[2015] SGCA 50

The Royal Bank of Scotland NV (formerly known as ABN Amro Bank NV) and others v TT International Ltd (nTan Corporate Advisory Pte Ltd and others, other parties) and another appeal
[2015] SGCA 50

Case Number	:	Summons No 5682 of 2012 in Civil Appeal No 44 of 2010 and Summons No 6520 of 2012 in Civil Appeal No 47 of 2010
Decision Date	:	30 September 2015
Tribunal/Court	:	Court of Appeal
Coram	:	Sundaresh Menon CJ; Andrew Phang Boon Leong JA; Chan Seng Onn J; Quentin Loh J; Vinodh Coomaraswamy J
Counsel Name(s)	:	Edwin Tong SC, Kenneth Lim, Peh Aik Hin, Tan Kai Liang, Jasmine Tham and Chua Xinying (Allen & Gledhill LLP) for nTan Corporate Advisory Pte Ltd; Lee Eng Beng SC, Low Poh Ling, