1 This appeal, CA/CA 30/2023 (“CA 30”), is against a recent decision of the General Division of the High Court in HC/RA 42/2023 (“RA 42”) filed in HC/S 267/2022 (“Suit 267”) ordering security for costs (“SFC”) in the manner and form of a joint undertaking furnished by the respondents’ litigation funder’s parent company and a Singapore-incorporated subsidiary of that parent company. The decision in RA 42 highlights and reinforces the wide discretion of the court as to the infinite forms of SFC that the court may accept and order.

2 In AD/OA 33/2023 (“OA 33”), the Appellate Division of the High Court (the “Appellate Division”) granted permission to appeal against the decision in RA 42 and crafted two questions of law for consideration in the appeal. The two questions of law pertain to the nature of the principles governing the exercise of the court’s discretion to accept and order the form of SFC offered by a party. A preliminary issue in the present appeal is the scope of the Appellate Division’s order in OA 33 granting permission to appeal. The respondents argue that the permission to appeal is confined to the two legal questions posed by the Appellate Division whereas the appellant adopts the position that the scope of the permission to appeal is wider and allows for an assessment of the merits of the High Court judge’s (the “Judge”) decision.

3 The appellant is Ms Lum Ooi Lin (“Ms Lum”). The respondents in this appeal are three companies – Hyflux Ltd, Hydrochem (S) Pte Ltd and Tuaspring Pte Ltd – and the joint and several liquidators of Hyflux Ltd and Hydrochem (S) Pte Ltd, Mr Cosimo Borrelli and Mr Patrick Bance.

4 The respondents are the five remaining plaintiffs in Suit 267 following the withdrawal of 33 other parties from the proceedings in 2022. The respondents’ litigation funder in Suit 267 is a company incorporated in the Cayman Islands, Omni Bridgeway (Fund 5) Cayman Invt Ltd (the “Omni Funder”). Ms Lum sought security for her costs in Suit 267 for the period until the filing and/or exchange of affidavits of evidence-in-chief (“AEICs”) in the sum of $90,000. The respondents were agreeable to furnishing SFC in the sum of $90,000. However, the parties could not agree on the form of security to be provided by the respondents. The parties duly appeared before Senior Assistant Registrar Cornie Ng (“SAR Ng”) for a decision on the matter.

5 On 14 February 2023, SAR Ng made the following orders:

6 By SAR Ng’s order (ie, HC/ORC 784/2023), the respondents would have to make payment into court of the agreed quantum of security if the parties could not come to an agreement on any one of the other forms of security mentioned in SAR Ng’s order. Unfortunately, the parties could not agree on an appropriate arrangement. The respondents maintained that an undertaking from two companies (the “Omni Undertaking”) would be a satisfactory security arrangement. The first company was Omni Bridgeway Limited, the parent of the Omni Funder and a company listed on the Australian Securities Exchange. The second was Omni Bridgeway (Singapore) Pte Ltd, Omni Bridgeway Limited’s Singapore-incorporated subsidiary (collectively, the “Omni Related Companies”). On 28 February 2023, the respondents appealed against SAR Ng’s order by way of RA 42. The Judge allowed the appeal. He accepted the Omni Undertaking as an adequate form of security and ordered that the respondents furnish security for Ms Lum’s costs for the period until the filing and/or exchange of AEICs by way of the Omni Undertaking (see Hyflux Ltd (in compulsory liquidation) and others v Lum Ooi Lin [2023] SGHC 113 (“RA 42 Judgment”) at [49]–[50]).

7 OA 33 was Ms Lum’s application for permission to appeal against the order made in RA 42. The Appellate Division allowed OA 33 on 26 June 2023 on the grounds that there were questions of general principle to be decided for the first time, which were also questions of importance upon which further argument and a decision of a higher tribunal would be to the public advantage. These questions (the “PTA Questions”) are:

8 On 28 June 2023, Ms Lum filed an appeal against the whole of the decision in RA 42 in AD/CA 64/2023 (“AD 64”). On 25 August 2023, AD 64 was transferred to the Court of Appeal on the Court of Appeal’s own motion pursuant to s 29D(1)(a) read with s 29D(2)(a) of the Supreme Court of Judicature Act 1969 (2020 Rev Ed) (“SCJA 1969”). AD 64 was renumbered as CA 30.

