1 This is the first, second, 16th, 37th, and 38th plaintiffs’ (“the plaintiffs”) appeal against the decision of the learned Senior Assistant Registrar (“the SAR”), who, among others, ordered that the plaintiffs furnish security for the defendant’s costs for the period until the filing and/or exchange of affidavits of evidence-in-chief by way of (a) the provision of a costs undertaking jointly by Omni Bridgeway Limited (“OB”) and Omni Bridgeway (Singapore) Pte Ltd (“OBS”) on terms satisfactory to the defendant (“the Undertaking”); (b) if not (a), a banker’s guarantee on terms satisfactory to the defendant; (c) if not (a) or (b), a solicitor’s undertaking on terms satisfactory to the defendant; and (d) if the parties are unable to agree on the terms of the costs undertaking, banker’s guarantee, or solicitor’s undertaking, then the security shall be provided by way of payment into court. I should mention that the plaintiffs have refined the Undertaking by a letter dated 6 April 2023. When I refer to the “Undertaking” below, I am referring to the Undertaking as refined by the said letter.

2 In essence, the plaintiffs are appealing against the learned SAR’s decision as to the form of the security. More particularly, the plaintiffs would like the form of security to be restricted to the Undertaking only, instead of effectively vesting the eventual form of security on the defendant’s wishes.

3 After hearing the parties and considering the relevant documents, I allow the plaintiffs’ appeal. I therefore order that the plaintiffs furnish security for the defendant’s costs for the period until the filing and/or exchange of affidavits of evidence-in-chief by way of an undertaking to the court jointly by OB and OBS, which I have abbreviated as “the Undertaking” above. Because there is no Singapore decision on the applicable principles in the determination of the appropriate form of security for costs, I set out my reasons below.

4 The parties’ positions can be stated briefly. The plaintiffs and the defendant had agreed on the quantum of security, which is $90,000. However, they disagree as to the form of the security.

5 The plaintiffs’ position is that they can provide security by way of an undertaking from their litigation funders, OB, and OBS. In particular, the plaintiffs argue that the Undertaking provides adequate protection and does not impose an unacceptable disadvantage to the defendant. The plaintiffs emphasise, among others, that the Undertaking is an irrevocable and unconditional promise by OB and OBS that is akin to a bank guarantee. This is because the Undertaking is a promise given by entities with the means to pay, who would in all likelihood make payment without the need for enforcement, and in any case, against whom enforcement can be readily obtained.

6 In contrast, the defendant’s position is that the plaintiffs should provide security by way of a solicitor’s undertaking from the plaintiffs’ lawyers. The defendant argues that the Undertaking is plainly inadequate. Among others, the defendant says that because the plaintiffs do not state whether OBS has any assets and where those assets may be, this suggests that OBS does not have any assets against which the Undertaking can be enforced. As such, the defendant argues that the Undertaking is effectively only being offered by OB, which is a foreign company. Thus, in the event that the Undertaking needs to be enforced, the defendant would be put to the time and cost of ascertaining whether OB has assets in Singapore or undertaking enforcement action overseas. This would certainly exceed $90,000, being the agreed quantum of security.

7 For reasons that I will now develop, I allow the appeal. To begin with, it is trite law that the court has a discretionary power to order, at any stage of the proceedings, a person in the position of a plaintiff to give security for his or her opponent’s costs. This would enable the defendant to recover costs from the plaintiff out of a fund within the jurisdiction in the event that the claim against him or her by the plaintiff proves to be unsuccessful. The purpose behind this is clear. While a plaintiff has a choice whether to commence proceedings against another party, and therefore run the risk of suing a party who may not be good for costs, the same cannot be said of a defendant. Indeed, a defendant cannot choose not to be sued. As such, the law treats the defendant in this regard slightly more favourably (see Singapore Civil Procedure 2021 vol 1 (Cavinder Bull gen ed) (Sweet & Maxwell, 2021) (“White Book”) at para 23/0/2).

8 The principal sources of law in relation to security for costs include O 23 of the Rules of Court (2014 Rev Ed) (“ROC 2014”) (which the parties rely on in the present case), as well as O 9 r 12 of the Rules of Court 2021 (“ROC 2021”). I pause to observe that while O 23 of the ROC 2014 and O 9 r 12 of the ROC 2021 are slightly different in wording, they are materially the same in substance. As such, the principles relevant to the provision of security for costs applicable to the ROC 2014 should apply, with the appropriate adaptations, to the ROC 2021 as well. Indeed, there is nothing in either the Report of the Civil Justice Commission (2017) (Chair: Justice Tay Yong Kwang) or the Report of the Civil Justice Review Committee (2018) (Chair: Indranee Rajah SC), both of which laid the foundations for the ROC 2021, to suggest otherwise.

