned in terms of the audited company’s trading losses being “not caused” by the auditor’s negligence, or in terms of such losses falling “outside the scope of duty” of auditors based on a principle apparently established in the decision of the House of Lords in South Australia Asset Management Corporation v York Montague Ltd [1997] AC 191 (“SAAMCo”) as a limit on legal responsibility that is distinct from causation and remoteness. The decision in SAAMCo has generated confusion as to when this “scope of duty” principle applies and how it fits within the general structure of negligence; indeed, this was the issue at the heart of Meadows and the case of Manchester Building Society v Grant Thornton UK LLP [2022] AC 783 (“Manchester Building Society”) which was decided in parallel with Meadows. The consensus that has been reached by the English courts appears to be that the so-called “SAAMCo principle” is a doctrine with a distinct identity and which applies in the specific context of professional negligence or situations analogous to it. But, as we explain below, we consider that the SAAMCo principle is an application of, and can be explained through, orthodox principles of contractual remoteness. This avoids having to define a principle by reference to a specific case – which, as this court remarked in Natixis, Singapore Branch v Seshadri Rajagopalan [2025] 1 SLR 1020, is often “a sign of difficulty in understanding its rationale” (at [126]) – and limiting its application in a somewhat arbitrary fashion to a specific type or context of negligence claims.

64 The concept of “legal causation” is usually invoked where it is alleged that there is some necessary condition other than the defendant’s breach – a novus actus interveniens – that absolves the defendant of legal responsibility for the claimant’s damage. As mentioned, this is often expressed in the familiar language of a breakage in the chain of causation, but what is broken is not so much the chain of causation but the chain of responsibility for the consequences of the defendant’s breach.

65 The types of events which can constitute a novus actus interveniens are usually expressed in three categories: (a) a natural event independent of any human agency; (b) an act or omission of a third party; or (c) the conduct of the claimant himself: Clerk & Lindsell on Torts at para 2-107; Andrew Burrows, Remedies for Torts, Breach of Contract and Equitable Wrongs (Oxford University Press, 4th Ed, 2019) at pp 107–117. We propose to only address the last of these briefly in so far as it will be relevant to our analysis below. The legal position is that it will generally take rather exceptional or abnormal conduct by the claimant to break the chain of responsibility as between the defendant’s breach and the claimant’s damage. Thus, earlier decisions of this court have set the benchmark of the claimant’s conduct having to be “wholly unreasonable”: ACB at [84]; TV Media Pte Ltd v De Cruz Andrea Heidi [2004] 3 SLR(R) 543 at [76].

66 Finally, even when a different cause is not treated as so significant as to completely sever the chain of responsibility between the defendant’s wrong and the claimant’s damage, that is not necessarily the end of the matter as attribution of responsibility does not only occur on an all-or-nothing basis at the liability stage. Instead, it may also take place in the quantification of damages through the application of the principles of contributory negligence: see, for example, PlanAssure PAC v Gaelic Inns Pte Ltd [2007] 4 SLR(R) 513 at [99] and [117]–[120] (in the context of a claim for auditor negligence).

67 Apart from attribution of responsibility as between multiple causes, the law also recognises that foreseeability of damage may be a different reason for attenuating the defendant’s legal responsibility. The leading case in the context of tort law is the decision of the Judicial Committee of the Privy Council in Overseas Tankship (UK) Ltd v Morts Dock & Engineering Co Ltd (The Wagon Mound) [1961] AC 388 (“Wagon Mound”), which establishes that loss is too remote to be recoverable if it is of a type that was not reasonably foreseeable at the time of the defendant’s breach of duty. The test in Wagon Mound has been affirmed and applied by our courts: see, for example, Ho Soo Fong v Standard Chartered Bank [2007] 2 SLR(R) 181 at [39]; Sunny Metal (CA) at [56].

68 In this case, an important dimension that the parties have tended to overlook in their submissions is that HLT has mounted a claim for negligence not only in tort but also in contract based on an implied term that Deloitte would exercise reasonable skill and care in the conduct of the audit (see [12] above). This is important because the contractual remoteness rules established in Hadley v Baxendale (1854) 9 Exch 341 (“Hadley v Baxendale”) are narrower than the Wagon Mound test, and it is generally accepted that the contractual rules would apply in a case where the defendant owes the claimant concurrent duties of care in contract and tort: Sunny Metal & Engineering Pte Ltd v Ng Khim Ming Eric [2007] 1 SLR(R) 853 (“Sunny Metal (HC)”) at [139]–[140]; BDW Trading Ltd v URS Corporation Ltd [2025] 2 WLR 1095 at [33]. The rule in Hadley v Baxendale, in brief, contemplates that a defendant should be liable for such damage that he is taken to have assumed responsibility for based on it having been in his contemplation at the time of contracting (see [149] below). In this regard, the defendant’s horizon of contemplation is made up of: (a) “ordinary” or “natural” damage arising from the breach of contract which the defendant as a reasonable person would be imputed with knowledge of; and (b) “extraordinary” or “non-natural” damage not arising in the usual course of things but from special circumstances which the defendant has actual knowledge of: Robertson Quay Investment Pte Ltd v Steen Consultants Pte Ltd [2008] 2 SLR(R) 623 (“Roberston Quay”) at [81]–[82]. The rationale behind the narrower scope of recoverable damage under the Hadley v Baxendale test as compared to the Wagon Mound test is that contracting parties are not strangers and would therefore have had an opportunity before the defendant’s breach to make arrangements to allocate the risk of damage as between them through the contract: Koufos v C Czarnikow Ltd [1969] 1 AC 350 (“The Heron II”) at 385G–386C; Sunny Metal (HC) at [138]; Robertson Quay at [75]–[76].

69 Finally, upon the claimant establishing the five elements above, the defendant may nonetheless escape liability either in full or in part through certain defences. We have already noted contributory negligence at [66] above as one example. It suffices for us to make mention of another, the defence of illegality, which denies a claim on public policy grounds and which, as we will briefly note below, is of some relevance to Suit 237 even if it does not strictly arise in this appeal.

70 The purpose of the exercise above of identifying the different elements of a negligence claim is to promote clarity in identifying what the true issue in dispute is in any given case. Each element raises specific issues and it does not assist for parties to attempt to frame what are in substance arguments on breach of duty or causation as a question of duty of care. Unfortunately, conflation of this kind has proved rather pervasive in the parties’ pleadings and submissions in this case.

71 Before turning to the case at hand, it would be useful to consider some examples of cases in which the wrong question has been asked and answered. We highlight two types of cases which reflect similar errors as those committed by the parties before us. The first is where what is really a question of breach is framed as a question of duty of care. This is typically done by defining the duty of care not as a general duty to take care but in terms of the specific conduct that the defendant (allegedly) should or should not have done. The second is where what is in truth an issue of remoteness is brought forward to the issue of duty of care by defining the duty not in terms of damage generally but as a duty in relation to a specific risk of damage.

72  An example of the first type of case is the decision of the English Court of Appeal in Darnley v Croydon Health Services NHS Trust [2018] QB 783 (“Darnley”). The claimant there had received a serious blow to his head in an assault. He attended at a hospital managed by the defendant, where he was told that he would not be seen for up to four or five hours. This information was incorrect as the hospital’s protocol required a person with a head injury to be seen by a triage nurse within 30 minutes. After waiting for about 19 minutes, the claimant elected to go home. His condition subsequently deteriorated and he was reconveyed to hospital for surgery. By then, however, the claimant had unfortunately suffered serious and permanent brain damage.

73 The defendant admitted that, if the claimant had remained present when called for triage, his treatment would have been prioritised, and he would have made a full recovery (at [18]). At first instance, the claim was dismissed on the basis, among others, that the defendant did not owe the claimant a duty of care to give accurate information about waiting times. A majority of the English Court of Appeal upheld this decision. Jackson LJ, as part of the majority, said that it would not be “fair, just and reasonable to impose upon [the defendant] a duty not to provide inaccurate information about waiting times” as this would “add a new layer of responsibility to clerical staff and a new head of liability for NHS health trusts” (at [53]).

74 The majority’s approach was criticised by Professor Goudkamp as an “erosion of the customary divide between the duty of care and breach elements of the action in negligence”: James Goudkamp, “Breach of Duty: A Disappearing Element of the Action in Negligence?” (2017) 76 CLJ 480 (“Breach of Duty”) at 481. The majority’s focus on the defendant’s duty of care was “surprising given that it is, of course, trite law that hospitals owe a duty to their patients”. In his view, the real issue was not whether the defendant owed a specific duty of care to give accurate information but whether the defendant had breached its duty of care by giving inaccurate information (see Breach of Duty at 482):

75 The UK Supreme Court later allowed an appeal against the English Court of Appeal’s decision: see Darnley v Croydon Health Services NHS Trust [2019] AC 831. Lord Lloyd-Jones JSC, delivering the judgment of the court, explained that “[t]his [was] a distinct and recognisable situation in which the law imposes a duty of care” (at [16]), and agreed with Professor Goudkamp that the majority of the English Court of Appeal had “elide[d] issues of the existence of a duty of care and negligent breach of duty” (at [21]–[23]).

76 In our view, Darnley illustrates that the duty of care inquiry is concerned with the existence of a duty of care, that is, whether there is a need on the facts for the defendant to take care to avoid causing harm to the claimant. The duty of care inquiry is not concerned with what steps would or would not amount to sufficient care on the facts of the case. If the duty of care is defined in terms of specific acts which the defendant should or should not have done, the questions of duty and breach merge entirely into one another. This is essentially the same caution made by this court in Go Dante Yap which we have referred to at [50] above.

77 Keeping duty and breach distinct is not merely a matter of conceptual tidiness. Asking the wrong question can sometimes lead the parties and the court into error. We highlight two potential pitfalls. First, while the issue of whether the defendant owes the claimant a duty of care is a question of law, the issue of whether the defendant has breached his duty of care is a question of fact: JSI Shipping at [32], [47] and [65]; Gary Chan Kok Yew, The Law of Torts in Singapore (Academy Publishing, 2nd Ed, 2015) at para 06.003. The fact-law distinction was historically of especial importance when the arbiter of fact was a jury rather than the judge as it would have been an error on the part of the judge to usurp the fact-finding role of the jury. While the judge in our legal system is both arbiter of fact and law, the division of issues as between judge and jury remains a useful reminder of the true nature of the issue(s). In this regard, it is well-established that “[w]hat is reasonable care in a particular circumstance is a jury question”: Wooldridge v Sumner [1962] 2 All ER 978 at 988G. Consequently, a finding in one case that the defendant was liable due to a failure to do a certain act is, strictly speaking, a finding of fact and does not establish any precedent of law: Qualcast (Wolverhampton) Ltd v Haynes [1959] AC 743 (“Qualcast”) at 757–758. It would thus be fallacious to treat such a finding as authority for a legal proposition that a failure to do the same act necessarily constitutes a breach of duty on the facts of a different case. As Lord Denning explained in Qualcast, while the Highway Code “contains many propositions of good sense which may be taken into account in considering whether reasonable care has been taken … it would be a mistake to elevate them into propositions of law” (at 759). More recently, the same point was made by McHugh J in the High Court of Australia’s decision in Vairy v Wyong Shire Council (2005) 223 CLR 422 (at [29]):

78 The different natures of the duty and breach inquiries can have consequences in practice. One is that, as a question of fact, the applicable standard of care and whether the defendant failed to act in accordance with that standard would rarely be appropriate for determination on a summary basis. Another is that an appellate court would be slow to intervene in a trial judge’s findings on breach of duty as the usual restraints on appellate intervention in matters of fact would apply, whereas an appellate court is at greater liberty to intervene in a lower court’s decision on duty of care as it is an issue of law: James Plunkett, The Duty of Care in Negligence (Hart Publishing, 2018) (“Plunkett”) at p 127.

