1 A company commences litigation. Before judgment in the litigation, the company goes into insolvent liquidation. A liquidator is duly appointed. She has a duty to form an independent view on the litigation and to decide whether to adopt it or to discontinue it. Adopting the litigation carries with it the reward of a judgment in the company’s favour, but also the risk of a costs order against the company. Discontinuing the litigation limits the company’s risk of having to pay the costs of a potentially unsuccessful claim but could also amount to abandoning the reward of a successful claim.

2 The liquidator is well-advised. She knows that a rule of insolvency law – known as the estate costs rule – accords priority to a company’s liability under an adverse costs order made while the company is in liquidation. Indeed, the estate costs rule places such great importance on a company in liquidation satisfying its liability under a costs order in full that the rule accords super priority to the liability. Thus, the liability ranks ahead even of the liquidator’s own remuneration and expenses and the company’s other preferential debts, not to mention the company’s unsecured debts.

3 This liquidator is not responsible in any way for the company’s pre-liquidation decision to commence the litigation. She is merely deciding post-liquidation whether to adopt the litigation. Does the estate costs rule distinguish between litigation which a liquidator commences and litigation which a liquidator merely adopts? And if it does, does the rule accord super priority only to the successful opponent’s post-liquidation costs?

4 Those, in a nutshell, are the questions which the liquidator of Sembawang Engineers and Constructors Pte Ltd (“the Company”) raises on this application.

5 The liquidator seeks the court’s directions on these questions under s 273(3) of the Companies Act (Cap 50, 2006 Rev Ed) (“the Act”). This application poses the following questions to me:

6 The defendant in the litigation which the liquidator must now decide whether to adopt or discontinue is Metax Eco Solutions Pte Ltd. I shall refer to this company for convenience as “the defendant”, even though it is not a defendant in this application. The defendant was permitted to intervene in this application in order to present written and oral submissions on the questions posed. Not surprisingly, the defendant opposed the liquidator’s position on those questions.

7 Having heard the submissions and reviewed the authorities, I have held in favour of the defendant. In my view, the estate costs rule encompasses both litigation which a liquidator commences and litigation which she adopts. It is also my view that the estate costs rule accords priority to the company’s entire liability under an adverse costs order, because that liability is incapable of being resolved in a legal sense into a pre-liquidation component and a post-liquidation component.

8 The liquidator then invites me to apply my discretion under s 283(3) of the Act to disapply the estate costs rule on the facts of this case. I do not consider that that would be a proper exercise of my discretion under that section. I find it impossible to distinguish in any principled way this Company’s litigation against this defendant from the general type of litigation which typically falls within the scope of the estate costs rule. Accepting the liquidator’s invitation would amount to an unprincipled and illegitimate circumvention of the estate costs rule.

9 The liquidator has appealed against my decision. I now set out my reasons in full.

10 The underlying facts are not in dispute and are not material other than as background or context.

11 The Company commenced suit against the defendant at the end of 2012 claiming about $3.6m in damages. The defendant rejected the claim and counterclaimed about $2.1m from the Company. The fact that the Company is the plaintiff in the suit is significant. It means that the parties’ dispute cannot be resolved simply by the liquidator adjudicating a proof of debt. The adjudication procedure operates only to resolve claims against the Company. It has no application to claims by the Company. The fact that the defendant has a counterclaim in the suit is also significant. It means that each party is simultaneously in the position of a plaintiff and a defendant in the suit.

12 The claim and counterclaim were tried before me over eight days in 2015. After the trial, parties filed and exchanged written closing submissions and written reply submissions in the usual way. A date was fixed in September 2015 for both parties to make their oral closing submissions. ¹

13 It appears, however, that the Company was in severe financial difficulties well before trial. ² In September 2015, before I could hear the parties’ oral closing submissions, the Company secured an order under s 210(10) of the Act staying all litigation against it ³ in anticipation of proposing a scheme of arrangement to its creditors under s 210(1) of the Act. In January 2016, the Company secured leave to propose a scheme of arrangement to its creditors ⁴ (see Re Sembawang Engineers and Constructors Pte Ltd [2016] 3 SLR 1057).

14 The scheme failed. ⁵ In February 2016, the Company applied to place itself in judicial management. ⁶ In June 2016, a judicial management order was made. ⁷ The applicant was appointed as one of the Company’s judicial managers.

15 The judicial management failed. In August 2017, the judicial managers applied for and secured their discharge. ⁸ The Company was immediately placed in compulsory liquidation. ⁹ The liquidator immediately took office as one of the Company’s liquidators.

16 The defendant’s counterclaim in the suit against the Company has been automatically stayed, first under s 210(10) of the Act, then under ss 227C(c) and 227D(4)(d) of the Act and finally under s 262(3) of the Act. The plaintiff’s claim against the defendant in the suit was, however, not stayed. That is because the automatic stays apply only to suits against the Company and not to suits by the Company. But because the Company’s claim cannot proceed to judgment separately from the defendant’s counterclaim, its claim against the defendant has been adjourned in parallel with the counterclaim.

17 In September 2017, the defendant lodged a proof of debt for $2.7m. ¹⁰ That sum comprises the full value of the defendant’s counterclaim of $2.1m and a further $0.6m in legal costs incurred thus far in defending the Company’s claim and advancing the defendant’s counterclaim. The liquidator has yet to adjudicate upon the defendant’s proof of debt.

18 The defendant has little prospect of recovering any meaningful dividend on its proof, even if the liquidator admits it. That is because the Company is deeply insolvent. It has assets estimated to be worth $24m but owes over 700 creditors $191m and owes contingent creditors a further $177m. ¹¹

19 The Company’s deeply insolvent position means that the questions which the liquidator poses in this application (see [5] above) are not academic. They are of practical significance. The defendant’s pre-liquidation costs of the litigation are virtually the whole of the costs of the litigation. These costs arise ab initio, and include receiving the writ and the statement of claim, drafting the defence and the counterclaim, conducting discovery and the trial, and run right up to filing the written closing submissions. The defendant’s post-liquidation costs, by contrast, are confined virtually to the costs of and incidental to presenting and responding to the oral closing submissions. The defendant’s post-liquidation costs are therefore nominal by comparison to its pre-liquidation costs. The liquidator candidly concedes that, if she adopts the litigation and the Company loses, its assets are insufficient to pay the defendant’s pre-liquidation costs in full. ¹² But it appears that the Company does have enough assets to pay the defendant’s post-liquidation costs in full.

20 Before answering the questions posed to me, I should point out that the issues which this application raises are confined to how our insolvency law treats discretionary costs awarded against a company in insolvent liquidation, ie costs which the court awards under para 13 of the First Schedule to the Supreme Court of Judicature Act (Cap 322, 2007 Rev Ed) (“SCJA”) and O 59 r 2(2) of the Rules of Court (Cap 322, R 5, 2014 Rev Ed) (“the Rules”). The issues which this application raises do not touch on costs which are payable as of right, ie costs which a litigant has a legal right to recover from the counterparty, whether by reason of contract (see, eg, Gomba Holdings (UK) Ltd and others v Minories Finance Ltd and others (No 2) [1993] 1 Ch 171 or more recently Chee Kheong Mah Chaly and others v Liquidators of Baring Futures (Singapore) Pte Ltd [2003] 2 SLR(R) 571 (“Baring Futures”) at [49]) or (in limited circumstances) as damages for a legal wrong (see British Racing Drivers’ Club Ltd and another v Hextall Erskine & Co [1996] PNLR 523 and more recently Hermann v Withers LLP [2012] EWHC 1492 (Ch)).

21 I make two points about the nature of discretionary costs.

22 First, discretionary costs are awarded on the indemnity principle. The indemnity principle dictates that a successful party ought to recover an indemnity – no less, but also no more – for the costs which it has had to incur as between solicitor and client in order to vindicate its legal rights in the litigation (see Then Khek Koon and another v Arjun Permanand Samtani and another and other suits [2014] 1 SLR 245 at [153]–[160] and Harold v Smith (1860) 5 II & N 381 at 385). For that reason, the discretion to award costs is to be exercised ordinarily so that costs follow the event (see O 59 r 3(2) of the Rules). A court will award discretionary costs in the course of litigation as and when the event to which those costs relate becomes known. Thus, the court typically awards costs for an interlocutory application when the application is decided. And the court typically awards costs for the overall litigation ab initio together with or immediately after judgment. In both instances, the costs are awarded when the event which they follow becomes known by judicial determination.

23 Second, the indemnity principle yields only a partial indemnity to the receiving party for the solicitor and client costs it has incurred to secure the event in its favour. The indemnity principle dictates that a receiving party recovers only a reasonable amount for the costs which it has reasonably incurred as between solicitor and client to secure its success in the litigation. Who bears the burden of proving reasonableness or unreasonableness, as the case may be, depends on whether the costs are awarded on the standard basis or on the indemnity basis. But even where costs are awarded on the indemnity basis, the costs ordered are not a full indemnity to the receiving party, ie precisely equal to its actual solicitor and client costs (see Maryani Sadeli v Arjun Permanand Samtani [2015] 1 SLR 496 at [13]). In our system, there will always be a gap between the costs which the receiving party has actually incurred as between solicitor and client and the costs which the court holds it ought reasonably to have incurred as between party and party.

24 Before going further, I make four points on terminology.

25 First, when this application was filed and decided, the Insolvency, Restructuring and Dissolution Act 2018 (Act 40 of 2018) (“IRDA”) had been passed but was not yet in force. I therefore refer to the relevant legislative provisions as they stood in the Act at the time of this application and not to their current counterparts in the IRDA. Having said that, I do not consider that the questions posed on this application would be answered any differently had the IRDA been in force at the material time.

26  Second, I assume for ease of exposition that a company’s liquidation commences on the day that the court orders it to be wound up and appoints a liquidator (contrary to the position deemed by s 255(2) of the Act). I also assume that a liquidator will take a decision whether to adopt or discontinue the litigation as soon as she is appointed (contrary to the practical realities). I therefore use the terms “pre-liquidation” and “post-liquidation” as convenient albeit inaccurate shorthand to refer to each side of the scintilla temporis on which the issues in this application turn.