9 Ms Lum is represented by Davinder Singh Chambers LLC (“DSC”) while the respondents are represented by Tan Kok Quan Partnership (“TKQP”).

10 The Judge considered the powers of the court under O 23 of the Rules of Court (Cap 322, 2014 Rev Ed) (“ROC 2014”). After noting the wide discretion afforded to the court under O 23 r 2 of the ROC 2014 to order security in any form that it deems fit, the Judge then turned to consider the principles that govern the court’s exercise of discretion when determining the adequacy of the form of security that is offered. The Judge adopted as a matter of Singapore law the principles enunciated by Hargrave J in DIF III for the determination of the form of SFC, namely that (RA 42 Judgment at [11]–[12]):

11 For simplicity, the Judge synthesised the four principles into two principles to adopt as the legal framework governing the court’s exercise of discretion when determining the adequacy of the form of security that is offered. We refer to the synthesised principles as the “Mode Principles”. They are (RA 42 Judgment at [12]):

12 The Judge then went on to apply the Mode Principles to the facts of the case. In light of the first Mode Principle that there is no limitation on the form of security that the court may order, the respondents were free to propose the Omni Undertaking as a form of security (RA 42 Judgment at [39]). With respect to the second Mode Principle, the Judge was satisfied (at [40]–[49]) that the Omni Undertaking was adequate as it provided a fund or asset against which Ms Lum could readily enforce an order for costs if necessary. The Judge gave five reasons for accepting the Omni Undertaking as an adequate form of security:

(a) First, the Omni Undertaking was an irrevocable and unconditional promise to pay Ms Lum the amount of any costs order (up to $90,000) in her favour, and hence was akin to a bank guarantee.

(b) Second, the Judge was satisfied that the Omni Related Companies had sufficient assets to satisfy a costs order of up to $90,000, as the quantum of security sought was not too high and was a small fraction of their assets.

(c) Third, there was little to no risk of the Omni Related Companies not honouring the Omni Undertaking due to the substantial reputational damage they would suffer in not doing so.

(d) Fourth, Ms Lum would be able to enforce the Omni Undertaking easily, as she would have immediate recourse against Omni Bridgeway (Singapore) Pte Ltd which is based in Singapore. Even if she had to bring proceedings against Omni Bridgeway Limited in Australia, Singapore judgments can be enforced in Australia with relative ease and the Omni Undertaking provided that Omni Bridgeway Limited would not seek to set aside any Australian court judgment or seek security for costs in any proceedings by Ms Lum to enforce the Omni Undertaking there.

(e) Finally, the Omni Undertaking also required the Omni Related Companies to notify Ms Lum if the litigation funding agreement was terminated and to meet any adverse costs orders made during the term of the funding agreement up to parties’ agreed quantum for the security for costs.

13 The Judge therefore allowed the respondents’ appeal in RA 42 (RA 42 Judgment at [50]).

14 Ms Lum’s position is that the Appellate Division’s permission to appeal is not confined to examining the two PTA Questions; an assessment of the merits of RA 42 is covered by the permission to appeal. DSC for Ms Lum submits that the Judge misdirected himself in the approach taken and therefore his order should be set aside and the respondents should be ordered to furnish SFC either by way of a solicitor’s undertaking or a payment into court. The first part of this argument is that the Judge incorrectly adopted Hargrave J’s principles from DIF III as the applicable principles in determining the adequacy of a proposed form of SFC. This relates to the PTA Questions. The second part of DSC’s argument is that even if the Judge correctly determined the applicable principles, he made a number of errors in finding that the Omni Undertaking was adequate. This second part is not covered by the PTA Questions. In contrast, TKQP for the respondents contends that Ms Lum was granted permission to appeal the Judge’s decision solely on the PTA Questions. Therefore, there is no room for examining the Judge’s exercise of his discretion other than in relation to his decision on the principles covered by the PTA Questions.