9 Returning then to the ROC 2014, O 23 r 1(1) provides that if, having regard to all the circumstances of the case, the court thinks it is just to do so, then it may order security for costs in any one of four specified situations. Since the parties are here agreed as to the provision of security for costs, it is not necessary to consider whether security for costs should be ordered pursuant to O 23 r 1(1). Rather, the real dispute is as to the form of the security, to which I now turn.

10 The starting point is O 23 r 2 of the ROC 2014 (“O 23 r 2”), which provides as follows:

As can be seen, O 23 r 2 does not prescribe a particular form of security. This is despite it being well-accepted that the traditional forms of security are by bank guarantees, solicitors’ undertakings, or payments into court (see the High Court decision of PT Bumi International Tankers v Man B&W Diesel S E Asia Pte Ltd and another [2004] 3 SLR(R) 69 at [1], where the court characterised the provision of security by a banker’s guarantee as the “more conventional way” of doing so). The court therefore has a wide discretion under O 23 r 2 to order security in any form that it deems fit.

11 However, it is clear that the discretion to order security in any form must be exercised pursuant to clear principles. In this regard, Hargrave J in the Victoria Supreme Court decision of DIF III Global Co-Investment Fund, LP v BBLP LLC [2016] VSC 401 (“DIF III Global”) undertook an extensive analysis of the English and Australia case law, and laid down the following principles to guide a court in the determination of the form of security for costs (see DIF III Global at [40]):

(a) the plaintiff is entitled to propose security in a form least disadvantageous to it;

(b) the plaintiff bears a “practical onus” of establishing that the proposed security is adequate and does not impose an “unacceptable disadvantage” on the defendant;

(c) in order to be adequate, the proposed security must satisfy the protective object of a security for costs order, namely, to provide a fund or asset against which a successful defendant can readily enforce an order for costs against the plaintiff; and

(d) based on these and any other relevant considerations, the Court will determine how justice is best served in the particular circumstances of the case.

12 While Mr Jaikanth Shankar (“Mr Shankar”), who appeared for the defendant, sought to cast doubt on the correctness of Hargrave J’s principles above, I think that the principles, when viewed more broadly, represent an accurate summary of the various cases the learned judge examined. I therefore respectfully agree with and adopt these principles for the determination of the form of security for costs in Singapore. To my mind, these principles can be reduced to two for ease of application: (a) the plaintiff is not restricted to any fixed form of security for costs; and (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. While Hargrave J had proposed an overarching “justice of the case” principle in DIF III Global, I do not think it is necessary to include this in the applicable principles to guide a court in the determination of the appropriate form of security. This is because, in most cases, there will be a coincidence between the satisfaction of these two principles and the justice of the case. Of course, it goes without saying that the function of a court is to do justice in the case at hand, and so a court can consider the particular facts of each case in deciding on the appropriate form of security.

13 With these observations in mind, I turn to examine each of the two principles I outlined above.

14 First, the plaintiff is not restricted to any fixed form of security for costs. In this regard, the foreign cases have tended to frame this as the plaintiff being able to propose security for costs in a form that is least disadvantageous to him. Examples of such cases include the Victoria Supreme Court decision of DIF III Global, which I have referred to earlier, and the English Court of Appeal decision of Rosengrens Ltd v Safe Deposit Centres Ltd [1984] 1 WLR 1334 (“Rosengrens”). Indeed, in Rosengrens, Parker LJ said as follows (at 1337):

15 In my view, the principle that the plaintiff is entitled to propose a form of security that is least disadvantageous to him is a corollary to the principle that the “form of a fund or asset will be immaterial so long as it is adequate to achieve its object as a security” (see the Victoria Court of Appeal decision in Yara Australia Pty Ltd v Oswal (2013) 41 VR 425 at [10], per Redlich JA). In other words, the focus is on whether the defendant will recover the costs of the action if he succeeds. It is not the function of the court, after the defendant is assured of the adequacy of its security, to further disadvantage the plaintiff by insisting on certain forms of security while giving no legitimate advantage to the defendant (see Rosengrens at 1336). However, I would prefer not to frame this to say that the plaintiff propose security for costs in a form that is least disadvantageous to him. In my view, that creates the wrong impression that the law favours the plaintiff and distracts the analysis from the crux of the principle.