27 Third, and following from my second assumption, I shall use “pre-liquidation costs” to refer to legal costs which the opponent of a company in liquidation incurs as between solicitor and client before the commencement of the liquidation (as I have defined it at [26] above). “Post-liquidation costs” thus means the solicitor and client legal costs which the litigation opponent incurs after the commencement of the liquidation.

28 Finally, when I use the term “costs order”, I mean a costs order which a court makes against a company after it goes into formal liquidation. A costs order made against a company before it goes into formal liquidation is simply a provable debt and is, of course, outside the scope of the estate costs rule. And when I use the term “liquidation”, I mean an insolvent liquidation. A costs order made against a solvent company which is in voluntary liquidation is likewise outside the scope of the estate costs rule.

29 Mr Paul Seah appears as counsel for the liquidator. Mr Chandra Mohan appears as counsel for the defendant.

30 A long line of Commonwealth precedent has established two propositions about the scope of the estate costs rule: ¹³

(a) The estate costs rule applies whether the liquidator commences litigation or merely adopts it.

(b) The estate costs rule accords priority to a company’s entire liability under a costs order even if the order in part indemnifies the company’s opponent for pre-liquidation costs.

31 These two propositions are directly contrary to Mr Seah’s case. But it is common ground that there is no precedent which is binding on me which has established these two propositions as part of Singapore’s insolvency law.

32 Mr Seah’s primary argument ¹⁴ is that the two propositions at [30] above are not part of Singapore’s insolvency law. He therefore submits that a costs order is not part of “the costs and expenses of the winding up” within the meaning of s 328(1)(a) of the Act in so far as it relates to pre-liquidation costs.

33 Mr Seah’s alternative argument is that these two propositions at [30] above should not be part of Singapore’s insolvency law. He submits it is consistent with fairness, justice and equity to accord super priority to a costs order only in so far as it indemnifies the company’s opponent for its post-liquidation costs. ¹⁵

34 Mr Mohan relies on precedent, principle and policy to argue that the two propositions at [30] above are and should be part of Singapore’s insolvency law. As a matter of precedent, he relies on the Commonwealth authorities which establish these two propositions. As a matter of principle, he submits that the distinctions urged by Mr Seah between commencing and adopting litigation and between pre-liquidation and post-liquidation costs are inconsistent with the rationale underlying the estate costs rule. Finally, as a matter of policy, he submits that rejecting these distinctions strikes the correct balance between the competing policy considerations which underlie the estate costs rule.

35 The Court of Appeal first confirmed that the estate costs rule was a part of Singapore’s insolvency law in Baring Futures. The leading judgment now on the rule is the Court of Appeal’s decision in Ho Wing On Christopher and others v ECRC Land Pte Ltd (in liquidation) [2006] 4 SLR(R) 817 (“ECRC Land”). In ECRC Land, the Court of Appeal stated the content of the estate costs rule in the following terms (at [9]):

I should point out in passing that the super priority accorded by the estate costs rule is not absolute: it is subject to the direct costs of realising the assets (see ECRC Land at [4]; Re Movitex Ltd [1990] BCC 491 (“Re Movitex”) at 496E–F).

36 I consider the ratio of ECRC Land to comprise four propositions. I am, of course, bound by each of these propositions.

37 First, the estate costs rule is a rule of Singapore’s insolvency law. It is true that counsel for the liquidators in ECRC Land conceded the point (at [18]). Counsel in Baring Futures also conceded the point (at [47]). To dispose of the appeal in ECRC Land, therefore, it sufficed for the Court of Appeal simply to rely on the Commonwealth authority establishing the estate costs rule as stating Singapore law (at [9]).

38 I am bound to accept that the estate costs rule is a part of Singapore’s insolvency law. But the concession which counsel made in ECRC Land is significant. It meant that the Court of Appeal did not need: (a) to address explicitly whether the estate costs rule should be a part of Singapore’s insolvency law; or (b) to identify the rationale underlying the estate costs rule. To answer the questions posed to me, I have attempted to identify the rationale of the rule at [72]–[90] below. That analysis leads me to conclude that, even if I were not bound by the ratio of ECRC Land, I would nevertheless hold that the estate costs rule should be a part of Singapore’s insolvency law.

39 Second, a costs order is one of “the costs and expenses of the winding up” within the meaning of s 328(1)(a) of the Act. A costs order is therefore what I shall call for convenience a “liquidation expense”.

40 Again, because of the concession by counsel in ECRC Land, the Court of Appeal did not have to address expressly why that is so. Thus, it sufficed for the Court to assume the point when holding that a costs order is accorded super priority over all the other liquidation expenses which have priority under s 328(1)(a) of the Act. As Chan Sek Keong CJ said (at [9]):

41 Third, a liquidator who breaches the estate costs rule is personally liable for any shortfall in the opponent’s costs recovery which is caused by the breach of the rule (ECRC Land at [20]–[21]). The purpose of the personal liability is to uphold the efficacy of the estate costs rule and to avoid the super priority which it confers from becoming illusory (ECRC Land at [24]). The basis for the liquidator’s personal liability is not the court’s adjudicatory jurisdiction to order a non-party to pay the costs of litigation (see [50] below). The basis for this personal liability is the court’s supervisory jurisdiction over liquidators under s 313(2) of the Act (ECRC Land at [47] and [53]). I add that there is also a common law supervisory jurisdiction which the court can exercise over liquidators as officers of the court to the same effect (Ex parte James; In re Condon (1874) LR 9 Ch App 609; Re PCChip Computer Manufacturer (S) Pte Ltd (in compulsory liquidation) [2001] 2 SLR(R) 180).

42 This third proposition is the only proposition which was in contention between the parties in ECRC Land (at [21], see also [15]). I therefore consider it to be the primary ratio of ECRC Land.

43 Fourth, the super priority which the estate costs rule accords to a costs order is a conscious normative choice which strikes the appropriate balance between the competing policies which underlie the issue (ECRC Land at [90]). The Court of Appeal identified these policies as: “(a) the interests in facilitating liquidators’ duties to recover insolvent companies’ assets for the benefit of their creditors; and (b) the need to protect defendants from having to shoulder the legal costs of defending unmeritorious suits”.

44 My answer to the questions posed in this application begins with an analysis of the Commonwealth cases cited by both parties. These cases establish the two propositions at [30] above in the insolvency law of the respective jurisdictions.

45 The estate costs rule originates from the English cases. I therefore begin in England.

46 The origins of the estate costs rule lie in the English courts of Chancery. The earliest case cited to me is the House of Lords’ decision in Charles Boynton v George H Boynton and others (1879) 4 App Cas 733 (“Boynton”).

47 In Boynton, a plaintiff commenced litigation and won it. She secured the usual order for costs to follow the event (at 735). The plaintiff then died. The defendant served a notice of appeal on the plaintiff’s executor. The executor intended to resist the appeal. He therefore applied for and obtained an order allowing him to carry on the litigation “in like manner as [his testator, the plaintiff] might have done if she had not died” (at 736).

48 The executor lost the appeal. The Court of Appeal ordered the executor personally to pay the defendant’s costs of the entire litigation, ie, the entire costs of the appeal and the entire costs of the proceedings below. The executor appealed further to the House of Lords. He argued that he had merely continued the litigation on appeal in a representative capacity for his testator, as the appellant, and not in his personal capacity. Therefore, the executor submitted, the defendant’s costs should be paid out of the appellant’s estate and not by the appellant’s executor personally.

49 The House of Lords rejected the executor’s argument. Lord Cairns LC held that, by applying for the order to allow him to carry on the litigation, the executor had “adopted it ab initio” and therefore should be liable for the costs of the litigation ab initio (at 736). The headnote to Boynton thus bases the executor’s personal liability for costs on his procedural status as a substantive party to the litigation in all but name:

50 The rule thus stated is a rule of procedure which made it unnecessary for the court to consider whether the executor came within the wholly exceptional power to order a non-party to bear the costs of litigation (see Aiden Shipping Co Ltd v Interbulk Ltd [1986] AC 965; Symphony Group plc v Hodgson [1994] QB 179; DB Trustees (Hong Kong) Ltd v Consult Asia Pte Ltd and another appeal [2010] 3 SLR 542).

51 Be that as it may, cases which have followed Boynton have treated it as authority for a general, substantive rule of equity that an executor who adopts litigation commenced by his testator will be personally liable for an adverse costs order if the litigation fails. As such, Boynton has become a fundamental building block in subsequent cases for the two propositions at [30] above.

52 The rule in Boynton was extended from probate to insolvency in In re London Drapery Stores (1898) 2 Ch 684 (“London Drapery”). This is the earliest direct authority cited to me for the two propositions at [30] above.

53 In London Drapery, a company commenced litigation and lost. It was ordered to pay the defendant’s costs. While the litigation was pending, the company went into what ultimately became an insolvent liquidation. The issue on appeal was whether the defendant was entitled to priority for all of his costs or only for his post-liquidation costs. In the latter case, he would have no alternative but to prove for a dividend on his pre-liquidation costs.

54 In a brief judgment, Robert Wright J (as he then was) held that the defendant was entitled to priority for his costs ab initio (at 686):

The ratio of London Drapery comprises the two propositions at [30] above.

55 Cases in England – as well as in jurisdictions based on English law – continue to cite London Drapery for these two propositions to the present day.

56 In Re Movitex a company commenced litigation. It went into insolvent liquidation before judgment. The liquidators took advice from counsel and adopted the litigation. The company eventually lost the litigation. The defendants secured the usual order for costs to follow the event. On the authority of London Drapery, Mervyn Davies J held that the defendants’ costs were within the estate costs rule even though the liquidator had adopted the litigation rather than commencing it (at 495H):

57 The House of Lords has relatively recently endorsed the two propositions at [30] above. In Norglen Ltd (in liquidation) v Reeds Rains Prudential Ltd and others [1999] 2 AC 1 (“Norglen”), a company commenced litigation. It then went into insolvent liquidation. On two alternative grounds, the English Court of Appeal declined to order the company to furnish security for the defendants’ costs under the English equivalent of s 388 of the Act: (a) because the company had sufficient assets to meet any order for costs which might be made in the litigation; and (b) because the estate costs rule would accord super priority to any order for costs in favour of the defendant (at 20).