79 Second, framing the duty of care in the binary terms of a duty to do or refrain from doing a certain act can create the misleading impression that the defendant’s liability is strict as opposed to being tested on a standard of reasonableness. The formulation that A owes B a duty to do a certain thing when used in the context of a claim for negligence contains an a priori assumption that a failure to do that thing is invariably a failure to act with reasonable care: Plunkett at p 126. But as Lord Somervell of Harrow warned in the House of Lords’ decision in A C Billings & Sons Ltd v Riden [1958] AC 240, an ossified definition of the content of A’s duty of care should be avoided as the dictates of reasonable care are necessarily specific to the facts of each case (at 264):

In the final analysis, one should not lose sight of the point that the duty of care is “a broad duty to take such care as is reasonable in the circumstances”, and not a specific obligation to do or not do any specific acts: Go Dante Yap at [19].

80 The second category of case involves the framing of duties of care in relation to specific forms of damage or the risk of such damage. An oft-cited basis for such an approach is the following statement of Lord Bridge of Harwich in the famous decision of the House of Lords in Caparo Industries plc v Dickman [1990] 2 AC 605 (“Caparo”) (at 627D):

As will be seen below, this statement in Caparo is the bedrock of the “SAAMCo principle”, which is a principle that purports to limit the defendant’s liability by defining his duty of care as excluding the specific loss that the claimant has suffered. Our observations here on the difficulties with such an approach would therefore telegraph the reasons for our view that the SAAMCo principle is no more than an application of the principles of remoteness of damage.

81 In our judgment, framing a duty of care in relation to a specific type of damage confuses the question of duty of care and the question of remoteness. Whether the defendant should be responsible for a certain type of damage or loss is properly the territory of the latter. The duty of care as formulated by our courts is a wide concept that is focused on the need to take care to avoid causing harm generally. Correctly understood, the concept of a duty of care to avoid a specific type of harm is not consistent with the way our courts have approached the matter. This stands to reason as the purpose of laying down a universal framework for determining the existence of a duty of care in Spandeck was to do away with traditional distinctions based on the kind of loss that the claimant had suffered.

82 This can be seen from Spandeck itself. In that case, we held that the threshold requirement of factual foreseeability only required that “the defendant ought to have known that the claimant would suffer damage from his (the defendant’s) carelessness”, and it was because of this that “it would be fulfilled in almost all cases” [emphasis added] (at [75]). Indeed, we went on to expressly reject the respondent’s submission that the requirement was not met on the facts because the respondent had approached the issue at too granular a level in a manner that was more suited to the issue of remoteness (at [89]):

More recently, in NTUC Foodfare Co-operative Ltd v SIA Engineering Co Ltd [2018] 2 SLR 588 (“NTUC Foodfare”), we reiterated that the Spandeck test had made it “no longer necessary to characterise the nature of the plaintiff’s loss before examining whether a duty of care arises in tort” (at [4]).

83 It might be argued that it is unduly pedantic to draw a clear separation between the duty of care and remoteness inquiries, and that there is nothing wrong with framing the duty of care in relation to a specific kind of damage that the claimant has suffered. We disagree. If the defendant wishes to defend the claim against it by conceding the issue of duty of care and focusing instead on arguing that the claimant’s damage is too remote, there is nothing wrong in this. But what is not correct is to distort the concept of a duty of care by building notions of remoteness into it:

(a) Fundamentally, the duty of care and remoteness elements serve different purposes. While both are concerned with the general problem of indeterminacy of negligence liability, the duty element is specifically concerned with indeterminacy in the persons to whom the defendant may be liable, whereas remoteness is concerned with indeterminacy in the extent of loss suffered by the determinate class of persons to whom the defendant is liable: NTUC Foodfare at [42]–[43]. The issue of duty is thus a logical antecedent to the issue of remoteness. Thus, in a case like the present where the parties are in a contractual relationship and the duty of care arises in both contract and tort, there is plainly no question of the defendant being liable to an indeterminate class of persons. The question is the extent of the claimant’s damage that the defendant should be legally responsible for.

(b) Moreover, a failure to correctly approach an attempt to limit the extent of the defendant’s liability through the lens of remoteness can cause the parties and the court to lose sight of the relevant considerations in the analysis. We have said at [62] above that the law should avoid an unnecessary proliferation of novel doctrines and principles which bear on the question of the extent of the defendant’s legal responsibility for the claimant’s damage as this would create uncertainty and potential inconsistency. As prefaced there, we consider that the law took a wrong turn in developing the “SAAMCo principle” as a result of attempting to resolve an issue of remoteness at the duty of care level. If no such conflation had occurred and the issue was approached as one of remoteness, it would have been apparent that the orthodox principles on remoteness of damage were an adequate solution.

84 We return to Caparo to illustrate these points. The facts of the case are well known. The claimant, Caparo, began acquiring shares in a publicly listed company, Fidelity, after Fidelity’s accounts for the year ended 31 March 1984 were audited by the defendant firm of auditors and the results were announced to the market. By the time of the annual general meeting where Fidelity’s 1984 accounts were approved, Caparo was a shareholder of Fidelity. However, Caparo continued to purchase shares in Fidelity until its stake reached the 30% threshold which triggered an obligation under the stock exchange rules to make a general offer for all the shares in Fidelity. Caparo did so and acquired all the shares in Fidelity by end-1984. Subsequently, Caparo brought a claim against the defendant alleging that it had relied on the 1984 accounts in deciding to acquire shares in Fidelity, but that the accounts had been misstated and showed a profit instead of a loss.

85 The House of Lords dismissed the claim on the basis that the defendant did not owe Caparo a duty of care in its capacity as an investor who had acquired new shares in Fidelity. Lord Bridge, delivering the lead judgment, considered it clear that there was “a relationship between the auditors and the shareholders of a company on which the shareholder [was] entitled to rely for the protection of his interest” (at 626C). But the crucial question, his Lordship said, was “the extent of the shareholder’s interest which the auditor ha[d] a duty to protect” (at 626C). In this regard, Lord Bridge considered that an auditor had to protect the shareholders’ “collective interest in the company’s proper management” by providing them with accurate information that would allow them to “call the directors to book and to ensure that errors in management are corrected”. In practice, however, since the company would likely have its own cause of action against the auditor, it would not be necessary for an individual shareholder to bring proceedings against the auditor as the losses of the shareholders would be recouped by a claim by the company on its own cause of action against the auditor (at 626C–F). We pause here to observe that, as regards the last point made by Lord Bridge, although his Lordship did not refer to it specifically, he appears to have had in mind the principle against claims for reflective loss, which bars recovery by shareholders for loss suffered in the form of diminution in the value of their shares or distributions from the company where such loss is merely reflective of loss the company could recover pursuant to its own rights: see Simon Salzedo & Tony Singla, Accountants’ Negligence and Liability (Bloomsbury Professional, 2nd Ed, 2021) at para 5.24. In that event, it may be more accurate to say that it is impermissible as a matter of principle, rather than merely unnecessary “in practice”, for the shareholders to bring a claim against the auditor as such a claim would be for loss for which recovery is barred by the reflective loss principle: see generally, Miao Weiguo v Tendcare Medical Group Holdings Pte Ltd [2022] 1 SLR 884. This, however, does not change the fact that Lord Bridge accepted that an auditor did owe a duty of care to a shareholder to the extent of the shareholder’s interest in ensuring proper management of the company.

86 Following from this, Lord Bridge went on to state his conclusion, which began with his oft-cited statement which we have referred to at [80] above, as follows (at 627D–F):

87 As noted earlier, this reasoning conflates the questions of duty of care and remoteness. Lord Bridge accepted (or, at least, was content to assume) that an individual shareholder was owed a duty of care. What his Lordship was really concerned with was that, as far as Caparo was seeking to recover loss suffered from its acquisition of new shares in Fidelity, and not loss suffered in respect of its existing shares, such loss was outside the “scope” of the defendant’s duty of care to Caparo. In our view, this makes it clear that the true issue in Caparo was not so much whether the defendant owed a duty to Caparo – it did, since Caparo was a shareholder – but whether the loss that Caparo sought to recover was too remote to be recoverable.

88 Seen from a different perspective, there was no real concern of liability to an indeterminate class of persons (which is the issue addressed by the duty of care element) in Caparo. Given that Caparo was a shareholder of Fidelity, the defendant’s liability was limited to a defined class of persons (ie, the shareholders of Fidelity). Instead, the concern which the House of Lords sought to deal with in Caparo was that of indeterminacy in the extent of the defendant’s liability for the loss suffered by a person in the determinate class who it owed a duty of care to – ie, whether the defendant should be liable for all of Caparo’s losses. As Lord Bridge put it, the issue was the “extent of [Caparo’s] interest which the [defendant] ha[d] a duty to protect”.

89 Applying the established test of remoteness in Wagon Mound to the facts of Caparo, loss suffered by persons from acquiring new shares in Fidelity (the loss which Caparo was seeking to recover) was not reasonably foreseeable. This is a better explanation for Caparo, and one that avoids the criticism of being “[an] unhelpful and pointless reformulation of the remoteness issue as whether or not a particular consequence of the defendant’s negligence fell within the scope of the duty of care owed”: Donal Nolan, “Deconstructing the Duty of Care” (2013) 129 LQR 559 at 580.

90 Having set out the conceptual points above, we turn to the specific issues in this case.

91 We begin with the Creditor Duty Question. We dispose of it on the basis that it is an academic issue. This is because of two connected reasons. First, it has become abundantly clear from the parties’ arguments in the appeal that HLT is not advocating that there is some independent duty on the part of an auditor to take into account the interests of the creditors of the audited company if, during the time when the audit is conducted, the company is insolvent. In fact, both sides are in agreement, and rightly so, that when an auditor performs his audit, the standard by which he will be judged to be negligent is the objective yardstick of an auditor acting with reasonable skill and care, and not whether he has subjectively taken into account the interests of the company’s creditors when the company is insolvent. This will be expanded upon below at [112]–[116].