16 Indeed, the crux of the principle is simply that the court will not insist on a fixed form of security for costs. This much is clear from the cases which have allowed various forms that are not considered conventional. For example, the Victoria Supreme Court in DIF III Global allowed security for costs to be provided in, among others, the form of a deed of indemnity from the plaintiff’s UK-based insurer. Also, the English High Court in Versloot Dredging BV v HDI Gerling Industrie Vesicherung AG [2013] EWHC 658 (Comm) (“Versloot”) allowed security for costs to be provided in a deed of indemnity from the claimants’ insurer, in substitution for a first-class London bank guarantee which the claimants were initially ordered to obtain. More broadly, in the New South Wales Supreme Court decision of Re Tiaro Coal Ltd (in liq) [2018] NSWSC 746, Gleeson JA said (at [23]) that it was wrong for counsel to have suggested that the “normal” forms of security by cash deposit or bank guarantee should be viewed as preferable to other forms.

17 At this juncture, I should note that there are some foreign authorities that have considered the relative disadvantages to the parties in the action in deciding whether to allow a particular form of security to be furnished. Indeed, Mr Shankar had referred to a few of these authorities, which include the English Court of Appeal decision of Infinity Distribution Ltd (in administration) v The Khan Partnership LLP [2021] 1 WLR 4630 (“Infinity Distribution”), the New South Wales Supreme Court decision of In the matter of Pioneer Energy Holdings Pty Limited [2013] NSWSC 1366 (“Pioneer Energy”), and the Victoria Supreme Court decision of Nylex Corporation Pty Ltd v Basell Australia Pty Ltd [2009] VSC 97 (“Nylex Corporation”). These decisions were primarily justified by taking the conventional forms of security as a default starting point. In this connection, Mr Shankar had ably argued that if there was no such default starting point, then the issue of the form of security for costs will become unnecessarily complex because parties would be litigating about that endlessly. While I can see the force of Mr Shankar’s argument, I think that any such concern may be slightly overplayed. Indeed, as Mr Kenneth Tan SC (“Mr Tan”), who appeared for the plaintiffs, suggested, once the court lays down a rule, then would-be litigants in the future will structure their arrangements around that rule and reduce any unwarranted litigation.

18 Beyond this broad point, I am also of the view that those decisions should not be followed in Singapore. I begin by describing the reasoning in those decisions which have taken the conventional forms of security as a default starting point. To start off, in Infinity Distribution, the English Court of Appeal declined to order security in the form of a deed of indemnity, which the claimant proposed, and instead ordered that security be paid by way of payment into court, as the defendant proposed. In explaining his decision, Nugee LJ stated that the court should consider not only the disadvantage to the defendant if security is ordered by way of deed of indemnity, but also the disadvantage to the claimant if it is ordered by way of payment into court (at [63]). In this regard, the learned judge opined there would be substantial disadvantage to the defendant if security was provided by way of deed of indemnity (at [67]). Notably, while it was argued on behalf of the claimant that the deed of indemnity would give adequate security to the defendant, the learned judge rejected this argument (at [68]).

19 The New South Wales Supreme Court applied a substantially similar approach in Pioneer Energy, where it declined to order security in the form of a lien over the third plaintiff’s shares in the first defendant. Black J held, among others, that in the absence of an agreement between the parties, the usual form of an order for security for costs involves the payment of money into court or the provision of a bank guarantee by an Australian trading bank (at [28]).

20 Likewise, in Nylex Corporation, Mandie J declined to order that security be provided in the form of a written undertaking from the plaintiff’s third-party’s insurers to meet any costs orders in favour of the defendant because, among other things, the plaintiff failed to adduce any evidence that the insurers would have difficulty in obtaining a bank guarantee. The learned judge explained (at [29]–[30]):

21 In so far as these decisions are premised on there being a default starting point in the conventional forms of security, I respectfully disagree, especially in relation to the Singapore context. First, the plain wording of O 23 r 2 (as well as O 9 r 12 of the ROC 2021) does not fetter the court’s discretion as such. Second, there is no principled reason why some forms of security should be preferred over others. Indeed, as Tuckey LJ suggested in Mohammad Ali Aoun v Hassan Bahri [2002] EWCA Civ 1390 (“Aoun”) (at [11]), what is regarded as a “traditional” form of security today may well not be so tomorrow. Third, as long as the plaintiff’s proposed security is adequate, the purpose of the defendant’s security for costs application has been met. There is no further reason why the court should not only compel the plaintiff to provide security, but to provide it in a particular form.