15 We will deal with the scope of the permission to appeal and the further question as to whether a party’s arguments on appeal may exceed the scope of the questions framed by the Appellate Division granting permission to appeal in two parts:

16 We begin with considering the true construction of the Appellate Division’s order and its effect.

17 The Appellate Division’s order states at [7] that “[t]he court thus grants permission to appeal on the following question(s)…”. As TKQP submits, this wording is comparable to that of the order in the case of The “Jeil Crystal” [2021] SGHC 292 and its subsequent appeal in The “Jeil Crystal” [2022] 2 SLR 1385 (“Jeil Crystal (CA)”) (collectively referred to as the “Jeil Crystal”) where permission to appeal had been granted with respect to a single question (at [3] and [60] of Jeil Crystal (CA)). In that matter, the wording of the Appellate Division’s order had been similar in that the court had granted leave to appeal “on the issue at paragraph 6 above”. This was treated by the first instance and appellate courts as a limitation on the issues being appealed. Both the courts in Jeil Crystal and in OA 33 had, in granting permission to appeal, expressed the grant of permission to be “on” the stated issue/question. In our view, the Appellate Division in fact intended to limit its permission; the Appellate Division would have granted “general leave to appeal” if permission to appeal against the whole of the Judge’s decision was intended. Consequently, the appeal in CA 30 is limited to the PTA Questions.

18 We now turn to DSC’s two arguments raised to oppose a narrow reading of the scope of the permission to appeal. DSC’s submission is essentially that an appeal concerning the issue of the adequacy of the Omni Undertaking is nonetheless an appeal against the whole order made by the Judge. In other words, this court should decide any issue that would properly dispose of the appeal. DSC relies on the decision of Hong Kong Island Development Ltd v The World Food Fair Ltd & Anor [2006] HKCU 449 (“Hong Kong Island”) to advance this argument. However, Hong Kong Island does not assist the appellant at all. In fact, in Hong Kong Island, the Hong Kong Court of Final Appeal rejected the suggestion that, in a civil case, the grant of leave on any single ground entitles the appellant to raise any other matter it may wish as a ground of appeal (at [20]). It held that the appellant was prima facie confined to arguing the appeal on the basis stated by the court when granting leave, as this was part of the intended filtering process to prevent the court’s time from being taken up by unmeritorious or trivial points or with a mere re-hearing of points properly dealt with by the intermediate court (at [21]). That said, what was permissible was for the applicant to apply to the appellate court if dissatisfied with the limited basis on which leave to appeal was granted, following which the appellate court could exercise its discretion afresh. It was also noted that it was within the powers of the Hong Kong Court of Final Appeal to inform parties in advance that it would require other points to be dealt with (at [22]–[23]).

19 In further argument, DSC said that an appeal that is confined to the two PTA Questions would not make sense because it would make the appeal meaningless. There is no merit in this contention. The appeal is not rendered academic or meaningless if it is confined to the PTA Questions. If Ms Lum succeeds in showing that the Judge erred in framing the Mode Principles, the appeal would be allowed. Conversely, the appeal would be dismissed if the Judge was right in his formulation of the Mode Principles. Any which way, the appeal is capable of disposal on determination of the two PTA Questions.

20 We have already explained why DSC’s challenge on the adequacy of the Omni Undertaking is outside the scope of the permission to appeal based on the PTA Questions. That said, a related question that could possibly arise is whether this court is open to grant permission to raise other issues or additional points where there is a desire on the part of the appellant to advance arguments which are not covered by the appeal (see generally, for this court’s jurisdiction, s 59(7) of the SCJA 1969 and O18 r 8(4) of the Rules of Court 2021 (2020 Rev Ed) (“ROC 2021”)). DSC has also highlighted that s 41(6) of the SCJA 1969 provides that the Appellate Division may draw any inference of fact, give any judgment and make any order. In light of the transfer of this appeal to this court, the corresponding provision for the Court of Appeal is s 59(6) of the SCJA 1969.

21 However, having regard to the conclusions reached earlier, it is not necessary to address this related question and this court’s jurisdiction. In any case, DSC has not explained why this court should go beyond the Appellate Division’s limitation of the questions on appeal by adding a further issue/question, namely whether the law was correctly applied. We say no more on this matter.