58 On appeal to the House of Lords, counsel for the defendants attacked the Court of Appeal’s second ground on the basis that the estate costs rule accords super priority only to post-liquidation costs. The defendants therefore argued that they should be awarded security for their pre-liquidation costs. Lord Hoffmann rejected this argument, following Boynton and London Drapery (at 21C):

59 Norglen is thus recent and high authority that the two propositions at [30] above continue to form part of English insolvency law.

60 The Australian cases, following the English cases, have established the two propositions at [30] above also as part of Australia’s insolvency law: see Hypec Electronics Pty Ltd (in liq) v Mead and others (2004) 50 ACSR 448 at [104] and Re Mendarma Pty Ltd (No 2) (2007) 61 ACSR 601 at [45].

61 The most relevant Australian authority for present purposes is Re Lofthouse; In the matter of Riverside Nursing Care Pty Ltd (subject to deed of company arrangement) [2004] FCA 93 (“Lofthouse”). In Lofthouse, a company became insolvent and appointed an administrator. The company then entered into a deed of company arrangement (“DoCA”) with its creditors. At the risk of oversimplification, a deed of company arrangement is analogous to a scheme of arrangement under s 210 of the Act, save that it does not require the court’s sanction in order to take effect.

62 The DoCA, as amended, established two trust funds with the administrator as trustee. The administrator was to distribute the trust funds to the company’s creditors in accordance with the terms of the trust deed and in composition of the company’s debts.

63 The company’s management had commenced litigation against a defendant before the administrator’s appointment. That litigation continued after his appointment and after the DoCA took effect. The court ultimately ordered costs against the company in that litigation. The administrator sought directions from the court as to how to deal with these costs in the administration and under the terms of the trust deeds.

64 Raymond Finkelstein J began his analysis by citing the English cases which I have analysed, amongst others. He derived from them the proposition that the liability of a company in liquidation under an adverse costs order forms part of the expenses of the winding up and is accorded super priority, whether the liquidator commences the litigation or adopts it, and whether the company is the plaintiff or the defendant (at [26]):

65 Finkelstein J’s decision in Lofthouse eventually turned on the construction of the terms of the trust deed rather than on a direct application of the estate costs rule. However, Lofthouse remains authority that the two propositions at [30] above form part of Australia’s insolvency law.

66 The Hong Kong cases, following the English cases, have established the two propositions at [30] above also as part of Hong Kong’s insolvency law.

67 In Typhoon 8 Research Ltd v Seapower Resources International Ltd and another [2002] 2 HKLRD 660 (“Typhoon 8”), a tenant took possession of premises under a lease. At the landlord’s request, the tenant paid the security deposit to the landlord’s parent company instead of to the landlord. As a result of the landlord’s breach, the landlord’s mortgagee terminated the lease prematurely. The tenant sued the parent company to recover the deposit. The tenant failed at first instance and appealed. Between the time of the first instance proceedings and the hearing of the tenant’s appeal, the parent company went in liquidation.

68 The Hong Kong Court of Appeal allowed the tenant’s appeal. Doreen Le Pichon JA held that the provisional liquidators had adopted the litigation after their appointment by actively resisting the tenant’s appeal. She also held that, as a result, the tenant’s costs were to be accorded priority ab initio (at [31]–[32]), citing Norglen with approval.

69 In Re Kenworth Engineering Ltd [2005] 2 HKLRD 97 (“Kenworth”), a company implementing a scheme of arrangement sought directions from the court. The scheme permitted certain arbitrations to continue in order to quantify the claims against the company. The terms of the scheme expressly accorded priority to post-scheme costs (at [47]). The company sought directions as to whether an award of costs against the company in one of those arbitrations was a post-scheme cost or was merely admissible to proof under the scheme for a dividend (at [14]).

70 Susan Kwan J held that the costs award was a post-scheme cost and therefore entitled to priority. Kwan J bore in mind, as part of the factual matrix in construing the scheme, how the costs award would have been treated in Hong Kong’s liquidation regime (at [30]). She held that the express provision in the scheme conferring priority on post-scheme costs was consistent with the position which would apply in a liquidation (at [48]), citing the leading English authorities including London Drapery and Norglen. From those cases, Kwan J extracted the following propositions (at [30]):

71 Both Typhoon 8 and Kenworth are authority that the two propositions at [30] above are part of Hong Kong’s insolvency law.

72 I now turn to consider Mr Seah’s arguments as to why, despite the overwhelming weight of Commonwealth authority, these two propositions should not be accepted as part of Singapore’s insolvency law. To deal with his submissions, however, it is necessary to extract a rationale from these cases and from ECRC Land for the estate costs rule.

73 In my view, the cases show that the rationale for the estate costs rule rests on two closely related principles. Both principles are two senses of the overarching concept of mutuality. The first principle goes to the liability to pay costs and the second goes to the extent of that liability.

74 The first principle is what I shall call the risk/reward principle. This principle dictates that those who stand ultimately to reap the rewards of litigation succeeding ought ultimately to bear the risks of that same litigation failing. The primary risk, of course, is the liability to pay costs to a successful opponent. The second principle is what I shall call the reciprocity principle. A party to litigation who succeeds is ordinarily awarded its costs ab initio. The reciprocity principle dictates that, by the same token, if that party fails in the litigation, it should ordinarily be held liable for its opponent’s costs ab initio. The reciprocity principle is an obvious by product of the risk/reward principle: a subsidiary reward of litigation and its primary risk is an order for costs ab initio.

75 The Court of Appeal in ECRC Land expressly relied on the risk/reward principle (at [66]; see also [68]): “where a company is insolvent and the liquidator has insufficient resources at his disposal to take a particular course of action, the individual who reaps the fruit of that action should be the one to shoulder the risk of taking it” [emphasis added]. So too in Baring Futures (at [46]).

76 Boynton is the ultimate authority for both the risk/reward principle and the reciprocity principle as the rationale for the estate costs rule. An important feature of Boynton is that the executor in that case was a co-defendant at first instance. As a result, the executor had had a costs order made in his favour at first instance against the main defendant. The executor was also the residuary legatee of the deceased. Thus, the executor had personally benefited from the award of costs in favour of the estate at first instance and, as the residuary legatee, stood personally to benefit from a costs order against the main defendant ab initio if the estate succeeded on appeal. Thus, Lord Cairns LC said (at 736):

77 Further, the risk/reward principle allows the court to look through matters of form to have regard to who in substance will benefit economically from the litigation and from any costs order in it. When a company goes into liquidation, the purpose of the limited liability which statute affords to those who reap its ultimate economic rewards has come to an end (see ECRC Land at [24] and [67]). In that situation, the risk/reward principle allows the court to have regard to the position of the creditors as the ultimate beneficiaries of the rewards of the litigation. Thus, Chan Sek Keong CJ held (at [67]):

78 This is not a novel proposition. The Court of Appeal in ECRC Land cited with approval (at [9]) the following passage in In re Trent and Humber Ship-Building Company (1869) LR 8 Eq 94 (“In re Trent”) which relies on the risk/reward principle to bring home the risks of post-liquidation litigation to the creditors:

79 Wright J in London Drapery cited Boynton and expressly adopted the reciprocity principle from Lord Cairns LC’s speech in the passage I have set out at [54] above. It is true that there is no mention of the risk/reward principle in London Drapery. But that does not mean that the risk/reward principle was not also a factor in Wright J’s decision. Lord Cairns LC had to state the risk/reward principle expressly in Boynton because the executor had been involved in the litigation in a representative capacity (as the executor on appeal), had been involved in the litigation in his personal capacity (as the second defendant at first instance) and had a direct pecuniary interest in the outcome of the litigation including in an award of costs (as the residuary legatee under the deceased’s will). In explaining why the executor should be held liable to pay costs personally, it was therefore necessary for Lord Cairns LC to cite the risk/reward principle, ie, to point out that the executor had a direct personal economic interest in the outcome of the litigation in his personal capacity. In London Drapery, on the other hand, the paying party was simply the plaintiff. His direct pecuniary interest in the litigation went without saying. London Drapery was therefore nothing more than the usual case where it is implicit (and therefore need not be articulated as in Boynton and in ECRC Land) that a party who stands to reap the ultimate economic rewards of litigation ought to bear the risks of the litigation.

80 It is true that the Court of Appeal in ECRC Land did not expressly rest its decision on the reciprocity principle. That is because it was the liquidators themselves who commenced the litigation. Therefore, all of the costs which the defendants incurred in that case were post-liquidation costs and thus indisputably within the scope of the estate costs rule. However, the Court of Appeal in ECRC Land cited with approval and relied on the English authorities which have established the reciprocity principle. Further, the entire tenor of the Court of Appeal’s judgment in ECRC Land shows that its primary concern was the disconnect between risk and reward in any litigation against a company in liquidation and the lack of true reciprocity in such litigation.

81 Both the risk/reward principle and the reciprocity principle underlie the indemnity principle. The indemnity principle in turn underpins the discretion to award costs. Both principles are so obvious in ordinary litigation between solvent parties that neither of them is ever articulated explicitly when exercising the discretion to award costs.