22 Moreover, while Mr Shankar asked that I treat with caution Hargrave J’s approach in DIF III Global in light of the decisions of the Victoria Court of Appeal in Trailer Trash Franchise Systems Pty Ltd v GM Fascia & Gutter Pty Ltd [2017] VSCA 293 (“Trailer Trash”) and of the English Court of Appeal in AP (UK) Ltd v West Midlands Fire and Civil Defence Authority [2001] EWCA Civ 1917 (“AP (UK)”), I do not think that the reasoning in these decisions justifies a departure from the approach in DIF III Global.

23 I begin with Trailer Trash, where the Victoria Court of Appeal stated in obiter that where the court has a choice, it should ordinarily prefer security in a liquid form over a personal undertaking by a third party other than a financial institution (at [59]). The court justified this on the basis of the provisions of the Civil Procedure Act 2010 (Vic) (“Civil Procedure Act 2010”), which, in brief, provides that the court should give effect to the overarching purpose of “facilitate[ing] the just, efficient, timely and cost-effective resolution of the real issues in [the] dispute”. In doing so, it should have regard to a number of objects, including the efficient conduct of the business of the court, the efficient use of judicial resources, and the timely determination of the civil proceeding. The court concluded that a form of security which does not provide a fund that can be accessed without the cooperation of the opposing party or a person who is connected to that party, and may require the commencement of proceedings to enforce it, has the potential to undermine that overarching purpose as it can give rise to satellite proceedings and lead to additional delay and costs, which would be contrary to the principle of finality in litigation.

24 In AP (UK), the English Court of Appeal expressed what was in substance the same concern. Longmore LJ (with whom the other Lord Justices agreed) distinguished Rosengrens (where the same court was concerned with the options of payment of money into court, a solicitor’s undertaking, and banker’s guarantee), from the option proposed by the appellant in AP (UK), which was a charge on real property. The learned Lord Justice opined that while the alternatives to payment of money into court in Rosengrens were “simple and straightforward” if enforcement became necessary, the principle in Rosengrens was not so broad as to include “a charge on real property with all the risks that would follow from enforcement of that charge and a forced sale” (at [15]).

25 While I agree with these authorities that the risk of satellite litigation is a factor that the court should take into account, it does not necessarily follow that all non-conventional forms of security would increase the risk of satellite litigation. Indeed, the comments by the Victoria Court of Appeal in Trailer Trash appear to have been clarified by the more recent Victoria Supreme Court decision of Iddles v Fonterra Australia Pty Ltd [2021] VSC 609 (“Iddles”). In Iddles, Efthim AsJ opined that it cannot be that the effect of the Civil Procedure Act 2010 was to eliminate undertakings or deeds of indemnities from insurers as forms of security (at [47]–[48]). In any event, any concerns about the likelihood of satellite litigation may be more appropriately addressed under the inquiry of whether the proposed form of security is adequate, which is distinct from the question of whether there is a default starting point as to the form of security to begin with. This is supported by a closer reading of AP (UK), which reveals that Longmore LJ’s real concern was whether the proposed form of security was adequate. Tellingly, following his discussion of Rosengrens and another decision, the conclusion reached by the learned Lord Justice was that it was “impossible to conclude that the security offered by the claimants was adequate security for costs” [emphasis added] (at [18]).

26 At the end of the day, as Mr Tan submitted, it bears repeating that the overarching consideration is whether the proposed form of security is adequate to ensure that the defendant will recover the costs of the action if he succeeds. That being said, there will be some forms of security that are more readily characterised as being adequate either due to their inherent advantages or historical usage. It may be easier for a plaintiff to establish adequacy in respect of these forms of security. However, this does not mean that forms of security outside of these traditional ones can never be adequate; it all depends on their characteristics and how they apply to the facts of the case at hand.

27 Second, the plaintiff bears the burden of showing that the proposed form of security is adequate to protect the defendant’s interest in the recovery of costs. The plaintiff bears the burden because he has commenced the action against the defendant, and the purpose of security for costs is to protect the defendant against unrecoverable costs.