92 Second, and more significantly, it is agreed between the parties that, even if the Creditor Duty Question can be (and is) resolved in Deloitte’s favour and the Negligence Claim is struck out, any pleadings which form part of the Negligence Claim that are necessary for the Reporting Duty Claim should stand unaffected. It appears, however, that the significance of this common ground may not have been fully appreciated by the parties. The short point is that the Reporting Duty Claim, properly understood, is not a distinct cause of action from the Negligence Claim. This is apparent from the way in which HLT’s case in its Statement of Claim has been pleaded as well as the correct analysis of its claims in Suit 237 (see [93]–[99] below). This ties in with our first point above, which is that the Creditor Duty – much like the Reporting Duty – really does not have a separate existence outside of the Negligence Claim and the framework for negligence. Indeed, once that is fully appreciated, it becomes clear that the “Creditor Duty” and the “Reporting Duty” share the same substance. We propose to first explain how we reach this conclusion based solely on an examination of the pleadings, before going on to rationalise the position on principle. In the course of the latter discussion, we will also identify the difficulties in the approaches taken by the parties which have led to this somewhat unusual and unsatisfactory state of affairs. It will be seen that, once the parties’ cases are properly situated within the framework and elements of negligence, most of the confusion falls away (see [101]–[119] below).

93 To begin with, we find it rather odd that the parties have apparently faced no resistance thus far with proceeding on the basis that the Negligence Claim and the Reporting Duty Claim are distinct causes of action. Even a cursory review of the Statement of Claim contradicts this assumption as the Reporting Duty is pleaded as part of HLT’s duty of care, along with a slew of other duties:

94 The pleading of the Reporting Duty is at para 4(m) of the Statement of Claim. As the chapeau of para 4 makes clear, the Reporting Duty is part of Deloitte’s duty of care, which HLT advances both on a contractual – as an implied term under the engagement letters between the parties (at para 3) – and tortious basis (at para 5). Given this, it is not clear to us how the parties have come to accept that the Reporting Duty Claim and the Negligence Claim are distinct causes of action. The pleadings make clear that there is really only one cause of action – negligence – albeit it is put across in tort and contract. Since the Reporting Duty is a subset of HLT’s duty of care, any claim based on a breach of the Reporting Duty must necessarily be a claim for breach of a duty of care (ie, negligence).

95 The pleadings on the breach of the putative Reporting Duty tell a similar story, and indeed, make the point in even clearer terms:

96 In the above extract from HLT’s Statement of Claim, we have placed in emphasis the repeated pleading of the counterfactual of Deloitte having acted with reasonable skill and care in the conduct of the audit. The significance of this is that HLT’s pleaded case on the Reporting Duty Claim is that if Deloitte had not been negligent in the conduct of its audit, it would have uncovered the fraudulent or improper conduct of the Lim Family in their management of HLT’s affairs, and in turn reported what it had uncovered to third parties other than the Lim Family. In our view, this makes clear that the Reporting Duty Claim is not distinct but part of the Negligence Claim as it is premised on HLT successfully establishing that Deloitte had failed to act with reasonable skill and care – ie, Deloitte was negligent – in the conduct of the audit. We therefore find it difficult to conceive how the parties have proceeded on the basis that the Reporting Duty Claim and the Negligence Claim are distinct. In reality, the only cause of action that HLT is advancing is the Negligence Claim (albeit in contract and tort).

97 In fairness to Deloitte, it had highlighted the peculiarity of taking the Reporting Duty Claim as distinct from the Negligence Claim in its submissions below. Before the AR, Deloitte had submitted that the Reporting Duty Claim was unsustainable and indeed a logically inconsistent pleading with the Negligence Claim because the latter was founded on the premise that Deloitte did not know of the fraud and/or irregularities in HLT’s affairs. Deloitte suggested that the Reporting Duty Claim was illogical because it could not have failed to report matters which, on HLT’s own case, it did not know of (see [25] above). In our view, Deloitte was correct in making this observation, but it drew the wrong conclusion from this. What this demonstrated was that the Reporting Duty Claim was not a distinct cause of action but simply a part of the Negligence Claim.

98 Furthermore, as we have noted at [28] above, in its own submissions before the AR, HLT had itself stated that the Reporting Duty was “founded on a claim for breach of [a] duty of care”. That can only mean, as a matter of logic, that the Reporting Duty Claim is not distinct from the Negligence Claim since the cause of action for a breach of a duty of care is negligence. Indeed, at the hearing before the AR, HLT’s counsel once again confirmed that the “duty to report [was] a subspecies of a duty of care” and that “[t]he duty to report fraud and irregularities to the authorities ... [was] an aspect of the duty of care owed by auditors to companies”.

99 Coming back to the Creditor Duty Question, the issue before us, as framed by the parties, relates only to the Negligence Claim. Thus, in a series of letters between the parties that was put before the court in the lead-up to this appeal, the parties confirmed that the resolution of the Creditor Duty Question in Deloitte’s favour would result in the striking out of the Negligence Claim, but as both parties conceived of the Reporting Duty Claim as a distinct cause of action, it was also agreed that “a striking out of the Negligence Claim would not result in portions of HLT’s pleadings that remain relevant to other claims being struck out”. Given that the Reporting Duty Claim is part of the Negligence Claim for the reasons we have explained above, it follows that it is not possible for the resolution of the Creditor Duty Question to have any effect on HLT’s pleadings. No part of the Statement of Claim in relation to the Negligence Claim would be struck out since they would remain necessary for the Reporting Duty Claim. In the circumstances, it is pointless for us to address the Creditor Duty Question.

100 In any event, as we explain next, we do not think that the Creditor Duty Question raises an issue that is appropriate for summary determination. In this regard, the parties’ framing of the Creditor Duty Question stems from a failure to identify which element(s) of the negligence framework are implicated by their respective arguments. The proliferation of labels like “Negligence Claim”, “Reporting Duty Claim” and “Creditor Duty” have, in our view, caused the parties to miss the forest for the trees. We therefore propose to unpack the substance of the Creditor Duty and to explain a number of conceptual difficulties we have with the parties’ approaches to the matter, with a view to injecting clarity in the dispute in Suit 237 and providing some general guidance for future cases.

101 The starting point is that what has come to be referred to as the “Creditor Duty” is a misnomer. Before us, counsel for HLT, Mr Cavinder Bull SC (“Mr Bull”), confirmed that it was not HLT’s case that Deloitte owed any direct duty to HLT’s creditors. Indeed, no such suggestion could sensibly be made seeing as the claimant in Suit 237 is not any individual creditor of HLT but HLT itself. Given this, the reference to a “Creditor Duty” – or, to put it in the exact terms that HLT has framed it, a “duty to have regard to the interests of HLT’s creditors” – is unhelpful.

102 The genesis of the “Creditor Duty” is Deloitte’s submission that the fact that HLT’s shareholders and directors – the Lim Family – were not misled by any alleged negligence on Deloitte’s part and were aware of the true position at all material times provides Deloitte with a complete answer to HLT’s claims in Suit 237. In its submissions before us, Deloitte has christened this as the “No Reliance Principle”, another label which, with respect, is equally unhelpful. The Creditor Duty is HLT’s reply to the No Reliance Principle. It entails, in short, the submission that Deloitte’s duty of care is not limited to providing information to the Lim Family but may extend to third parties in certain circumstances. The use of all these different labels in the parties’ submissions, however, seems to have caused them to overlook how and where the No Reliance Principle – or, more helpfully, the fact of the Lim Family’s knowledge of the true state of affairs – fits within the negligence framework. But, in our judgment, the No Reliance Principle and the Creditor Duty feature at no less than three different elements of the negligence inquiry:

(a) Breach: First, to the extent that Deloitte submits that its duty of care did not require it to report information to persons other than the Lim Family, this is in substance a submission that the content of its duty of care, or the standard of care of an auditor acting with reasonable skill and care, did not require Deloitte to report information to third parties other than the Lim Family even if it had detected the fraud and irregularities such as the misstatements in HLT’s financial statements. This is an issue of whether Deloitte had acted in breach of its duty of care (see [112]–[118] below).

(b) Factual Causation: Second, the upshot of Deloitte’s submission that the Lim Family was not misled is that nothing would have changed even if Deloitte had not been negligent and had detected the fraud or irregularities (as HLT alleges) since, in that event, Deloitte would only have reported the same to the Lim Family, who already knew of these matters. The submission that “nothing would have changed” is an argument on factual causation, or more specifically, that Deloitte’s negligence was not a “but for” cause of HLT’s losses because the Lim Family (and, in turn, HLT) would not have acted any differently. This argument, however, stands and falls with Deloitte’s case on breach of duty above because, if the appropriate standard of care did require Deloitte to report information to persons other than the Lim Family, Deloitte would have done so if it had not been negligent and it cannot be said that the same events would have transpired (see [119] below).

(c) Defences: Third, the fact that all the directors and shareholders of HLT were aware of and complicit in the fraudulent conduct in HLT’s affairs may be significant in so far as the attribution of their knowledge and acts to HLT would mean that HLT itself is tainted with fraud, which would raise the question as to whether HLT’s claim in Suit 237 should be barred by illegality or the ex turpi causa principle (see [103]–[111] below).

103 We take things in the reverse order and begin with the issue of illegality, which was canvassed at some length at the hearing of the appeal. We had raised the issue of whether the illegality defence operated against HLT’s claim in Suit 237 based on attributing the Lim Family’s knowledge and conduct to HLT as this, in our view, would properly have raised a question of law appropriate for summary determination in this appeal: see, for example, United Overseas Bank Ltd v Lippo Marina Collection Pte Ltd [2016] 2 SLR 597 (summary determination of the issue of whether a claim by a company was barred by the illegality principle based on the attribution of its director’s fraud to it).

104 To elaborate, the facts of the present case seemed to us to be on all fours with those of the decision of the House of Lords in Stone & Rolls Ltd v Moore Stephens [2009] 1 AC 1391 (“Stone & Rolls”), in that this was a case where there were no “innocent” shareholders or directors of the company who were not complicit in the fraud. In Stone & Rolls, the liquidators of a company had brought a suit against the company’s former auditor for negligently failing to detect the fraud which the sole director and shareholder of the company had procured the company to commit on various banks. The auditor applied to have the claim struck out. In the House of Lords, it was resolved by a majority of three to two that the company’s claim should be struck out as it was barred by the illegality defence.