22 We now turn to consider the PTA Questions. It appears to us that the two PTA Questions can be considered together as a broad overarching question of what the principles governing the court’s exercise of its discretion to determine the adequacy of the form of SFC are. The first Mode Principle provides as a starting point that a party is free to offer any form of security and that the court has a wide discretion to accept and order the form of SFC to be furnished, and this wide discretion should be, as held by the Judge, undergirded by the second Mode Principle.

23 In our view, the two principles upon which the Judge summarised and synthesised the existing case law are correct. Specifically, there is no default expectation that security will take the form of money or a bank guarantee. A plaintiff can offer any form of security that is workable. The court’s inquiry is not whether the proposed form is worse than cash or any other conventional security but whether it has the characteristics of a fund or asset against which a successful defendant can readily enforce an order for costs against the plaintiff.

24 Briefly, factors that shape and affect the court’s assessment of whether the form of security offered does provide or have the characteristics of a fund or asset against which a successful defendant can readily enforce an order for costs against the plaintiff are case-specific depending entirely on the circumstances of each particular case. This means that in each case the court’s inquiry into the adequacy of the form of security depends on the evidence provided and the court will weigh the evidence accordingly. Generally, evidence provided must be sufficiently persuasive for the court to form an objective belief as to the suitability of the third-party surety or guarantor in that any personal undertaking offered has characteristics of a fund or asset against which a successful defendant can readily enforce an order for costs against the plaintiff. Available evidence speaking to the third-party surety’s financial capacity to meet an adverse costs order (such as financial reports) will in most circumstances also satisfy the second Mode Principle.

25 We now elaborate on the court’s discretion under the Rules of Court and discuss in detail the parties’ submissions on the Mode Principles.

26 As rightly highlighted by the Judge, a wide discretion is afforded to the court under O 23 r 2 of the ROC 2014 to order security in any form that it deems fit. O 23 r 2 provides that:

27 For completeness, we note that in the ROC 2021, O 9 r 12(1) provides that:

28 It should also be noted that under O 9 r 12(3), it is permissible for defendants to apply for SFC from non-party litigation funders.

29 Pertinently, O 9 r 12(1) of the ROC 2021 does not expressly provide for any specific form in which security is to be provided (see Singapore Rules of Court: A Practice Guide (Chua Lee Ming and Paul Quan eds, 2023) at para 09.034). In our view, the court’s discretion to determine the form of SFC under the present ROC 2021 remains as wide as it was under O 23 r 2 of the ROC 2014.

30 The next question is how this discretion is to be exercised by the court. What are the principles to guide the exercise of discretion to accept and order security in the manner and form that is offered?

31 In DIF III, Hargrave J, having outlined the principles guiding the court’s exercise of discretion to order security in any form it deems fit (mentioned above at [10]), noted that the question for the court is not one of relative adequacy but of whether the form of SFC put forward is adequate to achieve its object as security, namely, of giving a successful defendant a fund or asset against which it can readily enforce an order for costs (DIF III at [63]–[64]; see also, Blue Oil Energy Pty Limited v Tan [2014] NSWCA 81 at [21]–[23]). As mentioned, the Mode Principles provide that: (a) the plaintiff is not restricted to any fixed form of SFC; and that (b) the plaintiff bears the burden of showing that the proposed form of security is “adequate”, ie, whether it provides a fund or asset against which a successful defendant can readily enforce an order for costs against the plaintiff (see above at [11]).

32 DSC argues that the Judge erred in adopting the principles in DIF III. Therefore, the Judge applied the wrong principles in deciding the form of SFC. DSC submits that the correct legal approach is that unless the parties agree on another form or unless the party furnishing security shows with evidence that it is unable to or will be disadvantaged if it is made to provide security in one of three conventionally and frequently employed forms (namely, bank guarantees, solicitors’ undertakings, or payments into court) (the “conventional modes”) in the sense that it would be unable to pursue its claim, the court should order that security be furnished in one of the conventional modes.