28 As for the adequacy of the form of security, Tuckey LJ said in the English Court of Appeal decision of Aoun that while other forms of security apart from the traditional ones cannot be ruled out, these must be “copper bottomed – in the sense that they can be enforced in a simple and straight forward way – otherwise the purpose of ordering security is defeated” (at [11]). In other words, the core consideration is whether the security is adequate, in the sense of being enforceable in a simple and straightforward manner, such that the defendant can recover the costs of the action if he succeeds.

29 This core consideration as articulated in Aoun has been applied in subsequent English cases. For example, to recapitulate the facts in Versloot, the plaintiffs applied to vary an order that security for costs be provided by way of a “first class” London bank guarantee. The plaintiffs had done this so that they could provide a deed of indemnity from an insurer in substitution for the bank guarantee. Clarke J varied the order. In so varying, the learned judge said this (at [10]):

30 Similarly, in the English High Court decision of Harlequin Property (SVG) Ltd and another v Wilkins Kennedy (a firm) [2015] EWHC 1122 (TCC), Coulson J had to consider whether an “after the event” (“ATE”) insurance policy could be sufficient security under English law. The learned judge decided that adequate security for costs could be provided by a defendant through a means other than a payment into court or a bank guarantee. As such, depending on the terms of the insurance and the circumstances of the case, there was no reason why an ATE insurance policy could not provide adequate security. However, there would be circumstances where an ATE insurance policy could not do so. For example, if there are provisions in the ATE insurance policy that reduce or obliterate the security on the occurrence of certain events, then it may well be the case that the policy cannot provide adequate security.

31 The same core consideration also applies in Australia. Thus, in DIF III Global, the plaintiffs appealed against a lower court’s decision that security should be provided in the form of “deposit into Court, or by way of [a] guarantee from an agreed Australian bank or other authorised deposit-taking institution” (see the Victoria Supreme Court decision of DIF III Global Co-Investment Fund, LP v BBLP LLC [2015] VCS 484 at [131]). Instead, the plaintiffs wanted to provide security in the form of, among others, a deed of indemnity to be given by a United Kingdom insurance company, AmTrust Europe Ltd (“AmTrust”).

32 Applying these principles which he had formulated (see [11] above), Hargrave J allowed the appeal. He concluded that, among others, the provision of the AmTrust indemnity would provide adequate protection to the defendants in case they needed to call upon the security to satisfy any costs order in their favour. In particular, the learned judge noted (at [83]) that (a) the AmTrust indemnity was irrevocable and unconditional; (b) the AmTrust indemnity was directly enforceable against AmTrust in Victoria and was governed by the laws of Victoria; (c) AmTrust was based in the United Kingdom, which was a jurisdiction that had clear and straightforward arrangements for the enforcement of Victorian judgments; (d) the plaintiff had offered extra security to cover the cost of any enforcement in the United Kingdom which might be necessary; (e) the evidence showed that AmTrust had significant assets in the United Kingdom and was generally of good financial standing; and (f) as a large regulated insurer, which was also in the business of underwriting legal expense risk, it was unlikely that AmTrust would default on the deed. More broadly, the learned judge’s careful consideration of the facts demonstrates how the ascertainment of whether a given form of security is adequate is an intensely fact-intensive enquiry.

33 With these principles in mind, I come to the facts of the present case.

34 To begin with, the plaintiff here is not restricted to any fixed form of security for costs. As such, the mere fact that the Undertaking is one from the plaintiffs’ litigation funders does not, without more, mean that it is inadequate to protect the defendant’s interest in the recovery of costs. I therefore cannot agree with the defendant in so far as she suggests as a general proposition that the courts do not consider undertakings from funders to be an adequate form of security.

35 In this regard, the defendant cites two cases which she claims support the general proposition being advanced. However, these cases all turned, as they should, on their particular facts. I do not think they can be extrapolated to stand for any broader principle beyond that a security in the form of an undertaking by funders was not adequate on the particular facts of those cases.