105 Stone & Rolls has proven to be a controversial case, as quite apart from the issue of whether it was correctly decided, the search for the precise basis of the majority’s decision itself has proven elusive. Indeed, the difficulty on this front is apparent from how, in the subsequent decision of the UK Supreme Court in Bilta (UK) Ltd v Nazir (No 2) [2016] AC 1 (“Bilta”), Lord Neuberger of Abbotsbury PSC commented that “it is very hard to seek to derive much in the way of reliable principle” from Stone & Rolls (at [21]–[23]), while Lord Toulson JSC and Lord Hodge JSC opined that Stone & Rolls should be regarded as a decision without a majority ratio decidendi and “authority for the point which it decided, namely that on the facts of that case no claim lay against the auditors, but nothing more” (at [154]). On the other hand, Lord Sumption JSC (who stood in the somewhat unique position of having acted as counsel for the successful auditor in Stone & Rolls) considered that the case was at least authority for the proposition that a third party who was not involved in the fraud or illegal conduct, such as an auditor, could rely on the illegality defence to defeat a claim by the company if there were no innocent shareholders or directors in the company (at [80]–[81]). If his Lordship was correct in identifying this as the ratio of Stone & Rolls, then there would be basis for Deloitte to submit, by analogy to the facts of Stone & Rolls, that HLT’s claim in Suit 237 should be struck out on the basis of identifying HLT with the fraud and illegal conduct of the Lim Family. In fact, there appears to be a pleading of this in Deloitte’s Defence in Suit 237:

106 Nevertheless, it was common ground before us at the hearing of the appeal that the question of whether the Lim Family’s knowledge and conduct should be attributed to HLT so that the illegality defence may operate to bar HLT’s claim is not an issue within the scope of this appeal. In light of this, bearing in mind that Deloitte does appear to have pleaded this defence and may rely on it if the matter proceeds to trial, we confine ourselves to only the following brief observations which tie in with what was traversed in the course of the parties’ submissions at the hearing.

107 The question of whether the fraudulent knowledge and conduct of a company’s directors should be attributed to the company in any given case is not a straightforward one. It seems well-established that attribution would not be appropriate in a claim by the company against a director who has acted fraudulently, essentially because it would make nonsense of the director’s duties to the company for him to be entitled to rely on his own fraudulent conduct to defeat the company’s claim: Ho Kang Peng v Scintronix Corp Ltd [2014] 3 SLR 329 (“Scintronix”) at [71]; Ong Bee Chew v Ong Shu Lin [2019] 3 SLR 132 at [140]–[172]; Bilta at [7].

108 But, outside of the context of directors, there does not seem to be a clear consensus. On the contrary, the only consensus seems to be that any question of attribution will depend on the specific context in which it is asked: Scintronix at [67]; Red Star Marine Consultants Pte Ltd v Personal Representatives of Satwant Kaur d/o Sardara Singh, deceased [2020] 1 SLR 115 (“Red Star”) at [37]–[41]. Where a defendant seeks to attribute the fraud or illegal conduct of a company’s directors to the company so as to invoke illegality as a defence to the company’s claim against him, the case law appears to consider if attribution would be appropriate from two dimensions:

(a) First, the courts have considered, from the perspective of the claimant company, if attributing the director’s wrong to the company would allow the wrongdoer to benefit from his own wrong. This was emphasised in the decision of this court in Red Star, where a claim was brought by a solvent company against an employee for misappropriating the company’s moneys in circumstances where the company’s primary director and shareholder, one Mr Singh, was aware of and had consented to the misappropriation. Although the company had another director and shareholder, Mr Singh’s wife, it was for all intents and purposes a “one-man company run and owned by Mr Singh” as Mr Singh held more than 99% of the company’s shares and was the only director involved in running it (at [43]). We held that the illegality defence should apply to bar the company’s claim as Mr Singh’s acts and knowledge should be attributed to it. As the company was solvent, allowing the claim would have meant that Mr Singh would be allowed to profit from his fraudulent conduct (at [47]).

(b) Second, the courts have also considered, from the defendant’s perspective, if attributing the director’s fraudulent acts and knowledge to the company would defeat the purpose of the defendant’s duty which the company is seeking to enforce or the cause of action which the company is pursuing. Claims by companies against errant directors, where attribution has been held to be inappropriate, fall within this category. In the same vein, in the decision of the UK Supreme Court in Singularis Holdings Ltd v Daiwa Capital Markets Europe Ltd [2020] AC 1189 (“Singularis”), it was held to be inappropriate to attribute the fraudulent conduct of the claimant’s director, who had misappropriated the claimant’s funds by issuing payment instructions to the defendant bank, to defeat the claimant’s claim against the defendant for breach of the Quincecare duty. Baroness Hale of Richmond PSC, delivering the judgment of the court, reasoned that since it was the purpose of the defendant’s duty to “protect the company against just the sort of misappropriation of its funds as took place here”, attributing the director’s fraud to the company would denude the Quincecare duty of any value as “there would in reality be no Quincecare duty of care or its breach would cease to have consequences” (at [35]).

109 In the present context of a claim by an insolvent company against its auditor for failing to detect the fraudulent or illegal conduct of its management, the first consideration above is likely to be inoperative. HLT’s insolvency in this case means that it is not likely that the Lim Family would benefit from their wrongful conduct if HLT’s claim against Deloitte were allowed to proceed as any recovery would inure to the benefit of HLT’s creditors.

110 The second consideration, however, is less straightforward. And it is here that arguments as to whose interests an auditor’s duty extends to protecting may become relevant. In this regard, it seems to us that the parties’ arguments over the Creditor Duty have their genesis in the reference to the “scope” of the auditor’s duty vis-à-vis creditors in some of the speeches in Stone & Rolls. In particular, whereas Lord Phillips of Worth Matravers in the majority was of the view that the scope of an auditor’s duty only extended to the protection of the company’s shareholders (at [18], [85] and [86]), which informed his conclusion that the auditor should be entitled to raise the defence of illegality against the company where the company’s sole director and shareholder had perpetrated fraud through the company, Lord Mance in the minority did not think that such a narrow view of the auditor’s duty was warranted (at [269]), which in turn led him to the opposite conclusion that the auditor could not attribute the very fraud that it had failed to detect to the company to defeat the company’s claim against it (at [275]).

111 In the present case, what the parties appear to have done is to take the discussion of the “scope” of the auditor’s duty – in terms of the identity of the stakeholders of a company whose interests an auditor has a responsibility to protect – out of the context of the issue arising in Stone & Rolls (the illegality defence), and to formulate, by extrapolation, the proposition that an auditor has a “duty to consider the interests of the company’s creditors” (ie, the Creditor Duty). But nothing in Stone & Rolls suggests that the law lords had in mind subjecting an auditor to an obligation to do any specific act. What their Lordships had in mind in their references to the “scope” of an auditor’s duty was the purpose of the duty: was the purpose of an auditor’s duty of care to protect only the interests of the shareholders, such that it would defeat the purpose of their duty if they were allowed to attribute fraud which they had failed to detect to the company where the shareholders were aware of the true position? It is one thing to say that an auditor’s duty normatively includes the protection of creditors’ interests while the company is insolvent, so that the auditor should not be allowed to escape liability when a suit is brought against it at the instance of the liquidators of the insolvent company, but quite another to say that an auditor has a positive duty to take certain action or to give some sort of subjective consideration to the interests of creditors. It is not clear that the members of the House of Lords in Stone & Rolls ever contemplated that an auditor could incur liability for failing to consider the interests of creditors in the same way that a director might for causing the company to undertake certain activity that is prejudicial to its creditors. Indeed, as we explain next, it is difficult to see what meaningful content a duty to “have regard to the interests of creditors” has in the context of auditors, and it is unclear if any meaningful analogy may lie as between directors and auditors so that the “Creditor Duty” as framed in the former context can be transposed to the latter.

112 This segues into our next point, which is that since it is clear from the case presented on appeal that HLT’s contention on the Creditor Duty ultimately boils down to Deloitte having acted or failed to act in a particular way, that is nothing more than a question of whether Deloitte’s conduct in the circumstances fell below the standard of care of a reasonably skilled and competent auditor and, in turn, whether Deloitte had breached its duty of care. The Creditor Duty is a wholly unnecessary gloss on the issue.

113 As mentioned at [111] above, while HLT has relied on an analogy to the context of directors’ duties, it is far from intuitive what a “Creditor Duty” may entail in the context of an auditor. In the context of directors, the Creditor Duty refers to the need for a director, as a fiduciary of the company, to exercise his powers and discretion having regard to the interests of the company’s creditors when the company is insolvent or in a financially parlous position. It is thus a constraint on the director’s powers qua fiduciary. As we explained in Foo Kian Beng, the rationale of the Creditor Duty is that when a company is insolvent or near to insolvency, the creditors are the main economic stakeholders of the company, and a director, as a custodial fiduciary having stewardship of what is effectively the creditors’ money, must “take corporate decisions with the interests of creditors in mind” [emphasis added] (at [72]).

114 Auditors, however, play a different role to that of directors. They do not generally have any involvement in the company’s decision-making, and do not exercise any discretion or authority delegated to them by the company as to how its affairs should be run. Given this, the formula that an auditor like Deloitte has a duty to “have regard to the interests of creditors” is unintelligible.

115 For the Creditor Duty to begin to have any content, it would be necessary for HLT to tie it to Deloitte’s doing of or failure to do certain acts. In that event, it would suffice for HLT to plead those acts or omissions. The question then would simply be whether Deloitte had acted in breach of its duty of care in the sense that its acts or omissions were out of accord with the standard of an auditor acting with due skill and care. The reference to a “Creditor Duty” adds nothing to the analysis. Indeed, the added baggage of the meaning it holds in the context of directors, as a facet of a director’s fiduciary duty to act in the best interests of the company, may even create confusion inasmuch as it can give the misleading impression that the touchstone of an auditor’s liability is not a failure to exercise reasonable skill and care (the fault element of negligence) but disloyalty (the fault element in fiduciary obligations) (see [79] above).

116 To develop the point further, it is not HLT’s case that an auditor like Deloitte has a duty to give some sort of subjective consideration to the interests of the company’s creditors when the company is insolvent. That, however, is the import of the “Creditor Duty” in the context of directors, as the fiduciary duty of directors to act bona fide in the best interests of the company “focuses on the subjective intentions of a director in committing a company to a certain course of action”: Foo Kian Beng at [77]. By contrast, the substance of the Creditor Duty as framed by HLT in the present context of auditors is that, when the audited company is insolvent, reasonable skill and care requires an auditor to detect and report irregularities in the company’s affairs to persons other than the company’s shareholders as the shareholders are no longer the main economic stakeholders of the company when it is insolvent. This is an objective yardstick. Thus, while “it is theoretically possible … for a director to have honestly believed that he had acted in the best interests of the company … even though his actions are found to fall below the objective standard of care and diligence expected of a director” (Foo Kian Beng at [77]), an auditor’s subjective intentions and honesty do not offer any defence to a claim that it has fallen short of the requisite standard. The label “Creditor Duty” and the analogy to directors is inapposite in the context of auditors and, with respect, has confused the real issues more than it has illuminated.