82 The difficulty of course is that the indemnity principle applies to all litigation. And it is merely a principle of liability, not of recoverability. The indemnity principle does not guarantee that a receiving party will recover in full its party and party costs. Quite the opposite. Insolvency risk is one of the inevitable risks of litigation. The indemnity principle accepts that every litigant runs the risk that it may not be able to recover an entitlement to costs from its opponent, whether for practical reasons (eg, the opponent is a foreign or a nominal party) or by reason of insolvency. A defendants’ risk of irrecoverable costs is mitigated to a limited extent by procedural provisions in O 23 of the Rules and in s 388 of the Act (specifically for insolvent companies) which allow a defendant to seek security for costs from a plaintiff. But those provisions are exceptional. The estate costs rule goes above and beyond these exceptional provisions by doing everything possible to guarantee that the opponent of a company in liquidation will recover its costs in full. On top of that, as ECRC Land has established, the common law fortifies that virtual guarantee by holding that a liquidator who breaches the estate costs rule is personally liable for any shortfall in the opponent’s costs recovery which is caused by the breach. Why does the estate costs rule add this virtual guarantee to the liability created by the indemnity principle and then fortify it with a rule of personal liability? It does so because insolvency engages the rationale for the estate costs rule in a fundamental way.

83 Once liquidation commences, the insolvency risk which is an inevitable feature of all litigation has eventuated. The company’s insolvency is no longer merely a possibility, a probability or even a certainty. It is now a fact.

84 As a result, the commencement of liquidation renders the risk of irrecoverability asymmetric. If the company succeeds, it will recover its substantive claim or counterclaim plus its own costs in full in the usual way from the opponent. But if the opponent succeeds, it will recover at most a tiny fraction of its claim or counterclaim and its costs as a dividend in the company’s liquidation. Simply put, the company now has nothing to lose. Going into liquidation has, in practical terms, relieved the company of substantially all of its litigation risk and deprived its opponent of substantially all of its litigation reward. That gives the company a significant tactical advantage both in pursuing the litigation and also in any negotiations to compromise the litigation.

85 This asymmetry is why a company which is impecunious does not have unfettered freedom to commence litigation without making provision for its opponent’s costs (ECRC Land at [72]):

86 As ECRC Land held, the position of a company in actual liquidation is a fortiori that of a company which is merely impecunious. This fundamental alteration of risk and reward justifies looking through the company and directly at those who will ultimately reap the rewards of the litigation, ie, the creditors.

87 The obvious question then is why the estate costs rule applies only when a company is in liquidation and not simply where a company is insolvent but has avoided liquidation through wile, guile or the indulgence of its creditors. Indeed, in that sense, it could be said that the estate costs rule operates arbitrarily and perhaps unfairly by having regard to form over substance. A company which is insolvent but not in liquidation is entirely outside the scope of the estate costs rule even though the two principles which comprise its rationale apply equally to it and to its shareholders. Thus, a litigant who secures an order for costs against an insolvent company which has somehow avoided liquidation will have to take what it can get from the company by way of recovery, whether by levying execution pre-liquidation or by winding up the company and filing a proof of debt post-liquidation. But if liquidation had intervened just a day before judgment, the estate costs rule dictates that the litigation opponent should secure super priority for its costs ab initio.

88 There are two reasons for this apparent anomaly. First, until liquidation actually happens, insolvency may be a possibility or even a probability but is not a certainty let alone a fact. It remains possible that the insolvency may be overcome such that the company’s creditors, including a receiving party under an adverse costs order, will have their debts paid in full. In that situation, there is certainly cause for upholding the risk/reward principle by requiring the company to give security for costs under s 388 of the Act. But there is no cause for disregarding the corporate form and the limited liability which it provides to its shareholders by intervening in the company’s affairs and overriding the company’s entitlement through its usual organs to pay its debts in the manner and in the order it chooses to do so. Put another way, for broader policy reasons, the law continues to uphold a company’s separate legal personality, limited liability and autonomy so long as it remains outside liquidation, no matter how hopelessly insolvent it may be. In that position, its liability to be ordered to furnish security for costs suffices to address the risk/reward and the reciprocity principles. Further intervention by subjecting the company to the estate costs rule would undermine the public policy behind establishing the corporate form.

89 Second, it is only post-liquidation that the court has the legal means to intervene directly in the company’s affairs in order to uphold the risk/reward principle and the reciprocity principle. It is only post-liquidation that the court gains control over the company through its supervisory jurisdiction in an insolvent liquidation and gains control over its liquidator, as a court-appointed office holder. Thus, although the risk/reward principle and the reciprocity principle are conceptually engaged as soon as a company apprehends insolvency, it is only post-liquidation that the court can give practical effect to these principles through a rule of priority such as the estate costs rule. Equally, it is only post-liquidation that the court can uphold the estate costs rule by exercising its supervisory jurisdiction over liquidators to impose personal liability on a liquidator who breaches the estate costs rule and thereby causes a receiving party to suffer a shortfall in its costs recovery.

90 I now turn to consider Mr Seah’s substantive arguments on the questions posed to me.

91 Mr Seah’s first argument ¹⁶ is that the liability of a company in liquidation under a costs order is not a liquidation expense (see [39] above) to the extent that that liability is intended in part to indemnify the receiving party for its pre-liquidation costs. The argument proceeds as follows. The estate costs rule operates as a common law supplement to the statutory priority accorded to liquidation expenses under s 328(1)(a) of the Act (ECRC Land at [9]). It does so by according super priority to a particular type of liquidation expense. A liquidation expense must, by definition, be one which arises after the liquidation. ¹⁷ Therefore, if a court makes a costs order which is intended in part to indemnify the receiving party for its pre-liquidation costs, that part of the liability is not a liquidation expense and is not entitled to any priority whatsoever under s 328(1)(a). ¹⁸ A fortiori, therefore, that liability cannot be entitled to super priority under the estate costs rule.

92 I consider that this argument has been concluded against Mr Seah by the ratio in ECRC Land. That is so even though the costs ordered against the company in that case were entirely in respect of the receiving party’s post-liquidation costs. In any event, for the reasons which follow, I reject this argument even on principle.

93 Section 328(1)(a) of the Act provides as follows:

94 Mr Seah cites Halsbury’s Laws of Singapore vol 13 (Butterworths Asia, 2003). At para 150.460, the author draws a distinction between pre-liquidation creditors and post-liquidation creditors, and points out that s 328(1)(a) accords priority only to the latter:

Mr Seah relies on this passage to argue that the proposition at [30(b)] above contradicts s 328(1)(a) of the Act.

95 Mr Seah also relies on Barings Futures at [51] where the Court of Appeal held that the only post-liquidation liabilities are liquidation expenses. ¹⁹

96 Mr Seah is of course correct that a liquidation expense within the meaning of s 328(1)(a) must, by definition, arise post-liquidation. But this does not take his argument very far. His argument succeeds only if he can also establish that the liability which arises when a court makes a costs order can be resolved in a legal sense into a pre-liquidation component and a post-liquidation component.

97 I do not accept that it can. In my view, a paying party’s liability under a costs order arises only at the moment when the court actually exercises its discretion and makes the order. And at that point, the liability arises in the full amount of the costs awarded, whether the quantum is fixed at that time or by subsequent taxation. The liability cannot be resolved in any legal sense into a pre-liquidation component and a post-liquidation component. I say that for three reasons.

98 First, until the court exercises its discretion to award costs, which party will be paying costs and which party will be receiving costs is doubly unknowable. The outcome of the litigation itself is unknowable. Therefore, the event which the costs will ordinarily follow is also unknowable. Compounding that uncertainty, it is also unknowable whether the court will exercise its discretion to order the costs of the litigation to follow the event, make no order as to costs or even (exceptionally) award costs against the successful party or against a non-party. The only way that this argument can succeed therefore is by positing that, as soon as litigation commences, every litigant immediately becomes subject to a notional liability to pay costs to every opposing litigant. The liability of each litigant then increases gradually day by day throughout the litigation as the opposing litigants incur additional costs as between solicitor and client. This proliferation of accumulating liabilities is an unnecessarily complicated way to conceptualise the operation of a costs order. I do not consider that it describes accurately how a liability to pay costs under a costs order arises.

99 It is of course correct that a receiving party will incur solicitor and client costs day by day, from the beginning of the litigation until it ends. It is also true that, where a corporate litigant goes into liquidation while litigation is ongoing, its opponent will have incurred some of its solicitor and client legal costs in point of fact pre-liquidation and some post-liquidation. But I do not accept that it is conceptually correct to map the receiving party’s solicitor and client legal costs onto the liability which a costs order imposes as between party and party. That is because party and party costs are not awarded in a manner directly referable to the receiving party’s solicitor and client costs. Instead, party and party costs are awarded once and for all when the order is made, by way of an admittedly partial (see [23] above) indemnity to the receiving party against the total and undifferentiated quantum of reasonable legal costs which it has reasonably incurred in the litigation, from beginning to end.

100 Second, this conception of a costs order necessarily posits that when a court makes a costs order, it does no more than recognise and give effect to the paying party’s pre-existing liability to pay costs to the receiving party. But that is not the jurisprudential basis of a costs order. The basis of the liability to pay costs is the court’s exercise of a statutory discretion in the present, not the court’s recognition of a liability in the past. Until that discretion is actually exercised, no liability to pay costs exists. The basis of a paying party’s liability under a discretionary costs order (as opposed to an order for costs as of right) is not historical. The parties’ historical conduct, if it has increased or diminished the quantum of the solicitor and client costs paid by the receiving party, may be one of the factors on which the court’s discretion to award costs is exercised. But that is all it is: a factor. It is not the basis of the order.

101 That is to be contrasted with the court’s decision on substantive liability. In most cases, a decision on substantive liability is indeed the court’s present recognition by way of a judgment of a historical legal truth. When a court holds a defendant liable or not liable to a plaintiff, that involves the court looking back in time and deciding whether the defendant's conduct in the past was or was not sufficient, under our substantive law as applied in accordance with our procedural law, to generate a substantive liability to the plaintiff. The court then recognises that legal truth from the past by its judgment in the present. That is quite unlike the way in which a costs order operates. It is for this reason that interest on damages ordinarily runs from the date on which the plaintiff’s cause of action arose (see s 12(1) of the Civil Law Act (Cap 43, 1999 Rev Ed)) whereas interest on costs runs only from the date of the costs order (see O 59 r 37 of the Rules). A receiving party under a costs order is not kept out of its money until the day the costs order is made.