36 First, the defendant points out that Roberts-Smith J in the Supreme Court of Western Australia decision of Global Finance Group (in liq) v Marsden Partners [2004] WASC 52 (“Global Finance Group”) declined to order that security be provided by an undertaking from the plaintiff’s litigation funder although the funder was listed on the Australian Stock Exchange. However, a brief examination of the learned judge’s reasoning will demonstrate that his decision was specific to the facts of that case and should not be taken to stand for any broader proposition that an undertaking by litigation funders is always inadequate. In this regard, Roberts-Smith J had summarised (at [88]) the reasons for not allowing security to be provided by such an undertaking as including: (a) the conduct of the funder’s business was subject to commercial risk, especially since litigation funding was “a developing and still evolving area of business” (at [87]), (b) the action might not come to trial for possibly two years or more, and (c) the defendant’s costs were likely to be substantial. It is important to bear in mind that Global Finance Group was decided some 20 years ago in 2004. It is therefore understandable why Roberts-Smith J might have doubts about litigation funding as a sustainable business. Moreover, the undertaking in question there was A$700,000, which was substantial compared with the funder’s assets and state of present and future businesses.

37 Second, the defendant then points to the New South Wales Supreme Court decision of Northern Southern Western Supermarkets Pty Ltd v HIH Casualty and General Insurance (in Liq) [2002] NSWSC 541 (“Northern Southern Western Supermarkets”), where Einstein J also declined to order that security be provided by an undertaking from the holding company of the plaintiff’s litigation funder. However, as with Global Finance Group, the reasoning in Northern Southern Western Supermarkets was particular to the facts of the case. Thus, Einstein J had said this (at [29]):

As can be seen, the learned judge was convinced by the specific characteristics of the funder in that case in holding that an undertaking by the funder would not be an adequate form of security. Also, while the learned judge did not allude to this fact, it should be kept in mind that Northern Southern Western Supermarkets was decided in 2002, which is about the same time as when Roberts-Smith J in Global Finance Group described litigation funding as a developing area of business. Accordingly, I cannot think that the case stands for any broader proposition that the defendant appears to extrapolate from it.

38 Indeed, to my mind, it is not helpful to find examples of cases that point one way or the other without bearing in mind the overarching principles at play. This much is demonstrated by the plaintiffs’ citation of the Victoria Supreme Court decision in Iddles, in which Efthim AsJ allowed the security for costs to be provided in the form of an undertaking by the plaintiff’s funder. The learned judge had considered the cash position and assets of the funder, as well as the risk and reputational damage if the undertaking was not honoured. He concluded that there would be little or no risk of satellite litigation and therefore ordered the security to be in the form of an undertaking by the funders (at [48]). As with the two cases advanced by the defendant, I do not think, nor do the plaintiffs contend, that Iddles can be extrapolated to stand for any broader proposition other than a security in the form of an undertaking by litigation funders was adequate in that case.

39 Accordingly, as a starting point, the plaintiffs are entitled to propose the Undertaking as security for costs in the present case. The nature of that Undertaking, being an undertaking by its litigation funders, does not mean that it is inadequate. That is the subject of a separate inquiry on the facts of this case, to which I now turn to.

40 In my judgment, the Undertaking is an adequate form of security for costs, in that it provides a fund or asset against which the defendant can readily enforce an order for costs if necessary. I say so for the following reasons.

41 First, similar to the undertaking in DIF III Global, the Undertaking here is an irrevocable and unconditional promise by OB and OBS to pay to the defendant the amount of any costs order (up to S$90,000) that may be made in her favour. I agree with the plaintiffs that this makes the Undertaking akin to a bank guarantee, provided that it can be shown that OB and OBS have the means to pay, who would in all likelihood make payment without the need for enforcement.

42 Second, I am satisfied that OB and OBS have sufficient assets to satisfy a costs order of up to S$90,000, which is, relatively speaking, not substantial in the context of the present litigation. In this regard, the defendant complains that while the plaintiffs have disclosed that OB has substantial assets, they have said nothing about whether OBS has any assets. As such, the defendant suggests that this means that OBS does not have any assets against which the Undertaking can be enforced. I disagree. It is clear from OBS’s audited accounts for financial year 2022 that it has net assets of about S$2.2m and profit before tax of about S$533,739. This should alleviate the defendant’s concern that OBS has no assets in Singapore that it can enforce the Undertaking against. As for OB, which is a publicly listed company, its latest financial statements from the financial year that had ended on 31 December 2022 record that it has substantial net assets of about S$605m, with cash and cash equivalents of about S$101m.