117 This leads us to a more general point, which is to disentangle what the parties have referred to distinctly as the “Negligence Claim”, “Reporting Duty Claim” and the “Creditor Duty”. The short point is that the Reporting Duty and Creditor Duty are not distinct duties owed by Deloitte to HLT but arguments as to the breach of Deloitte’s duty of care. Indeed, to the extent that the Creditor Duty is HLT’s response to the No Reliance Principle, it is of identical substance to the Reporting Duty in so far as it is a submission by HLT that a reasonably skilled and competent auditor in Deloitte’s position would have reported matters to persons other than the Lim Family, such that Deloitte’s failure to do so constitutes a breach of its duty of care. The common denominator between the Negligence Claim, Reporting Duty Claim and the Creditor Duty is thus HLT’s duty of care. That makes clear that the parties’ dispute is not over whether Deloitte owed HLT a duty of care – as mentioned at [48] above, it is quite unarguable that an auditor does owe an audited company a duty of care – but whether Deloitte had acted in breach of its duty of care. The issue of breach is a question of fact (see [77]–[78] above). Indeed, in JSI Shipping, this court confirmed that the standard of care of an auditor may be the subject of expert evidence on the practices of the audit profession, in which case the issue of whether the auditor had breached its duty of care would be assessed based on the principles laid down in Bolam v Friern Hospital Management Committee [1957] 1 WLR 582 and Bolitho v City and Hackney Health Authority [1998] AC 232, namely whether the auditor’s conduct was supported by a responsible body of opinion within the audit profession (at [50]). Such issues are self-evidently matters of fact which cannot be determined on a summary basis in the present appeal.

118 Taking a step back, the parties have fallen into the error we have highlighted at [72]–[79] above of conflating the duty and breach inquiries of the negligence framework. Instead of directing their minds to the question of whether Deloitte had acted in breach of its duty of care in doing or not doing certain acts, or failing to report information to third parties other than the Lim Family, the dispute has morphed into one of whether Deloitte owed HLT one or more specific duties to do certain acts (ie, the Creditor Duty and the Reporting Duty). This in turn has led to the parties losing sight of the fact that the question of whether Deloitte should or should not have done those acts depends on the standard of an auditor acting with reasonable skill and care. Parties would do well to recall this court’s caution in Go Dante Yap against the formulation of an alleged “tortious duty to advise”, which is equally applicable in the present case vis-à-vis the Reporting Duty and the Creditor Duty (see Go Dante Yap at [27]):

119 Finally, as to the dimension of causation, if the standard of a reasonably skilled and competent auditor would have required Deloitte to report information to persons other than the Lim Family, that would not only mean that Deloitte had breached its duty of care in failing to do so, but would also mean that factual causation between Deloitte’s breach of duty and HLT’s losses would be established. This is because, as mentioned at [102(b)] above, the No Reliance Principle formulated by Deloitte seeks to negate causation between any alleged breach of duty by Deloitte and HLT’s loss on the basis that the no-breach position is that Deloitte would have informed the Lim Family of the fraud or irregularities in HLT’s affairs that it may have uncovered, so there is no difference between HLT’s (current) breach position and its (counterfactual) no-breach position since the Lim Family already knew of the true position. If, however, Deloitte’s duty of care did not only require it to detect the fraud or irregularities but, as HLT argues through the Reporting Duty and Creditor Duty, to also report such matters to government authorities or third parties, there would be no equivalence between the breach and no-breach positions as Deloitte contends. Deloitte’s breach of duty would thus be a factual “but for” cause of HLT’s losses, and the question that would then arise is the extent of HLT’s losses that Deloitte should be legally responsible for. The Trading Losses Question, which we now turn to, is a question of this kind in so far as it focuses on a particular head of loss – ie, the Trading Losses – which HLT seeks to visit on Deloitte.

120 We begin our analysis of the Trading Losses Question by observing that it is unclear what the proper characterisation of the issue is in terms of which stage(s) of the negligence inquiry it fits within. This is important because, even though Deloitte has been able to identify a steady body of authority that supports its position that a company’s losses from continued trading are typically not recoverable from a negligent auditor, there is little consistency in terms of the reasoning to this conclusion.

121 Based on a review of the case law cited by the parties, one can identify two main bases on which the courts have rationalised absolving an auditor from liability for losses suffered by a company from continued trading which, but for the auditor’s negligence, would not have occurred as the company would have ceased trading and entered insolvency proceedings sooner:

(a) Legal Causation: First, an approach that relies on the notion of legal causation as the reason why a negligent auditor should not be held liable for trading losses. This appears to be the way in which the AR and the Judge saw the issue (see [34] above).

(b) The SAAMCo Principle: Second, an approach that relies on the notion of the scope of a duty of care and rules out recovery of trading losses on the basis that they fall outside the scope of the auditor’s duty. The primary authority for this is SAAMCo, which, while not itself a case of auditor liability, has been applied to the context of auditors in more recent cases.

122 In our judgment, neither of these represents an entirely satisfactory approach. We explain our concerns with each below. It would be recalled, however, that we have already expressed certain general reservations with the second approach in that defining a duty of care with reference to particular damage (or risk of such damage) entails an unnecessary conflation of the issues of duty of care and remoteness of damage (see [80]–[89] above). We elaborate on those points with reference to the SAAMCo principle specifically which, as prefaced above, we consider can be rationalised through orthodox principles of remoteness of damage in contract. It follows that we view the Trading Losses Question as a question of the appropriate scope of Deloitte’s legal responsibility in terms of whether the loss being claimed – ie, HLT’s Trading Losses – is too remote.

123 The causation-based approach is exemplified by the decisions of the New South Wales Court of Appeal in Alexander v Cambridge Credit Corporation Ltd (1987) 12 ACLR 202 (“Alexander”) and the English Court of Appeal in Galoo Ltd v Bright Grahame Murray [1994] 1 WLR 1360 (“Galoo”). The facts of both cases were materially on all fours with each other. In both cases, the courts held that the defendant auditors were not liable for the losses suffered by the audited companies that continued trading and incurring losses as the losses were not caused by the auditors’ negligent failure to detect and report material misstatements in the companies’ accounts.

124 In Alexander, the defendant auditor negligently failed to make provision for certain doubtful investments in the company’s accounts for the year ending 30 June 1971. The company continued trading for some three years before it was later put into receivership in September 1974. The company claimed damages of $145m from the defendant, which sum represented the difference between its net asset deficit in September 1974 and September 1971 (when it would have entered receivership if the defendant had not been negligent).

125 The New South Wales Court of Appeal held, by a majority of two to one, that the company was not entitled to recover such loss from its auditor. The majority, comprising Mahoney JA and McHugh JA (as he then was), held that the loss suffered by the company in the deterioration of its net asset position between 1971 and 1974 was not caused by the auditor’s negligence. Mahoney JA opined that even though the company’s loss was “[i]n the broadest sense … a result of the [auditor’s] breach”, that did not suffice to make the auditor liable (at 225). While the auditor’s negligence had allowed the company to continue existing until 1974, this only meant that the company was “expose[d] … to all of the dangers of being in existence”, and “allowing the company to remain in existence [did] not, without more, cause losses from anything which [was] … a danger incident to existing” (at 226). McHugh JA, on the other hand, emphasised that the question of causation of loss was one of “practical commonsense” (at 241). In this regard, the learned judge held that to hold the auditor liable for the company’s loss would “depart from the commonsense notion of causation which the common law champions” (at 248).

126 McHugh JA’s reasoning was followed in Galoo. The defendant auditor in Galoo negligently failed to detect that, in each of the years from 1985 to 1990, the stock held by the company had been overstated in its annual accounts. The company alleged that it would have been put into liquidation by mid-1986 if the fraud had been discovered in 1985. As a result, the company sought to recover trading losses of £25m, which amount represented the difference between its net liabilities at the time it entered liquidation (£27m) and its net liabilities in mid-1986 (£2m).

127 The English Court of Appeal held that the company’s claim for trading losses should be struck out on the basis of an insufficient causal link to the auditor’s negligence. It was not enough that, but for the auditor’s negligence, the company would have ceased trading earlier. Glidewell LJ, delivering the judgment of the court, distinguished a breach of duty that “causes a loss to the plaintiff” from “one which merely gives the opportunity for him to sustain the loss”. Like McHugh JA did in Alexander, Glidewell LJ emphasised that which category a given case fell into depended on “the application of the court’s common sense”. Applying this test, his Lordship held that the auditor’s negligence had not caused the company’s losses “in the sense in which the word ‘cause’ is used in law” (at 1374H–1375B).

128 In our view, the reasoning in Galoo and Alexander, which rests on allusions to “common sense”, is not satisfactory for at least three reasons. The first is analytical uncertainty. Professor Stapleton has commented, rightly in our view, that the “common sense” approach “is so indeterminate that it is effectively worthless as an analytical guide for lawyers confronting the facts of a new situation”: Jane Stapleton, “Reflections on Common Sense Causation in Australia” in Torts in Commercial Law (Simone Degeling, James Edelman & James Goudkamp eds) (Thomson Reuters, 2011) at p 350. Indeed, Alexander itself demonstrates this. While McHugh JA asserted that his conclusion that the auditor’s negligence had not caused the company’s loss was based on common sense, the dissenting judge, Glass JA, thought that common sense pointed in the opposite direction (at 209).

129 The second reason is the lack of transparency in reasoning. As Justice James Edelman has observed extrajudicially, “common sense” reasoning “invites the judge … to reason by reference to unstated premises and to assert as ‘common’ a conclusion that is often highly contested”: James Edelman, “Unnecessary Causation” (2015) 89 ALJ 20 at 20. In the same vein, Lord Hoffmann – who as will be seen below, was the modern architect of the second approach and the SAAMCo principle – commented that characterising an issue as one of “common sense” was among “the most common judicial expressions which conceal, or perhaps reveal, a complete absence of any form of reasoning”: Lord Hoffmann, “Common Sense and Causing Loss”, Lecture to the Chancery Bar Association (15 June 1999) at p 1.

130 The final reason, which we elaborate on below, is that “common sense” causal reasoning in cases like Alexander and Galoo arguably involves a category error as it conflates a question of causation with a question of legal responsibility. The language of “causation” was used by the courts in Alexander and Galoo as a means of stating a conclusion on the appropriateness of the auditors being legally responsible for the companies’ trading losses, rather than whether the companies’ trading losses had been caused by the auditors’ negligence in the sense that the losses would not have occurred if the auditors had not been negligent. In this regard, we have suggested at [61] above that the language of “legal causation” should be reserved to cases where the question is one of attribution of responsibility as between two or more causes of the claimant’s damage such that the defendant’s wrongful conduct is not seen as the “legal cause” of the claimant’s damage due to the existence of some other factor – a novus actus interveniens – that severs the chain of “causation” (or, more accurately, legal responsibility). Absent a competing cause to the defendant’s wrongdoing, it is difficult to assess in the abstract if the defendant’s breach is a “legal” or “proximate” cause of the claimant’s damage.