102 Third, if making a costs order is conceived as merely recognising a pre-existing liability to pay costs which was imposed somehow upon each of the receiving party’s opposing litigants when the litigation commenced and which has grown over time from day to day, that necessarily implies that making a costs order operates to extinguish the liability of all of the opposing litigants to pay costs, save only for the paying party’s liability. It also necessarily implies that, if the court exercises its discretion to compel a non-party to the litigation to pay the receiving party’s costs (on the authorities cited at [50] above) or to make no order as to costs, the costs order extinguishes the liability of all of the receiving party’s opponents to pay costs to the receiving party and to all of the other opponents. That is not at all how a costs order works. A costs order by its terms extinguishes no liability, whether expressly or impliedly. A costs order by its terms only creates a liability.

103 In my view, therefore, a paying party’s liability under a costs order arises only at the time the order is made and arises in the full quantum of the costs awarded, whether that quantum is fixed then or by taxation subsequently. The liability cannot therefore, in any legal sense, be resolved into a pre-liquidation component and a post-liquidation component any more than the substantive claim in the dispute can be resolved into a pre-liquidation component and a post-liquidation component. To attempt to do so would be wholly artificial. That is so even though the costs order is awarded in a factual sense as an indemnity to the receiving party for both its pre-liquidation and post-liquidation costs.

104 In support of his submission that a costs order can be conceptualised as a liability coming into existence from the moment litigation commences, Mr Seah relies on the UK Supreme Court’s decision in In re Nortel GmbH (in administration) and related companies [2013] 3 WLR 504 (“Nortel”). In that case, Lord Neuberger of Abbotsbury PSC held that an order for costs is provable as a contingent debt in a liquidation (at [88]–[89]):

105 Once again, Nortel takes Mr Seah only so far. The case establishes only that a litigant comes under a contingent liability to pay costs from the moment the litigation commences. It does not go so far as to establish that a litigant comes under an actual liability to pay costs from that time. Indeed, that cannot be the case. For the reasons I have already given (at [96]–[103] above), an actual liability for costs does not and cannot arise unless and until the court exercises its discretion to award costs. Establishing that an award of costs is provable in a liquidation as a contingent liability therefore does not address the point that the actual liability to pay costs comes into existence only when the court exercises its discretion to make the order.

106 Further, Nortel’s conceptualisation of a costs order as giving rise to a contingent liability from the moment the litigation commences also runs into the same fundamental objection I adverted to earlier about the nature of a costs order (see [102] above). A costs order does not in any way extinguish a contingent liability any more than it extinguishes an actual liability. So too, a contingent liability to pay costs is even less capable of being resolved in a legal sense into a pre-liquidation component and a post-liquidation component than an actual liability to pay costs.

107 This is not, of course, to say that Nortel was wrongly decided. The result of characterising the potential liability of a company in liquidation to pay costs as a contingent liability is to render that liability provable in the liquidation when it would otherwise not be. That is undoubtedly a pragmatic solution to the specific issue which confronted the court in Nortel. Holding that a liability to pay costs is not even a contingent liability for the purposes of the proof of debt regime would force a plaintiff in litigation with a company to face a stark choice when the company goes into liquidation. The choice is between: (a) abandoning any attempt to recover the legal costs of the litigation, even on a dividend, and proving in the liquidation only for the substantive claim; or (b) expending more time and resources in pursuing pointless litigation for the sole purpose of securing an order for costs, thereby rendering the costs provable in the liquidation. That would defeat the policy in insolvency law of sublimating disputes about a company’s pre-liquidation rights and obligations as far as possible to the proof of debt regime in order not to deplete the company’s limited resources through litigation. That would also defeat the policy of our procedural law which discourages litigation over costs alone (see s 29A(1)(c) read with para 3(f) of the Fifth Schedule to the SCJA).

108 Leaving aside these practical concerns, however, and considering the issue in Nortel from a purely conceptual perspective, I find great force in the analysis of Mummery LJ in Glenister v Rowe [2000] Ch 76 at 84B–85B and Thorpe LJ in the same case at 85D–85F that the commencement of litigation creates merely a risk of liability under a costs order and therefore creates neither a contingent liability nor an actual liability.

109 For these reasons, therefore, I hold that the entirety of the liability under a costs order against a company in liquidation arises post-liquidation and is therefore a liquidation expense within the scope of s 328(1)(a) of the Act. That is so even if the costs order is intended in part to indemnify the receiving party for its pre-liquidation costs.

110 Mr Seah argues next that restricting the scope of the estate costs rule so that only post-liquidation costs come within its scope is consistent with Singapore’s judicial management regime.

111 Under s 227I(1)(b) of the Act, a judicial manager is personally liable on any contract of the company which he adopts while carrying out his functions. The judicial manager can, however, disclaim that liability by notice to the counterparty under s 227I(2) of the Act. Under s 227I(3), no act or omission by the judicial manager within the first 28 days of the judicial management order is to be taken as adopting a contract.

112 For this argument, Mr Seah relies on the High Court decision of Kotjo Johanes Budisutrisno v Ng Wei Teck Michael and others [2001] 2 SLR(R) 784 (“Kotjo”). In Kotjo, the judicial managers of a company at first adopted an employee’s employment contract and then disclaimed personal liability under it. Choo Han Teck JC (as he then was) observed in obiter dicta that a judicial manager who adopts a contract under s 227I is not liable for breaches of contract which occurred before he adopts it (at [15]):

113 Relying on Kotjo, Mr Seah contends that it is therefore consistent with the judicial management regime to hold that only post-liquidation costs are within the scope of the estate costs rule. This is because a liquidator too, should not be exposed to personal liability for the company’s pre-liquidation acts and omissions in the conduct of the litigation when the liquidator has no knowledge or control over those acts or omissions.

114 In my view, Mr Seah’s reliance on Kotjo and the analogy which he draws between a judicial manager adopting a contract under s 227I of the Act and a liquidator adopting litigation is inapposite for four reasons.

115 First, the rule in s 227I of the Act is a primary rule of personal liability. It is not a rule of priority. In contrast, the estate costs rule is a primary rule of priority. It is true that the estate costs rule is enforced by a secondary rule of personal liability. Under the secondary rule, a liquidator who breaches the estate costs rule is personally liable for any shortfall in the opponent’s costs recovery which is caused by the breach. Nevertheless, it remains the case that it is inapposite to draw an analogy between a primary rule of personal liability and a primary rule of priority.

116 Second, the subject matter of s 227I is a judicial manager’s personal liability for a breach of contract by the company. Liability for breach of contract is a species of liability for a civil wrong. Further, because it is liability imposed upon one legal person (the judicial manager) for a civil wrong committed by another legal person (the company), it is also a species of vicarious liability. The subject-matter of the estate costs rule has nothing to do with wrongs, civil or otherwise. The subject-matter of the estate costs rule is an order for the costs of litigation. These costs are awarded on the indemnity principle and are, of course, intended to serve as compensation for the successful party’s costs of litigation. But an order for costs is juristically best analysed as compensation ex gratia for having been sued without cause, not compensation for any type of civil wrong. If the juristic character of an order of costs was compensation for a civil wrong, the costs would be due as of right rather than being awarded in the court’s discretion (see [96]–[103] above). An order of costs as compensation for the civil wrong of being sued without cause is properly the subject matter of the tort of malicious prosecution. Further, although the estate costs rule is enforced by a secondary rule of personal liability, the secondary rule has nothing to do with imposing vicarious liability on the liquidator for the company’s breach of contract. The personal liability imposed under the secondary rule is direct liability imposed for the liquidator’s own breach of the estate costs rule. It is therefore inapposite to draw an analogy between a statutory rule of vicarious liability for a civil wrong and a common law rule of priority for ex gratia compensation.

117 Third, even in so far as it pertains to personal liability, s 227I of the Act does not focus on the distinction between pre-liquidation and post-liquidation personal liability. Instead, it focuses on the distinction between pre-adoption and post-adoption personal liability. That concept is alien to the estate costs rule. There is no statutory provision which gives a liquidator a window within which to consider whether to adopt or disclaim litigation. And none of the cases which establish the estate costs rule at common law have suggested that there is any such window. It is therefore inapposite to draw an analogy between a rule which turns on a statutory distinction between pre-adoption and post-adoption personal liability to answer questions which have to do with the pre-liquidation and post-liquidation priority accorded by the estate costs rule.

118 Finally, a judicial manager and a liquidator have completely different roles and duties to be discharged for completely different purposes. The roles and duties of a judicial manager are multi-faceted. One of the purposes of appointing a judicial manager is to rehabilitate a distressed company as a going concern, for the benefit of all stakeholders (Re Genesis Technologies International (S) Pte Ltd [1994] 2 SLR(R) 298 at [6]). A judicial manager is therefore empowered to continue the company’s business. That includes causing the company to perform its existing contracts. The statutory rule in s 227I addresses the risk which the contractual counterparty faces in that situation by making the judicial manager personally liable under the contract, if he adopts it. The gloss which Kotjo put on the statutory rule ensures that a judicial manager is not deterred from adopting a contract by the risk of potentially unlimited personal liability for the company’s pre-adoption breaches of the contract, when those are beyond the judicial manager’s knowledge or control.

119 A liquidator’s roles and duties are completely different. Her foremost duty is to maximise the realisation of the company’s assets and to distribute the net proceeds of those realisations to the company’s creditors with all practicable speed (Korea Asset Management Corp v Daewoo Singapore Pte Ltd (in liquidation) [2004] 1 SLR(R) 671 (“Korea Asset Management”) at [36]). She is thus not expected, let alone permitted, to take on any personal risk to rehabilitate the company as a going concern. That is why, unlike a judicial manager, a liquidator is prohibited from continuing the company’s business unless it is necessary for beneficial winding up (see s 272(1)(a) of the Act). There is no policy imperative which encourages a liquidator to take on risk or which points towards ensuring that she is not deterred from doing so. It is therefore inapposite to draw an analogy between a judicial manager adopting a pre-judicial management contract and a liquidator adopting pre-liquidation litigation.