43 In a related vein, the defendant also argues that OB and OBS are in the business of litigation funding, which is subject to commercial risks. Therefore, the defendant says that she cannot begin to consider the commercial risks she is exposed to as a result. Again, I disagree. It is true that litigation funding is subject to commercial risks. But so too are other commercial enterprises, including even banks, as recent events show. It is not realistic to eliminate all commercial risks. However, given the assets owned by OB and OBS, I consider the risk of either entity failing in the time that the security for costs is provided to be extremely low. Furthermore, it is also important to bear in mind that the quantum of security sought here is not too high, and is certainly a small fraction of the assets owned by OB and OBS.

44 Third, there is little to no risk of OB or OBS not honouring the Undertaking because of the substantial damage that they would suffer in not doing so. This is because OB and OBS are in the business of litigation funding. If they will not honour an undertaking in respect of an ongoing litigation, I can imagine that potential or current clients will not see them in a positive light. Thus, similar to Iddles, there would be little or no risk of satellite litigation surrounding the enforcement of the Undertaking.

45 Fourth, the defendant can enforce the Undertaking easily against OB or OBS. Crucially, the defendant would have immediate recourse against OBS, which is based in Singapore. As to the defendant’s argument that OBS does not have sufficient assets to meet any potential claim, I have already explained why I disagree with that contention. Even if the defendant has to go against OB in Australia, it is clear that Singapore judgments can be enforced in Australia with relative ease. Therefore, if the defendant even comes to that, I do not think it is correct for her to assert that there are serious questions whether the Undertaking can be enforced in Australia or that she will be put to the time and cost of enforcing the Undertaking there. Moreover, I note that it is a term of the Undertaking that OB will not seek to set aside any court judgment in the courts of Australia or seek security for costs in any proceedings by the defendant to enforce the Undertaking there. Indeed, the plaintiffs have agreed to strengthen this particular area of the Undertaking via the amended clause (c) in their letter dated 6 April 2023, which provides:

46 This term brings the Undertaking closer to the terms of the deed of indemnity in Iddles, which included clauses providing that if AmTrust, a third-party insurer, became liable to pay in accordance with an adverse costs order, it would undertake (a) to consent to judgment being entered against it in favour of the defendants in Australia, (b) to the Australian judgment being registered in the English High Court, (c) to not seek to set aside the registration of the Australian judgment in that court, and (d) to not seek security for costs against the defendants for proceedings for the registration or enforcement of the Australian judgment in the UK (at [60]). In this regard, I respectfully adopt the view of the Victoria Supreme Court in Iddles that these enforcement provisions narrow the scope of problems that the third party can create (at [61]) and constitute a factor that points in favour of finding that the Undertaking is adequate security.

47 Finally, it is also a term of the Undertaking that OB and OBS will (a) notify the defendant in writing within seven business days of the funding agreement being terminated, and (b) meet any adverse costs orders made during the term of the funding agreement up to parties’ agreed quantum of the security for costs. As such, I am satisfied that the defendant will be protected against the risk of incurring substantial costs without knowing that the funding agreement has been terminated.

48 In the end, Mr Shankar asked a rather obvious question. If the quantum of the security for costs in the present case is merely $90,000, why are the plaintiffs going through all this trouble of justifying the Undertaking and even fortifying it via the letter dated 6 April 2023? Indeed, during the hearing before me, Mr Tan had said his clients were willing to pay $20,000 into court as a cash component to the security. While I can only surmise that the plaintiffs have changed their minds about this because the revised Undertaking in the letter dated 6 April 2023 does not contain a clause to this effect, Mr Shankar asked, quite rightly, if the plaintiffs are willing to pay $20,000 into court, then why are the parties arguing over the difference of $70,000? I accept Mr Shankar’s concerns. However, I take Mr Tan’s point that the plaintiffs’ interest, beyond the form of security in the present case, is a point of principle about the appropriate form of security in other cases.

49 I accordingly find that the Undertaking is an adequate form of security, in that it provides a fund or asset against which the defendant can readily enforce an order for costs if necessary.

50 For all these reasons, I allow the plaintiffs’ appeal. I order that the plaintiffs furnish security for the defendant’s costs for the period until the filing and/or exchange of affidavits of evidence-in-chief by way of the Undertaking to the court jointly by OB and OBS.

51 Unless the parties are able to agree on an appropriate costs order for the present appeal, they are to tender written submissions, no longer than 5 pages each, within 14 days of this judgment.

52 In closing, I would like to thank Mr Tan and Mr Shankar for their most helpful and clear submissions.