131 In Alexander and Galoo, it does not appear that the courts identified any cause other than the auditors’ negligence for the companies’ trading losses that could have constituted a novus actus interveniens. Instead, the courts asserted, without any real elaboration or explanation, that the auditors’ negligence did not “legally cause” the companies’ losses based on “common sense”. But since the companies’ losses must obviously have been caused by something, to say that the auditors’ negligence was not the “legal cause” of the companies’ losses begs the question what other factor could have constituted a novus actus interveniens so that it, rather than the auditors’ negligence, was the “legal cause” of the losses. Similarly, in this case, while Deloitte has argued that HLT’s claim for its Trading Losses should “fail for want of legal causation”, no alternative cause other than the alleged negligence of Deloitte that could constitute an intervening cause has been pleaded.

132 Indeed, the flaws with the “common sense causation” approach are laid bare by how McHugh JA’s commitment to it waned not long after Alexander was decided. Less than five years later, in the decision of the High Court of Australia in March v E & M H Stramare Pty Ltd (1991) 171 CLR 506 (“March”), in stark contrast to the rest of the court who affirmed the “common sense” approach, McHugh J commented that, upon reflection, his reliance on “common sense” in Alexander was unsatisfactory as it had obscured the reality that the court was making a value judgment based on policy as to the extent of a defendant’s responsibility for the consequences of his wrongful conduct and risked devolving into an “unfettered discretion” (at 530–531):

133 McHugh J’s change in approach between Alexander and March mirrors the paradigm shift in more recent decisions of the English courts that have preferred to rationalise the exclusion of certain losses from the scope of the defendant’s legal responsibility by defining the scope of the defendant’s duty as excluding such losses. As noted above, this is often referred to as the “SAAMCo principle” based on the eponymous decision of the House of Lords, which we take as a convenient starting point.

134 The facts of SAAMCo are somewhat complicated and involved three different appeals, but they can be stated in simplified terms for present purposes. The case, in broad overview, concerned the extent of the liability of valuers who had negligently overvalued properties over which lenders had taken security for loans that they had extended. It was undisputed that, if the valuers had not been negligent, none of the loans would have been made. By the time that the borrowers defaulted on the loans, the value of the properties had fallen sharply due to a fall in the property market, which greatly increased the losses suffered by the lenders. The lenders sued the valuers for negligence in both contract and tort. The question before the House of Lords was whether the valuers were liable for the entire loss suffered by the lenders from entering into the transactions or a more limited loss owing to their limited role in the lenders’ decision to make the loans.

135 The House of Lords held that the valuers were not liable for the entire loss suffered by the lenders from making the loans, but only the more limited loss of the extent of the overvaluation. The sole judgment was delivered by Lord Hoffmann, with whom the other members of the House agreed.

136 Lord Hoffmann began by observing that it was not correct to approach the issue as concerning the “correct measure of damages for the loss which the lender ha[d] suffered” (at 210G–211A). This was “the wrong place to begin”, because “[b]efore one can consider the principle on which one should calculate the damages to which a plaintiff is entitled as compensation for loss, it is necessary to decide for what kind of loss he is entitled to compensation” (at 211A–B). In this connection, his Lordship referred to Lord Bridge’s statement in Caparo (at 627D) – which we have earlier considered above – that (at 211G–212C):

Given that there was no dispute that the valuers owed a duty of care to the lenders (in contract and in tort), the “real question” was “the kind of loss in respect of which the duty was owed” (at 212C).

137 Lord Hoffmann went on to explain that the inquiry into the scope of the defendant’s duty would “depend upon the purpose of the rule imposing the duty” (at 212D). Applying this to the context of valuers, his Lordship said that (at 212E–F):

It is useful to emphasise two points from this passage. The first is that, when Lord Hoffmann spoke of the “scope of the duty”, his Lordship was not referring to the valuer’s primary obligation – ie, what it was required to do under the contract – but the valuer’s secondary obligation to pay damages as reparation for the consequences of the breach of its primary obligation. This is clear from how his Lordship defined the “scope of the duty” in terms of “the consequences for which the valuer is responsible”. Indeed, in a subsequent essay, Lord Hoffmann conceded that his language of “scope of duty” may have been infelicitous, and that he would “try to mend [his] language in future”, as “[t]he scope of the duty of care is to take reasonable care to get the valuation right” and “has nothing to do with the extent of the consequences for which the valuer is liable”: Leonard Hoffmann, “Causation” (2005) 121 LQR 592 at 596. The second point is that Lord Hoffmann saw the extent of the valuer’s responsibility for the consequences of its breach of its primary obligation to be a matter of giving effect to the parties’ expectations under the contract. This can be seen from his Lordship’s statement that an appropriate balance had to be struck between not giving the lender “less than he was reasonably entitled to expect” and not imposing on the valuer “a liability greater than he could reasonably have thought he was undertaking”.

138 As to the question of what liability the defendant “could reasonably have thought he was undertaking”, Lord Hoffmann’s answer was that this was not “the whole risk of consequences which would not have happened but for the wrongful act”, but limited to “those consequences which [were] attributable to that which made the act wrongful” (at 212F–213D). His Lordship illustrated this distinction using what has become a celebrated parable of a doctor who advises a mountaineer prior to the latter embarking on an expedition (at 213D–E):

According to Lord Hoffmann, the reason for the intuitive result that the doctor should not be liable, even though the mountaineer would not have suffered the injury if the doctor had not been negligent, was that “[t]he injury ha[d] not been caused by the doctor’s bad advice because it would have occurred even if the advice had been correct” (at 213F).

139 Based on the summary above, the decision in SAAMCo can therefore be summarised in the following three propositions:

(a) First, the defendant is not liable for all of the damage that the claimant has suffered which would not have occurred but for the defendant’s wrongful conduct. Instead, it is necessary to identify the consequences that fall within the defendant’s “scope of duty” or, more accurately (as Lord Hoffmann later acknowledged), the proper scope of the defendant’s legal responsibility.

(b) Second, the scope of the defendant’s legal responsibility depends on the purpose of the defendant’s duty. The law must strike a balance between not giving the claimant less than he was reasonably entitled to expect and not imposing on the defendant an extent of liability greater than what he could reasonably have thought he was undertaking.

(c) Third, the consequences that the defendant would reasonably have thought he was undertaking would be those consequences which are the result of what made his act wrongful and not the entirety of the consequences flowing from his wrongful conduct. In particular, losses which would have been suffered even if the defendant’s conduct had not been wrongful would fall outside the scope of the defendant’s legal responsibility.

140 It is fair to say that the English courts’ experience with SAAMCo since it was decided has not been a happy one. In Hughes-Holland v BPE Solicitors [2018] AC 599, Lord Sumption JSC commented that the problem posed by Lord Hoffmann’s mountaineer example was “one of the main dilemmas of the law of damages” (at [1]). This culminated in the decision of the UK Supreme Court in Manchester Building Society, where the court sought to undertake a root-and-branch re-examination of the SAAMCo principle. As we have already examined the decision in SAAMCo itself in detail, we do not think it necessary to consider Manchester Building Society at any length. It suffices to say that the difficulty in identifying the basis of the SAAMCo principle continues to persist in so far as one can discern two distinct approaches from the three different judgments in Manchester Building Society: (a) first, an approach preferred by the majority (Lord Hodge DPSC and Lord Sales JSC, Lord Reed PSC, Lord Kitchin JSC and Lady Black agreeing) and Lord Burrows JSC, which focuses on the purpose of the defendant’s duty of care and seeks to limit the defendant’s liability to damage that is the fruition of risks which it was the purpose of the defendant’s duty to guard against (see [5], [13] and [17] per the majority, and [179], [203] and [212] per Lord Burrows JSC); and (b) second, an approach adopted by Lord Leggatt JSC which sees the issue as one of causation between the claimant’s loss and the subject matter of the defendant’s duty (at [95]–[99]).

141 The precise bounds of the SAAMCo principle are also unclear in terms of when it would apply. In Meadows and Manchester Building Society, while the UK Supreme Court seemed to conceive of it as a general principle that was inherent in the structure of the tort of negligence, all the members of the court also appeared to think that it was either confined to or most appropriate in the specific context of professional negligence: Manchester Building Society at [1]–[2] per Lord Hodge DPSC and Lord Sales JSC, at [107] per Lord Leggatt JSC and at [179] per Lord Burrows JSC. This, however, begs the question as to why there may be a specific principle that is applicable only in the context of a particular type of negligence claim. Absent a justification, limiting the SAAMCo principle to the context of professional negligence – leaving aside that what constitutes a “professional” is, in the first place, “not susceptible to easy definition” (see Jackson & Powell on Professional Liability (Mark Cannon, Hugh Evans & Roger Stewart gen eds) (Sweet & Maxwell, 9th Ed, 2021) at para 1-002) – seems rather arbitrary.

142 In our view, the problem which the SAAMCo principle seeks to resolve is the intuitive unfairness of visiting on the defendant all of the consequences that have resulted from his wrongful conduct. That is where Lord Hoffmann started from in SAAMCo, and what he had in mind when he drew a distinction between “the whole risk of consequences which would not have happened but for the wrongful act” and “those consequences which are attributable to that which made the act wrongful”. This is the same point made by: (a) the majority in Manchester Building Society, who considered that the defendant’s liability should be limited to loss that represented the fruition of risks which it was the defendant’s duty to guard against (at [17]); (b) Lord Leggatt JSC, who considered the “underlying policy rationale” of the SAAMCo principle to be the allocation of responsibility for loss suffered by the claimant “in a way which fairly reflects the assumption of risk implicit in the service which the [defendant] agreed to provide” (at [88]); and (c) Lord Burrows JSC, who said that the SAAMCo principle was “underpinned by the policy of achieving a fair and reasonable allocation of the risk of the loss that has occurred” (at [201]).

143 The primary source of difficulty, in our view, is that Lord Hoffmann’s reasoning in SAAMCo was not entirely precise in its focus. At one point, his Lordship reasoned that the courts should give effect to the parties’ expectations as to the allocation of risk of damage. But, in the course of his analysis, the focus shifted to causal reasoning as can be seen in the mountaineer example and the use of a counterfactual to identify those losses which are not “caused” by the defendant’s negligence.

144 In our view, the focus should be on the former. The SAAMCo principle, as we see it, has nothing to do with causation. To hark back to the mountaineer example which Lord Hoffmann used, his Lordship’s reasoning was that the mountaineer’s injury had not been caused by the doctor’s negligence. With respect, we do not see how that can be correct. Once it is accepted that the mountaineer would not have gone on the expedition and therefore would not have suffered any injury if the doctor had not been negligent, it is plain that the doctor’s breach of duty had caused the injury. The difficulty in his Lordship’s analysis is apparent from how the counterfactual which he employs to disprove the causal relevance of the doctor’s negligence is not one in which the doctor had given non-negligent advice (“what if the defendant had not acted in breach of duty”), but one where the doctor’s negligent advice is deemed to have been non-negligent (“what if what the defendant did was not a breach of duty”). We do not necessarily think that his Lordship’s reasoning was wrong, as we explain below, but what we do not agree with is that it is an exercise in causation.