120 Mr Seah’s next argument is that limiting the scope of the estate costs rule to a receiving party’s post-liquidation costs would achieve a result that best takes into account the interests of all the stakeholders in a winding up. He rests this argument on four limbs: ²⁰

(a) Limiting the estate costs rule to the receiving party’s post-liquidation costs strikes the right balance between the interests of a receiving party and the interests of the company’s general body of creditors.

(b) It is unjust for a receiving party to secure the windfall of super priority for its costs ab initio under the estate costs rule simply because judgment in the litigation is by happenstance rendered after the company has gone into liquidation rather than before.

(c) Not limiting the estate costs rule to post-liquidation costs will have a chilling effect on a liquidator litigating in the name of a company for the benefit of creditors, causing liquidators to abandon even meritorious litigation.

(d) Where a company in liquidation is the unsuccessful defendant in litigation, the plaintiff should not be accorded priority for its costs but should instead be left to prove for its costs together with its principal claim.

121 I do not accept any of these limbs of this argument. I deal with each limb in turn.

122 The first limb of the argument on justice and fairness is that the balance which the estate costs rule has struck between the interests of the receiving party and the interests of the company and its creditors leans too far in favour of the receiving party and against the creditors. ²¹ As Mr Seah puts it: “By according priority to costs incurred by the counterparty ab initio even where an action is subsequently continued or adopted by the liquidator, the creditors are in fact penalised for risks and decisions not taken by them and/or for their benefit.” [emphasis in original].

123 I cannot accept this argument as a matter of precedent. That is because the ratio of ECRC Land binds me and concludes this argument against the liquidator. The Court of Appeal in that case considered normatively the appropriate balance to be struck between the competing policies and concluded that the estate costs rule struck that balance (at [90]):

124 I cannot accept this argument even as a matter of principle. The risk/reward principle and the reciprocity principle are equally engaged regardless of whether a liquidator commences or adopts litigation. Mr Seah suggests that the English case of In re Lundy Granite Company; Ex parte Heavan (1871) LR 6 Ch App 462 (“Lundy Granite”) is authority for a principled basis on which to redress the imbalance against a receiving party and in favour of a company and its creditors. The general rule is that liabilities arising from pre-liquidation contracts are not liquidation expenses and are merely debts provable in the liquidation. The principle derived from Lundy Granite is a specific exception to this rule. According to Lord Hoffmann in In re Toshoku Finance UK plc [2002] 1 WLR 671 (“Toshoku”) at [29]:

Mr Seah submits that the estate costs rule can strike a more appropriate balance between the parties’ interests if a costs order in litigation which the liquidator adopts is held to fall within the Lundy Granite principle.

125 In In re Oak Pits Colliery Company (1882) 21 Ch D 322, Lindley LJ explained that the rationale underpinning the Lundy Granite principle is that a liquidator ought to pay as a liquidation expense any liabilities which arise from using the company’s property for the purposes of the liquidation (at 330–331):

126 In Toshoku, Lord Hoffmann explained that the effect of the Lundy Granite principle is not to convert the liability into a liquidation expense, but merely to treat it as though it were by an equitable fiction (at [27]):

127 In my view, the Lundy Granite principle is poor authority to support a principled basis on which to redress any imbalance in the balance which the estate costs rule has struck between the competing underlying interests. I begin by noting that I do not consider that the principle to be derived from Lundy Granite can be stated as broadly as Lord Hoffmann did in Toshoku. Lundy Granite involved a competition between a liquidator and the company’s landlord under a pre-liquidation lease. The liquidator adopted the lease for the benefit of the liquidation. The landlord was held entitled to retain the benefit of distress which he had levied pre-liquidation, but only to discharge the rent which had fallen due for the leased property post-liquidation.

128 In my view, the true principle to be extracted from Lundy Granite is this: where a pre-liquidation contract creates a recurring liability which continues post-liquidation and the liquidator takes the benefit of that contract for the purposes of the liquidation, it is just and equitable to treat that part of the recurring liability which accrues post-liquidation and which is necessary to secure that benefit as though it were a liquidation expense; but any part of the recurring liability which accrued pre-liquidation remains merely a provable debt (cf Roy Goode, Goode on Principles of Corporate Insolvency Law (Sweet & Maxwell, 5th Ed, 2018) at para 8-38).

129 Thus, on the facts of Lundy Granite, it was entirely just and equitable that the liquidator who retained possession of the leased property post-liquidation for the purposes of the liquidation should be required to treat the rent which accrued during that post-liquidation period as though it were a liquidation expense. If nothing else, the liquidator would have had to enter into a post-liquidation lease elsewhere and pay rent to a different landlord to store the goods there for the benefit of the liquidation. That rent would undoubtedly have been a liquidation expense. The Lundy Granite principle, as I have put it, therefore leaves the general body of unsecured creditors no worse off, even though it accords priority to one part of the debt of one particular unsecured creditor under a pre-liquidation contract.

130  A contract within the Lundy Granite principle is quite different from a costs order within the estate costs rule. A contract within the Lundy Granite principle is a contract created pre-liquidation. It creates a liability which binds the company regardless of liquidation. The liability falls contractually due for payment periodically, both pre-liquidation and post-liquidation. The recurring liability is divisible and can be mapped in a linear temporal relationship to the benefits accruing to the company under the contract. In the case of a lease, that mapping can be done by prorating the monthly liability for rent right down to the day.

131 On the other hand, for the reasons I have already given, a costs order within the estate costs rule is created in its entirety post-liquidation. It operates once and for all when the order is made. It creates a liability which falls due for payment immediately. The company’s obligation is therefore to pay the full amount of the costs awarded under the order as soon as it is made. The company receives no benefit in return for paying the costs ordered (other than simply discharging an existing liability) because ex hypothesi, the company has already lost the litigation. Further, even the risk of having to pay costs which a company undertakes in the hope of securing the reward of a judgment in its favour in litigation results in no benefit to the company whatsoever until there is actually a judgment in the company’s favour. In addition, the character of a costs order as a (partial) indemnity makes the liability which it creates legally indivisible. As such, it is not conceptually possible to map that liability in a linear temporal (or indeed any other) relationship into a pre-liquidation component and a post-liquidation component. That is so even though it may be possible as a practical matter to apportion a just estimate of the costs awarded under the order to the receiving party’s pre-liquidation and post-liquidation solicitor and client costs.

132 I therefore do not accept Mr Seah’s argument that the Lundy Granite principle offers a principled basis for redressing an imbalance between the competing interests under the estate costs rule, even if ECRC Land left it open to me to find that such an imbalance existed.

133 The second limb of the argument on justice and fairness is that failing to draw a distinction between litigation which a liquidator commences and litigation which a liquidator adopts leads to arbitrary results including the possibility of a receiving party securing a windfall of full costs recovery by happenstance.

134 For example, if a liquidator is appointed even one minute after the court issues its judgment and makes a costs order against a company, the receiving party will have to prove in the liquidation for both the judgment sum and the costs ab initio. Inevitably, the opponent will have to be satisfied with a dividend on both sums. However, if the liquidator had been appointed and had adopted the litigation even one minute before the very same court issued the very same judgment with the very same costs order in the very same litigation, the receiving party would be entitled to super priority for the costs ab initio and would prove in the liquidation only for the judgment sum. Mr Seah submits that it is arbitrary for this substantial difference in the recoverability of costs to turn on what can be as little as a minute’s difference. To do so would be “instinctively wrong”. ²²

135 ECRC Land, however, establishes that this result is not arbitrary. It is the intended result of the balance which the estate costs rule has appropriately and deliberately struck in regulating the underlying competing interests with a bright line rule. Every bright line rule produces seemingly arbitrary results at its margins. That does not detract from the utility of the rule in the general run of cases falling within its scope or from the justice of the principled rationale which underlies it. That rationale, as I have held, is equally engaged in litigation which a liquidator commences and litigation which she adopts.

136 Further, even the result which the estate costs rule produces at its margins is not unjust or unfair as between the interests of a receiving party and the interests of a company and its creditors. The liquidator has actual knowledge of the company’s financial state through the statement of affairs and will have taken control of the company’s books of account and other records (ECRC Land at [79] and [81]). Even if that knowledge is imperfect and incomplete because of the directors’ poor record-keeping pre-liquidation or their failure to cooperate post-liquidation, it is far better than any knowledge which the receiving party might have. As between a receiving party and a liquidator, therefore, the liquidator is almost invariably in the better position to arrive at a considered decision as to how to proceed. That is the case both when the liquidator is considering whether to commence litigation and when she is considering whether to adopt litigation. The risk/reward and the reciprocity principles are fully engaged in both situations.

137 The third limb of the argument on justice and fairness is that the estate costs rule will have a chilling effect on liquidators’ willingness to litigate in the name of the company for the benefit of the company’s creditors if it accords super priority even to pre-liquidation costs. This is a particular concern since the liquidator’s statutory discretion to bring or defend legal proceedings under s 272(2)(a) of the Act is apparently unfettered (cf s 144(1)(e) of the IRDA).

138 I do not accept this argument for two reasons. First, the estate costs rule is no more than a rule of priority. It merely accords to a costs order super priority out of the assets of a company in liquidation. It says nothing about the consequences which ensue if the company’s assets are insufficient to pay the receiving party’s costs in full even with the super priority. The estate costs rule thus does not in itself impose any incentive or disincentive on a liquidator litigating in the name of the company for the benefit of creditors. As a mere rule of priority, it cannot have a chilling effect on liquidators.