33 On the other hand, TKQP submits that the Mode Principles should be upheld. TKQP also submits that, in any event, the respondents would have met the threshold required under Ms Lum’s preferred approach as conventional modes of SFC would have come at additional cost to them. We note, however, that it is unclear what this additional cost would entail and TKQP’s position goes only so far as to say that the Omni Undertaking would be less disadvantageous than other forms of SFC.

34 DSC’s challenge to the Mode Principles essentially places emphasis on the conventional modes of SFC as the default starting point of the court’s determination. As such, in contrast to the Judge’s approach where the plaintiff’s burden is to prove that its chosen form of security is adequate and should be ordered, DSC argues that the burden of proof placed on the party furnishing security should be to establish that conventional modes should not be ordered. Only then should the court proceed to consider non-conventional modes. The effect is that DSC’s proposed approach envisions non-conventional modes of SFC as an exception to the three conventional modes, which should only be contemplated when the conventional modes have been removed from the court’s consideration.

35 In our view, DSC’s proposed approach is incorrect. We do not agree with DSC’s emphasis on conventional modes of SFC as the default starting point such that other forms of SFC should be regarded as exceptions to them.

36 First, there cannot be a default expectation that security will take the form of money paid into court, a solicitor’s undertaking or a bank guarantee. The court should not be closed to alternative forms of SFC in civil litigation, especially since, as highlighted by TKQP, what would constitute a “conventional” mode of SFC may change over time. In today’s contemporary litigation landscape, the courts should expect forms of security that are reflective of the times. To illustrate, in Tulip Trading Limited (a Seychelles company) v Bitcoin Association for BSV (a Swiss verein) and others [2022] EWHC 141 (Ch), the claimant offered to provide security in the form of Bitcoin by transferring the relevant cryptocurrency to its solicitors who would give written confirmation to the court and the defendants that they held the Bitcoin (at [34]–[36]). While the court declined to order security in the form proposed by the claimant because it was not workable for a variety of reasons (at [44]), the point here is that the court will not refuse to consider non-conventional modes of SFC just because conventional modes are still available to the claimant.

37 Further, the cases which DSC has relied on in support of Ms Lum’s position that conventional modes of security should be the starting point do not actually assist nor validate that approach. DSC has relied primarily on the cases of Infinity Distribution Ltd (in administration) v The Khan Partnership LLP [2021] EWCA Civ 565 (“Infinity”); In the matter of Pioneer Energy Holdings Pty Limited [2013] NSWSC 1366 (“Pioneer”); and Nylex Corporation Pty Ltd v Basell Australia Pty Ltd [2009] VSC 97 (“Nylex”).

38 In so far as the decisions of Infinity, Pioneer and Nylex may have been premised on there being a default starting point in conventional modes of SFC, the Judge reasoned that these decisions should not be followed in Singapore as O 23 r 2 of the ROC 2014 and O 9 r 12 of the ROC 2021 do not fetter the court’s discretion and there was also no principled reason why certain forms of security should be preferred over others (RA 42 Judgment at [17]–[21]). We agree. Going further, while the courts had in those cases ordered conventional modes of SFC instead of the non-conventional modes proposed by the parties, it is also possible to read these decisions as entailing a holistic consideration of all the various factors at play in each case. Any disadvantage, prejudice or difficulty (or lack thereof) arising from an order to furnish conventional modes of SFC was not a pre-requisite that had to be proven before non-conventional modes could even be considered; it was merely one of the factors taken into the court’s consideration as to whether a proposed form of SFC should be allowed (see Nylex at [24]–[33]; Pioneer at [28]–[29]; Infinity at [36]–[37] and [59]–[74]).

39 As such, we do not see any basis for prioritising conventional modes of security over alternative forms.

40 In light of our conclusion that there are no default forms of SFC, there is no real basis for Ms Lum’s call for the Judge’s synthesis of principles from DIF III to be rejected. DSC submits that its reading of English and Australian cases supports Ms Lum’s position (see below at [41]–[43]) and that DIF III is an outlier. We disagree. Contrary to DSC’s view, Australian and English jurisprudence is actually aligned with DIF III in saying that any proposed security, regardless of form, should be considered by way of an evidential assessment of the facts and circumstances of the case. We explain below.