145 What Lord Hoffmann’s counterfactual demonstrates, in our view, is not the causal relevance or irrelevance of the defendant’s breach of duty to the claimant’s damage, but that it is necessary to identify the scope of the risk that the defendant has assumed. We would explain it in this way. The starting point is that, when a person (“A”) undertakes any conduct, the risks of damage arising from such conduct rest with A. But a relationship between A and another person (“B”) can result in the transfer of some of the risks to B. The question, then, is the extent of the risk of damage to A which B has undertaken. Lord Hoffmann’s counterfactual seeks to filter out the risks of damage which B does not undertake by reference to the task that B performs. To situate this in the mountaineer example, if A embarks on a mountaineering expedition, A will, as a starting point, bear the entirety of the risk of damage arising from the activity. But, when A goes to meet B, a doctor, for advice on the fitness of his knee before going on the expedition, A is seeking to transfer onto B the risk of damage arising from the fitness of his knee. Since A has not asked B about the risks of mountaineering generally, and B has made no representation as to such risks, those risks remain with A. As Lord Hoffmann put it, there is “no reason of policy which requires that the negligence of the doctor should require the transfer to him of all the foreseeable risks of the expedition”: SAAMCo at 214A–C.

146 It is apparent that the concept of assumption of responsibility is implicit in the SAAMCo principle as the focus is on whether B assumed responsibility in respect of certain risks of damage to A. Where and how does this fit in the structure of negligence liability? One way would be to define B’s duty of care with reference to the specific damage, in the way Lord Bridge did in Caparo, and ask if B owed a duty with respect to the risk of this specific kind of damage. Lord Hoffmann appears to have been of the same mind as he cited Lord Bridge’s statement in Caparo with approval and criticised the English Court of Appeal for using the wrong starting point of imposing on the valuers liability for all the losses the lenders had suffered (see [136] above). But, as we have explained at [81] above, this unnecessarily rolls up the issue of remoteness of specific kinds of damage into the question of duty and is also not consistent with the way in which our courts have conceived of the duty of care following Spandeck: see, for a detailed analysis that makes essentially the same point in the New Zealand context, Routhan v PGG Wrightson Real Estate Ltd [2025] 1 NZLR 306 (“Routhan”) at [239]–[277] per Kós J.

147 That then leaves the question of remoteness. In this regard, we consider that a clear link can be drawn between the SAAMCo principle and the rules for remoteness of damage in contract law, which are the default rules that the law prescribes for ascertaining a fair and reasonable allocation of risk of damage as between the parties. Indeed, it will be recalled that, while SAAMCo involved a claim for negligence, it was founded on a contractual duty of care arising in the contract between the lender and valuer (see [134] above). And in what we regard to be a crucial part of his reasoning in SAAMCo, Lord Hoffmann expressly defined the scope of the defendant’s responsibility as “that which the law regards as best giving effect to the express obligations assumed by the valuer” based on the reasonable expectations of the parties (at 212E–F):

148 As mentioned at [137] above, the essential point made in this passage is that the scope of the primary obligation undertaken by the defendant under the contract defines the scope of the secondary obligation imposed by law on the defendant to pay damages as the next-best alternative in the event that the primary obligation is breached. Thus, since the doctor’s primary obligation was to guard against the risk of the mountaineer suffering injury from his knee, the secondary obligation to pay damages would extend to making good losses arising from the unfitness of his knee, but not other losses that arise from the general perils of mountaineering. Losses of the latter kind, as Lord Hoffmann put it, would constitute “a liability greater than [the doctor] could reasonably have thought he was undertaking”, seeing as he was never asked to opine on the safety of mountaineering generally.

149 This can be explained in terms of the rules for remoteness of damage in contract law established in Hadley v Baxendale, which this court has accepted embodies the concept of assumption of responsibility: MFM Restaurants Pte Ltd v Fish & Co Restaurants Pte Ltd [2011] 1 SLR 150 (“MFM Restaurants”) at [101]–[107]. Specifically, the Hadley v Baxendale test uses the contract-breaker’s knowledge at the time of contracting as a basis for imputing to him an assumption of responsibility for loss arising from his breach of contract: Out of the Box Pte Ltd v Wanin Industries Pte Ltd [2013] 2 SLR 363 (“Out of the Box”) at [15]–[22]. As the learned authors of The Law of Contract in Singapore (Andrew Phang Boon Leong gen ed) (Academy Publishing, 2nd Ed, 2022) have put the point (at para 22.076):

Given that the parties are likely to have contemplated performance rather than breach, it is unlikely that they would have actually considered how the risk of losses from a breach of contract should be allocated between them. In these premises, the Hadley v Baxendale test is “the best possible reflection of what the parties would have agreed to had they thought about a situation in which the contract was breached” [emphasis in original]: MFM Restaurants at [109].

150 Situating the SAAMCo principle within the contractual remoteness rules has the advantage of certainty. As noted at [141] above, English law appears to have reached a tentative consensus that the principle may be confined to, or at least most relevant in, the professional negligence context, but no justification has been identified for this other than this having been the trend in the case law. A likely explanation, in our view, is that most professional negligence claims are not brought as pure tort claims but as claims for breach of a contractual duty of care, as the contractual matrix would mean that the facts would line up with an analysis based on assumption of responsibility. It might be that the SAAMCo principle could have a peripheral relevance in cases where there is strictly no contract but the parties’ relationship can be considered “akin to contract” – the result being either the displacement of the Wagon Mound test by the Hadley v Baxendale test or the application of the former in a way that is consistent with the element of assumption of responsibility (which of these, it may not matter, as the result is likely to be the same). But it is not necessary for us to express any firm or conclusive view here on non-contractual cases. It suffices to say that, in a case like the present where the duty of care arises both in tort and contract, the Hadley v Baxendale test fulfils the function that the SAAMCo principle has been set up to address: Routhan at [161].

151 Reframing the SAAMCo principle in terms of a well-established rule like the Hadley v Baxendale test will also clear up uncertainties as to how it should be applied. It would be recalled that the SAAMCo principle seeks to define the scope of the defendant’s assumption of responsibility for loss caused by his breach of duty by reference to the purpose of the duty. As our analysis of the facts of the present case will show, there is no need for this to be considered as a separate step in itself since the purpose of the defendant’s duty is a factor that can be taken into account in the application of the Hadley v Baxendale test, along with other facts that would be in the defendant’s horizon of contemplation at the time of contracting, for the purpose of determining what extent of loss the defendant may fairly be said to have assumed responsibility for.

152 In our opinion, the view that it is necessary for the SAAMCo principle to exist separately outside the traditional rules for remoteness of damage may be engendered by the perception that the Hadley v Baxendale test is purely a test of objective probability. But as we previously explained in Out of the Box, in discussing the controversial decision of the House of Lords in Transfield Shipping Inc v Mercator Shipping Inc [2009] 1 AC 61 (“The Achilleas”), that perception is misconceived because it is possible for objectively probable loss to be too remote (see, for example, The Achilleas itself as well as the mountaineer example in SAAMCo); and equally, for objectively improbable loss to be not too remote (see, for example, The Heron II at 416–417). Indeed, it is for this reason that this court thought it unnecessary in MFM Restaurants and Out of the Box to adopt Lord Hoffmann’s reformulation of the contractual remoteness rule in the explicit terms of assumption of responsibility as set out in The Achilleas. The test in Hadley v Baxendale, if correctly applied, already achieves that end. In this regard, while a test of objective probability may be a rough approximation of whether loss is too remote, it is not itself the true test of remoteness. There is no difficulty in subsuming the SAAMCo principle into the Hadley v Baxendale test if this is borne in mind.

153 Based on the discussion above, the Trading Losses Question raises an issue of remoteness of damage: are HLT’s Trading Losses too remote to Deloitte’s breach of duty so as to be irrecoverable? As we have explained, this turns on an application of the rule in Hadley v Baxendale. While remoteness is inherently fact-sensitive, the factual uncertainty in this case is obviated by Deloitte’s agreement to have assumed as true all of the facts pleaded by HLT against it.

154 In our judgment, the issue is a relatively straightforward one. It will be recalled that the Trading Losses are losses incurred by HLT from its trading activities in the period between November 2015 and mid-April 2020. This period generally covers the audits conducted by Deloitte between FY2014 and FY2019. The claim is that, if not for Deloitte’s negligence in the carrying out of its audits during this period, HLT would have been put into liquidation at some earlier point in time and hence would not have continued to incur further losses that deepened its insolvency. In our judgment, there is no doubt that the Trading Losses are too remote, for the following two reasons.

155 First, Deloitte did not have any involvement in HLT’s trading activities. It did not have any sight over what trading strategies HLT employed or give any input to the Lim Family as to whether certain trading methodologies were advantageous or otherwise. This gaping hole in Deloitte’s knowledge makes it fanciful for HLT to suggest that it could have been in Deloitte’s reasonable contemplation that it had essentially signed up to insure HLT’s trading fortunes. It is inconceivable that an auditor who does nothing more than perform a statutory audit could be taken to have assumed liability for such losses since their occurrence depends on movements in the market and the decisions of the company’s management which the auditor has no control over or involvement in.

156 This, in our view, is the primary reason underlying the body of authority that Deloitte has cited in which courts have been unwilling to impose on auditors responsibility for trading losses suffered by their clients. This can be seen from the English High Court decision of Bank of Credit and Commerce International (Overseas) Ltd v Price Waterhouse [1999] BCC 351, which counsel for Deloitte, Ms Wendy Lin (“Ms Lin”), referred us to at the hearing. In that case, in striking out a claim against an auditor in circumstances not dissimilar to the present, Laddie J said (at [57]):

157 We agree. What would have been in Deloitte’s contemplation, if it had applied its mind to the consequences it would be liable for in the event that it was negligent, would unlikely have included HLT’s Trading Losses for the simple reason that the purpose that HLT had engaged Deloitte’s services for had nothing to do with its trading activities. It may be objectively foreseeable or not improbable that a company may trade at a loss – indeed, one could even say it is highly foreseeable since there are only three possible consequences from trading (a loss, profit, or breaking even) – but it is not a consequence that would be in the reasonable contemplation of an auditor in the context of the role it has undertaken.

158 Second, we consider the statutory context for liability for insolvent trading to be relevant inasmuch as it formed part of the background against which the parties contracted. As we indicated at the hearing, HLT’s case in Suit 237 comes very close to a claim against Deloitte for wrongful trading; this is because the gist of HLT’s case is that Deloitte should have acted at an earlier point in time to prevent HLT (or the Lim Family) from continuing to trade and deepening HLT’s insolvency. Given this, it is fair to assess Deloitte’s expectations as to its potential liability for HLT’s Trading Losses having regard to the relevant statutory context which would reasonably have been in its contemplation.