139 Second, in so far as estate costs rule does cause liquidators to think carefully before litigating on behalf of the company – no matter how meritorious the litigation – that is not an unintended or undesirable consequence. That is, in fact the intended and desired consequence of the balance which ECRC Land has struck between the competing underlying interests. As the Court of Appeal held, if a liquidator considers it consistent with her duties to exercise her power to bring or defend proceedings under s 272(2)(a) of the Act when the company’s assets appear to be insufficient to satisfy an adverse costs order, the liquidator ought to seek an indemnity from the creditors for the costs of the litigation (ECRC Land at [66], [68], [79] and [84]). The Court of Appeal recognised expressly that, if creditors were not forthcoming with an indemnity, this may result in the liquidator having to abandon a meritorious claim. Yet, the Court of Appeal deemed this to be a “justifiable consequence of the creditors’ refusal to provide an indemnity” (ECRC Land at [64]). If the position were otherwise, the very purpose of the estate costs rule would be undermined.

140 Mr Seah’s “chilling effect” argument does, however, reveal the liquidator’s real concern, and the concern which has motivated this application. That concern is with the secondary rule of personal liability – which ECRC Land has established as a part of Singapore’s insolvency law – which holds a liquidator who breaches the estate costs rule personally liable for any shortfall in the opponent’s costs recovery which is caused by the breach. It is the rule of personal liability, and not the estate costs rule itself, which is capable of having a chilling effect on liquidators’ willingness to litigate in the name of the company. But the questions posed to me on this application raise issues relating only the scope of the estate costs rule, not the scope of the rule of personal liability for breach of the estate costs rule.

141 I nevertheless pause at this stage to make some observations on the rule of personal liability. There are three situations in which a company which is in liquidation and involved in litigation may find itself unable to discharge in full its liability under a costs order in that litigation: (a) where the company had sufficient assets to pay the costs in full at the outset of the liquidation but does not now because the liquidator made payments out of the company’s assets after the court made the costs order; (b) where the company had sufficient assets to pay the costs in full at the outset of the liquidation but does not now because the liquidator made payments out of the company’s assets before the court made the costs order; and (c) where the company did not have sufficient assets at the outset of the liquidation to pay the costs in full.

142 ECRC Land is an especially egregious example of the first situation. The liquidators in that case caused the company to commence substantial litigation against its directors on multiple causes of action (at [3]). The litigation largely failed. The company was ordered to pay substantially all of the defendants’ costs (at [4]). The liquidators conceded that they were bound by the estate costs rule to accord super priority to the company’s liability under the costs order (at [18]). The defendants’ solicitors wrote to the liquidators twice to remind them of the effect of the estate costs rule (at [6]). The company appealed against the trial judge’s decision. The appeal failed. The company was once again ordered to pay the defendants’ costs, this time for the unsuccessful appeal as well (at [4]).

143 Throughout the litigation, the liquidators had substantially depleted the company’s assets by paying professional fees and disbursements to both themselves and to the company’s solicitors (at [6]). The defendants sought and secured an order from the trial judge expressly ordering the liquidators to comply with the estate costs rule and expressly requiring the liquidators to pay the defendants their costs in priority to all other claims and expenses, including the liquidators’ and the company’s solicitors’ fees and disbursements (at [4]). Despite all of these orders, the liquidators made two further payments of legal fees to the company’s solicitors. And they compounded their breach by failing to have the company’s solicitors’ legal fees taxed as they were obliged to (at [6]).

144 The direct result of these payments was that the defendants faced a shortfall in recovering their costs amounting to about $190,000 (at [5]–[7]). The defendants applied for an order holding the liquidators personally liable to the defendants for this shortfall. The liquidators accepted that all of the payments to themselves and to the company’s solicitors were made in breach of the estate costs rule (at [18]). At first instance, the liquidators were held personally liable to make good the shortfall, but only up to an amount equivalent to the payments which the liquidators had made to themselves, ie, excluding the amounts they had paid to the company’s solicitors (at [7]–[8]). On appeal, the liquidators were held personally liable to make good the shortfall up to an amount equivalent to all of the payments which the liquidators had made, ie, including the amounts they had paid to the company’s solicitors (at [30] and [54]). The basis for this liability is the court’s supervisory jurisdiction over liquidators (see [41] above). The defendants were thus able to recover the entire sum due under the costs order in their favour.

145 I make three points about ECRC Land before I consider the second and the third situations.

146 First, the issue before the Court of Appeal in ECRC Land was “whether [the liquidators] should be held personally liable for the [defendants’] outstanding shortfall which has resulted from [the liquidators’] breach of the estate costs rule” [emphasis in original omitted] (at [21]; see also [15]). The ratio of ECRC Land is therefore that a liquidator who breaches the estate costs rule is personally liable for any shortfall in the opponent’s costs recovery which is caused by the breach. Thus, a liquidator’s personal liability is only to the extent of the secondary insolvency which her breach of the estate costs rule has caused (at [81]). The liquidator’s personal liability is therefore subject to two limits: (a) it cannot exceed the amount of the receiving party’s shortfall; and (b) it cannot exceed the amount which the liquidator has paid to the company’s creditors (including to the liquidator herself) in breach of the estate costs rule (ECRC Land at [91(a)]).

147 Second, the species of personal liability which is the ratio of ECRC Land rests on two prerequisites. One prerequisite is that the liquidator has breached the estate costs rule. The second prerequisite is that the receiving party’s shortfall is caused by the liquidator’s breach. If these prerequisites are absent, then arguably the ratio of ECRC Land is not engaged. Further, because of the egregious nature of the breach in ECRC Land and because the breach was undisputed (at [18]), it was not necessary for the Court of Appeal to consider whether there is a third prerequisite: fault. In other words, it is an open question whether the secondary rule of personal liability established by ECRC Land is a species of strict liability. It is therefore open to a liquidator to argue, even after ECRC Land, that she is not personally liable within the ratio of ECRC Land because the breach of the estate costs rule occurred even though she acted in good faith or with reasonable care (whichever might be the appropriate test for fault in this context (see Deputy Commissioner of Taxation v Tideturn Pty Ltd [2001] NSWSC 217 at [8] and [16])).

148 Third, in so far as the liquidators’ breach of the estate costs rule in ECRC Land arose from their payments of debts due from the company to third parties (ie, not to the liquidators themselves), the payments were directly connected to the litigation in which the costs order was made. Thus, in ECRC Land, the third-party debts which the liquidators paid were paid to the company’s own solicitors for the legal fees which the company had incurred to sustain the very litigation in which the adverse costs order was made (at [25] and [56]). It is perhaps understandable, therefore, that the Court of Appeal should view with particular disapproval the liquidators’ decision to pay the company’s own solicitors for representing the company in ultimately fruitless litigation ahead of the company’s successful opponent in the very same litigation and in breach of the estate costs rule. It is by no means assured that the liquidator would have been held personally liable if, for example, the defendants’ costs recovery shortfall in ECRC Land had been caused by the liquidator making payments in the bona fide performance of their duties as liquidators to persons other than the company’s solicitors in the failed litigation, eg, preferential creditors such as employees, the Central Provident Fund or the tax authorities.

149 I now turn to consider the second situation I described at [141] above. Whether the ratio in ECRC Land extends to the second situation depends on how one conceives the temporal scope of the estate costs rule. If the temporal scope of the estate costs rule is conceived narrowly, it is a rule of immediate priority. It requires only that a liquidator pay costs which have just fallen due from the company to a receiving party before she pays any other debts of the company. On this narrow conception, there can be no breach of the estate costs rule if there are no costs due from the company at the time when the liquidator pays a particular debt. The liquidator cannot breach a duty to accord priority to a liability to pay costs over other debts when no such liability exists at the time those other debts are paid.

150 But ECRC Land adopts a broader conception of the temporal scope of the estate costs rule. In ECRC Land, the liquidators paid close to $360,000 to the company’s solicitors while the litigation was ongoing without making any provision for the company’s potential future liability under an adverse costs order. The Court of Appeal treated this as a breach of the estate costs rule, even though the payments took place before any adverse costs order was made against the company (ECRC Land at [25]). The Court of Appeal gave two reasons for this broader conception. First, when a liquidator litigates in the name of a company, he “pledges the assets of the company for the costs of the action”, and “[t]he liquidator therefore cannot use these assets to pay claims of a lower priority before the outcome of the litigation is determined” (ECRC Land at [83]). If the liquidator does so, that is a breach of the estate costs rule thus conceived. Second, the Court of Appeal saw the estate costs rule as extending to impose a duty on a liquidator to make a prospective assessment – before deciding to litigate, and also perhaps continuing throughout the course of the litigation – of the risk that the company will be unable to satisfy in full a potential future adverse costs order in that litigation, bearing in mind the liquidator’s knowledge of the company’s actual financial position (ECRC Land at [79] and [81]). A failure to make that assessment or to act upon it by making full provision for an adverse costs order – either by setting funds aside against that liability or by securing an indemnity from creditors in respect of that liability – is a breach of the estate costs rule thus conceived (ECRC Land (at [57] and [81]). The liquidator will be held personally liable for the loss which the receiving party suffers by reason of this breach.

151 As for the third situation, the Court of Appeal in ECRC Land expressed the view that the “rationale for imposing personal liability should apply equally to situations where a liquidator commences proceedings though the company’s coffers are completely empty and it has no assets to satisfy either its own or the defendant’s legal costs” [emphasis in original] (at [69]). ECRC Land thus concludes with this advice to a liquidator who finds herself in the third situation (at [91(c)]):

152 I point out only that what was said of the liquidator’s personal liability in the third situation in ECRC Land was obiter. The liquidators in ECRC Land did not choose to litigate when the company’s coffers were empty. Instead, they chose to litigate when the company had sufficient assets to satisfy the defendants’ entitlement to costs in full, but did not have sufficient assets to pay the defendant’s costs as a direct result of the liquidators’ decision to pay the company’s debts in breach of the estate costs rule on its broader temporal conception (ECRC Land at [10]). It is therefore open to a liquidator to argue that ECRC Land is not binding authority in the third situation. The Court of Appeal itself appears to allude to this possibility by stating its conclusion on this issue tentatively rather than prescriptively: “a liquidator who omits to do so in a future case may well be held personally liable” [emphasis added] ( ECRC Land at [91(c)]; see also [69]).