41 The principles in DIF III have been expressly endorsed in Australian cases, including where SFC was sought to be provided in the form of undertakings by a litigation funder (see especially Iddles & Anor v Fonterra Australia Pty Ltd & Ors [2021] VSC 609 (“Iddles”) at [25]–[26], where the court allowed SFC to be furnished by way of an undertaking by the plaintiff’s litigation funder coupled with a payment into court for the purposes of enforcement if needed; see also Trailer Trash Franchise Systems Pty Ltd and Dale Cooney v GM Fascia & Gutter Pty Ltd [2017] VSCA 293 (“Trailer Trash”) at [57]; Re Tiaro Coal Ltd (in liq) [2018] NSWSC 746 at [22] and [49]).

42 Moreover, we agree with the Judge’s observations (at [14], [16] and [29]–[30] of the RA 42 Judgment) that English law does not differentiate between conventional and non-conventional modes of SFC when it comes to the determination of whether a certain form of SFC should be allowed. In coming to this view, the Judge considered the authorities discussed in DSC’s written submissions (namely, Rosengrens Ltd v Safe Deposit Centres Ltd [1984] 1 WLR 1334 (EWCA); Versloot Dredging BV v HDI Gerling Industrie Vesicherung AG [2013] EWHC 658 (Comm) at [10]–[15]). The Judge said that these cases stand for the proposition that the court will not insist on a fixed form of SFC and that the central consideration is whether the security is “adequate”. We agree with his analysis.

43 DSC also argues that the cases of Trailer Trash and AP (UK) Ltd v West Midlands Fire and Civil Defence Authority [2001] EWCA Civ 1917 (“AP (UK)”) support the proposition that DIF III should be treated with caution given the risk of satellite litigation arising from the form of security proposed. The end point of DSC’s argument is that the party furnishing security must first explain why one of the conventional modes is prejudicial, as ordering one of the conventional modes would mitigate the risk of satellite litigation. However, we agree with the Judge that while the risk of satellite litigation is a factor that the court should take into account, it does not necessarily follow that all non-conventional modes of security would increase the risk of satellite litigation (RA 42 Judgment at [25]). Further, as the Judge highlighted (at [26] of the RA 42 Judgment), the fact that it is generally easier to satisfy the court that a liquid form of security is adequate due to its “inherent advantages or historical usage” does not mean that non-conventional modes of security can never be adequate. The adequacy of any form of security is necessarily case-specific.

44 Taking the line of English and Australian caselaw in its entirety, it appears that the law has been generally consistent in: (a) not preferring conventional modes of SFC as a default option; (b) placing the burden of proof on the party proffering SFC to justify its preferred form of SFC; and (c) considering the adequacy of the preferred form of SFC in terms of whether it provides sufficient protection for the defendant should the defendant need to recover costs at the end of the proceedings. As such, we answer the PTA Questions in the affirmative having approved the Mode Principles, which in turn means that CA 30 must be dismissed.

45 Finally, although the Mode Principles guide the court in the exercise of its discretion, it is important to consider each case on its facts. Whether the form of security offered is acceptable and adequate is a fact-sensitive matter. The consideration of whether a party is unable to or would be disadvantaged if required to put up conventional modes of SFC is only one of the various factual circumstances that the court will take into consideration when determining the adequacy of the proffered security.

46 In this regard, we also affirm the Judge’s interpretation of the case of Global Finance Group Pty Ltd (in liq) v Marsden Partners (a firm) [2004] WASC 52 (“Global Finance”), which was relied on by DSC both below and in the present appeal to suggest that undertakings by litigation funders should be subject to especial scrutiny. The Judge found that each case would turn on its particular facts and that Global Finance did not stand for any broader principle beyond that SFC in the form of a funder’s undertaking was inadequate on the particular facts of that case (RA 42 Judgment at [36]).

47 For the foregoing reasons, we dismiss CA 30 with costs fixed at the global figure of $25,000 inclusive of disbursements for both the costs of OA 33 and CA 30. The usual consequential orders will apply.