159 In this regard, as a preliminary point, we note that there may be some ambiguity as to whether an auditor could incur liability for wrongful trading to begin with. Under s 339(3) of the CA, which was in force at the time each of the engagement letters for the relevant audit years between FY2014 and FY2019 were signed, only an “officer” who was “knowingly a party to the contracting of a debt” which the company had no reasonable or probable ground of expecting that it would be able to pay would be guilty of the offence of wrongful trading. Wrongful trading was also only civilly actionable under s 340(2) of the CA if the “officer” had been convicted under s 339(3). It is not clear if an auditor is an “officer” of a company for the purposes of this section; it is not among the list of persons contained in the statutory definition of “officer” under s 4(1) of the CA, but it has been long-established in decisions of the English courts that an auditor is an “officer” amenable to a misfeasance summons under s 212 of the Insolvency Act 1986 (c 45) (UK): see Stone & Rolls at [189]–[190]. But, assuming without deciding in HLT’s favour that Deloitte could be liable for wrongful trading, liability would only attach if Deloitte had actual knowledge that HLT was incurring debts it could not repay. The same applies in respect of fraudulent trading under s 340(1) of the CA. While not limited in its operation to an “officer” of the company, s 340(1) of the CA only imposed liability on a person who was “knowingly a party to the carrying on” of the business of the company “with intent to defraud creditors of the company … or for any fraudulent purpose”. Again, the requisite mental state for liability was actual knowledge of fraudulent conduct by HLT (or, perhaps more accurately, the Lim Family).

160 Seen in this context, what HLT has sought to do in Suit 237 through a claim in negligence is to impose on Deloitte liability for its Trading Losses (and other heads of loss) based on a lower degree of knowledge than actual knowledge. In our view, it is unlikely that Deloitte would have assumed responsibility for HLT’s losses based on a lower degree of knowledge than that which was fixed by statute. The statutory framework would have informed Deloitte’s reasonable contemplation that it would only become liable for HLT’s Trading Losses if it acquired actual knowledge of the Lim Family’s fraudulent conduct or HLT’s irretrievable insolvency that would make it complicit if it allowed the fraud and/or wrongful trading to continue.

161 The two factors above suffice to establish that HLT’s Trading Losses do not fall within the first limb of Hadley v Baxendale. As against this, HLT’s case comes down to two main points which were emphasised by Mr Bull at the hearing. Neither, in our view, brings the Trading Losses within the scope of Deloitte’s reasonable contemplation under either the first or second limbs of Hadley v Baxendale.

162 First, Mr Bull emphasised that HLT’s pleaded case includes averments that Deloitte knew that HLT’s audited financial statements would be passed on to banks and financial institutions who would rely on them to extend credit to HLT, and that it was only because of these extensions of credit that HLT was able to continue trading despite being massively insolvent.

163 In our judgment, such knowledge, even if assumed in HLT’s favour, cuts no ice. It does not suffice as actual knowledge of the risk of the Trading Losses that would bring these losses within the second limb of Hadley v Baxendale. The pleading that Deloitte would have foreseen that there were creditors who would extend credit to HLT falls short of establishing that the Trading Losses, which arose from how HLT put the credit extended to it to use, were within Deloitte’s contemplation. The mere extension and obtaining of credit, in and of itself, did not cause HLT any loss as the increase in HLT’s liabilities was matched by a corresponding increase in its assets on the other side of the balance sheet: Galoo at 1369. The fact of the matter is that, even if it was known to Deloitte that it was enabling HLT to continue trading on credit, how HLT went about trading using such credit was not within Deloitte’s knowledge: see Deloitte & Touche v Livent Inc [2017] 2 SCR 855 at [172].

164 Second, HLT emphasised that, in the particular circumstances of this case, its business was “highly unprofitable and consistently suffered significant trading losses”. This meant that, if Deloitte was negligent and failed to identify the Trading Losses, HLT was likely to continue incurring further losses in future years as it did. In this regard, Mr Bull referred us to the decision of the English Court of Appeal in AssetCo plc v Grant Thornton UK LLP [2021] 2 BCLC 227 (“AssetCo”), and specifically to the court’s reasoning for holding the defendant auditor in that case liable for losses suffered by the company arising from two loss-making contracts due to the auditor’s negligence in detecting the fraud of its management (at [107]–[109]):

165 In our view, AssetCo is of no assistance to HLT. In the first place, the force of the authority is weakened substantially by the subsequent decision of the UK Supreme Court in Manchester Building Society, which Mr Bull also referred to, as Lord Leggatt JSC criticised the above reasoning as only going far enough to establish factual causation (at [122]). His Lordship attempted to reframe the case as turning on the fact that the company’s losses were caused by loss-making contracts which were “inherently likely to continue to generate further losses … unless or until they were terminated” (at [123]). The auditor’s negligence in failing to identify that the contracts were loss-making thus resulted in the company continuing to suffer losses from the same source in subsequent years. Drawing on Lord Leggatt JSC’s reasoning, Mr Bull sought to analogise the present case to AssetCo and submitted that it could be said that HLT’s business was also “inherently likely to continue to generate further losses” due to HLT’s (and the Lim Family’s) dismal record in trading over the years.

166 We reject this submission. First of all, it is not open for HLT to make this argument given that it is not part of its pleaded case that HLT’s trading activities were likely to generate further losses, whether due to the Lim Family’s dismal trading record over the years or some other reason. Rather, what has been pleaded is that Deloitte’s negligence enabled HLT to continue trading and thus exposed HLT to the possibility of incurring further losses (see [162]–[163] above). As far as the specifics of HLT’s trading activities are concerned, HLT has only pleaded that derivatives trading was "inherently complex and volatile”, which we take to mean that it was a type of trading that was capable of generating large losses. Nothing that establishes that HLT was likely to make losses, such as any apparent ineptitude on the Lim Family’s part in terms of their trading ability, has been pleaded. There is a clear difference between a pleading that HLT’s trading could generate significant losses and a pleading that its trading was likely to do so. HLT’s pleaded case only goes as far as the former. The analogy to AssetCo is thus untenable on the pleadings alone.

167 Second, even if AssetCo can be rationalised in the way Lord Leggatt JSC suggested and the deficiencies in HLT’s pleadings are overlooked, the analogy that Mr Bull seeks to impress upon us is in any event flawed. The crux of Lord Leggatt JSC’s reasoning was that the auditor had failed to detect a source of loss – the two contracts – which was ongoing and would thus cause the company to suffer loss from the same source unless they were terminated. To put the point another way, AssetCo involved the company having a particular wound which it continued to bleed from in subsequent years due to the auditors failing to identify the wound. The Trading Losses in this case are, however, qualitatively different. Indeed, Mr Bull fairly conceded that HLT’s case was weaker than the position in AssetCo as the latter involved “the same contract that bridge[d] over” into subsequent years. In the present case, the losses that HLT incurred year-on-year were losses from new trades that were entered into in each year. It cannot be said that Deloitte had failed to diagnose a specific source of loss which HLT continued to bleed from financially in subsequent years.

168 Faced with these difficulties, Mr Bull posited that a broader view of the “trading business as a whole” ought to be taken, such that HLT’s entire trading business could be seen as a single source of loss which continued across the years. We disagree. For one, the factual premise of Mr Bull’s argument does not hold because while HLT traded at a loss for the most part during the relevant years, it was not always loss-making. Indeed, this is likely why HLT has not pleaded that its trading activities were inherently loss-making (or likely to be so), as we have noted at [166] above, because the reality is that its trading fortunes were mixed. As Ms Lin highlighted, HLT had in fact turned a profit from derivatives trading for two of the six years between FY2014 and FY2019 (specifically, FY2017 and FY2018). And, based on its pleaded case in the Statement of Claim, HLT turned an overall profit in one of the six years between FY2014 and FY2019, even if this momentary upturn was far outstripped by the losses it had suffered in the other years:

We note that, in further and better particulars filed by HLT on 19 July 2023 in response to Deloitte’s request that it clarify a reference to its “trading losses” in HLT’s Reply and Defence to Counterclaim, HLT stated a different (and much higher figure) for its losses in FY2019 (US$1,318,000,000). It is not clear to us what the reason for this apparent discrepancy may be, but nothing turns on this as it is undisputed that HLT did not make losses for the entire six-year stretch between FY2014 and FY2019.

169 But even if HLT had not turned a profit in 2018 and it was continuously loss-making throughout the entire period, the point would remain that, since it was possible for HLT to make a profit through its continued trading, it could not be said that there was a single source of loss which Deloitte had failed to detect and which was likely or bound to continue producing losses for HLT in subsequent years. The somewhat fortuitous event of HLT having made a profit in FY2018 only serves to underscore this.

170 In the final analysis, taking a step back and looking at the circumstances as a whole, there is nothing in the relationship between the parties that indicates that Deloitte would reasonably have contemplated that it was exposing itself to potential liability for HLT’s ongoing trading simply by offering its services as a statutory auditor. Accordingly, we consider that the Trading Losses do not fall within the first limb of Hadley v Baxendale, and HLT has also failed to plead or establish any matter that brings the Trading Losses within Deloitte’s actual knowledge for the purposes of engaging the second limb of Hadley v Baxendale. The Trading Losses are therefore too remote and irrecoverable from Deloitte even if Deloitte were to be found to have been negligent.

171 For the reasons explained above, we decline to answer the Creditor Duty Question as it is academic and, in any event, it does not raise a question of law appropriate for summary determination in this appeal. However, we answer the Trading Losses Question and hold that the Trading Losses cannot be recovered from Deloitte as they are too remote. Accordingly, we order that the pleadings in Suit 237 relating to HLT’s claim for Trading Losses be struck out.

172 Finally, on the issue of costs, seeing as the Creditor Duty Question has turned out to be improperly formulated, we think it is fair that each party should bear its own costs on that issue. However, having regard to Deloitte’s success on the Trading Losses Question, we consider that Deloitte should be entitled to costs. But as our analysis on the Trading Losses Question differs somewhat from how Deloitte has argued it, we do not think that Deloitte should be awarded the full costs of that issue. In the circumstances, we fix the costs of this appeal and the costs of the underlying application for permission to appeal in CA 17, which were ordered to be in the cause of the appeal, in the total sum of $40,000 (all-in), to be paid by HLT to Deloitte. In light of Deloitte’s partial success in this appeal, we also set aside:

(a) the Judge’s costs order in favour of HLT for Deloitte’s appeal in RA 141, and substitute it with an order that HLT pay costs to Deloitte, which we fix in the sum of $15,000 (all-in); and

(b) the AR’s costs order in favour of HLT for Deloitte’s striking-out application in SUM 474, and substitute it with an order that HLT pay costs to Deloitte, which we fix in the sum of $10,000 (all-in).

173 The usual consequential orders will apply.