153 A liquidator in the third situation could also argue persuasively that the two prerequisites (and the third possible prerequisite) (see [147] above) for liability in the ratio of ECRC Land are absent. First, it is arguable that there can be no breach of the estate costs rule in the third situation. This liquidator finds herself in charge of a company with empty coffers even before the liquidator makes a single payment to any creditor. Second, it is arguable that there is no causation in the third situation. The proximate cause of the receiving party’s shortfall in the third situation is not any payment by the liquidator in breach of the estate costs rule but simply the liquidator’s decision to litigate. Even on the broadest of conceptions of the estate costs rule, it is difficult to characterise a decision to litigate as a breach of a rule of priority such as the estate costs rule. The liquidator could argue also that perhaps ECRC Land does not go so far as to constitute binding authority for a new rule of personal liability which does not arise from breach of the estate costs rule and which obliges a liquidator to have regard to the prospects of litigation opponents recovering their litigation costs in full whenever a liquidator decides to litigate. Certainly, the Court of Appeal was aware of the critical distinction between the court’s supervisory jurisdiction over a liquidator’s decision to litigate and the court’s supervisory jurisdiction over a liquidator’s breach of the estate costs rule (ECRC Land at [51]). Finally, it is arguable that there is no fault on the part of a liquidator in the third situation.

154 The Court of Appeal in ECRC Land expressly recognised and addressed the argument which Mr Seah makes. There is no principled basis upon which I can distinguish the Court of Appeal’s reasoning as between litigation which the liquidator commences and litigation which the liquidator adopts. Whether a liquidator who breaches the estate costs rule is personally liable for any shortfall in the opponent’s costs recovery which is caused by the breach is outside the scope of the questions posed to me on this application. The liquidator poses these questions prospectively, before she makes any distributions of the company’s assets. The questions posed to me therefore do not relate to any actual breach of the estate costs rule. Nor has the liquidator framed her questions in respect of the consequences of a hypothetical future breach of the estate costs rule.

155 The final limb of this argument is that it is unfair for the estate costs rule to apply where a company in liquidation is defending rather than prosecuting a claim. A defendant does not choose to commence litigation and cannot decide who its plaintiff will be. But a plaintiff does choose to commence litigation and does decide who its defendants will be. Thus, the ordinary rule in our system of procedure is that a plaintiff takes its defendant as it finds it, for better or for worse, in solvency and insolvency. Further, s 262(3) of the Act requires any plaintiff who wishes to commence or continue proceedings against a company in liquidation to secure the court’s leave to do so. A plaintiff who applies under s 262(3) is, ex hypothesi, aware of the company’s insolvency. The risk of securing an irrecoverable judgment, whether for substantive relief or even for costs, is an intrinsic risk which a plaintiff who chooses to sue a company in liquidation takes on voluntarily. Therefore, whatever the position may be where a company in liquidation is a plaintiff, the estate costs rule should not apply to a company in liquidation who is a defendant.

156 There are three difficulties with this argument.

157 First, the argument is contrary to precedent. Of the cases which I have analysed above, In re Trent (at [78] above), Lofthouse (at [61] above) and Kenworth (at [69] above) are direct authority against this limb of Mr Seah’s argument. Thus, Sir William James VC concluded in In re Trent that there “can be no distinction in principle between the costs of an action which fails and the costs of an action which has been unsuccessfully defended” (at 98). Citing In re Trent, ECRC Land likewise draws no distinction between a company in liquidation which prosecutes a claim and one which defends a claim (at [9]):

158 Second, the argument is contrary to principle. Both the risk/reward principle and the reciprocity principle are commutative. They both operate just as much in favour of a defendant against a plaintiff as they do in favour of a plaintiff against a defendant.

159 Third, the argument is contrary to the balance which the Court of Appeal held in ECRC Land the estate costs rule has struck appropriately between the competing underlying policy considerations. The Court of Appeal held that there was a “real and tangible justification for deterring insolvent companies from commencing litigation without any real possibility of satisfying a successful defendant’s costs” (at [75]). Although this justification is expressed in terms of a plaintiff who is in liquidation, the policy in favour of this deterrent is again commutative. If a company in liquidation ought to be deterred from prosecuting a claim as a plaintiff when it has no real prospect of satisfying an adverse costs order, it should equally be deterred from defending a claim in the same circumstances.

160 Mr Seah submits that allowing the estate costs rule to apply even where the company in liquidation is defending litigation runs the risk of subverting the policy of the proof of debt regime. The estate costs rule will then provide a perverse incentive for a plaintiff to insist on pressing on with its claim against the company purely to secure a costs order and the consequent super priority for those costs, instead of diverting its claim to adjudication under the proof of debt regime.

161 Mr Seah’s fear is overstated. A creditor can commence or continue litigation against a company in liquidation only if the creditor secures the leave of court to do so under s 262(3) of the Act. At the leave application, the court will no doubt be astute to prevent any attempt by a creditor to pursue litigation for the sole purpose of securing the benefit of super priority under the estate costs rule. That will especially be true if the liquidator is opposing the application for leave on the basis that she concedes the claim. Indeed, if a plaintiff is so confident of securing judgment on the claim and of securing the costs order which will ordinarily accompany it that it is prepared to accept the vagaries, delay and expense of litigation by insisting on pursuing the litigation purely to secure super priority for a costs order, any well-advised liquidator would be expected readily to concede the claim and to resist the application for leave on that basis. In any event, it will be quite obvious when a plaintiff is seeking leave purely to secure super priority for a costs order. And any plaintiff pursuing a case which is so strong, or which the liquidator is ready to concede, will almost certainly fail to establish a compelling case to have recourse to litigation rather than to the proof of debt regime (Korea Asset Management at [50]). Alternatively, the court may grant leave under s 262(3) on terms that the creditor shall not seek super priority under the estate costs rule for its costs.

162 For these reasons, I do not accept the final limb of Mr Seah’s argument that the scope of the estate costs rule does not or should not encompass claims against a company in liquidation.

163 In the alternative, Mr Seah submits that the particular facts of this litigation between this company and this defendant warrants a departure from the estate costs rule. He therefore asks me to exercise my discretion under s 283(3) of the Act to order that any costs order in favour of the defendant will not be accorded super priority in so far as it relates to the defendant’s pre-liquidation costs. ²³

164 Section 283(3) of the Act gives the court a discretion to alter the order of priorities in which liquidation expenses are paid:

165 Mr Seah advances several reasons for inviting me to exercise my discretion under s 283(3) of the Act in this way:

(a) The proceedings between the Company and the defendant are almost over. All that remains before a judgment can be rendered is a final hearing for the parties to present their oral closing submissions. The costs which the defendant will incur after the liquidator adopts the litigation will, on the facts of this case, be marginal.

(b) Allowing the proceedings to conclude with a judgment is the best way in which to resolve the Company’s claim and the defendant’s counterclaim. Diverting the defendant’s counterclaim to the proof of debt regime would require the liquidators to duplicate for the purposes of adjudication all of the forensic work which has already been undertaken for the purposes of the litigation. And if the liquidator rejects the proof of debt on the basis that the Company’s claim exceeds the defendant’s counterclaim, the parties will end up back in litigation anyway.

(c) The Company has sufficient assets to cover only an order for the defendant’s costs of the present application and the costs of getting up and preparing the oral closing submissions. Unless the court exercises the discretion under s 283(3) in the liquidator’s favour, the liquidator will have to abandon the claim. It is not fair on the creditors if the Company has to abandon its claim while the defendant remains entitled to prove for a dividend on its counterclaim.

166 I do not accept Mr Seah’s submissions.

167 Section 283(3) of the Act is phrased broadly. But the discretion it creates is applied narrowly. As the Court of Appeal held in ECRC Land, “[t]his discretion to reverse the erstwhile order of priorities will only be exercised in exceptional situations” [emphasis in original] (at [85], citing Re Linda Marie Ltd [1989] BCLC 46) and “is one that will be exercised sparingly” (at [87]).

168 The present circumstances are not exceptional and do not warrant the benefit of a discretion which is to be exercised sparingly.

169 The case law I have analysed shows that a liquidator having to decide whether to adopt litigation (whether the company is the plaintiff or the defendant) is a common and recurring problem. That is why there are so many authorities from so many jurisdictions of such antiquity on this very issue. There is nothing exceptional about this case which distinguishes it as a matter of principle from any other case to which the estate costs rule applies. To exercise my discretion in the manner suggested by Mr Seah would be to eviscerate the estate costs rule and the carefully balanced policy rationale on which it rests.

170 If the liquidator now adopts the litigation, it can only be because she has decided that: (a) the Company has good prospects of succeeding in the claim; (b) the Company has good prospects of defending the counterclaim; (c) that the rewards of adopting the litigation analysed in light of the probability of success outweigh the risks of doing so; and (d) that the proof of debt procedure is inappropriate for resolving the counterclaim. To put it bluntly, making these decisions – of course on professional advice – is what liquidators are paid to do. To the extent that she forms the view that all of these conditions are satisfied but cannot secure an indemnity from the creditors to pursue the litigation and therefore has to abandon it, that is an intended consequence of the estate costs rule as conceived in ECRC Land.

171 For the reasons above, I answer the questions posed to me as follows:

(a) If the liquidator decides to continue legal proceedings against a defendant (which were commenced before the winding up order was made), and the defendant ultimately succeeds in these legal proceedings, the successful defendant will be entitled to be paid its costs in priority to the other general expenses of the liquidation (ie, including the remuneration and expenses of the liquidator).

(b) In that event, the successful defendant will be entitled to be paid its entire costs from the beginning of the legal proceedings.

172 As the event on this application is in the defendant’s favour, I have ordered the Company to pay the defendant its costs of and incidental to this application, such costs fixed at $13,000 including disbursements.

173 Finally, I have also ordered that the Company’s solicitor and client costs of and incidental to this application be agreed with the Company’s committee of inspection or taxed by the court if not so agreed. These costs shall be deemed to be part of the costs and expenses incurred in the winding up of the Company and shall be paid out of the assets of the Company.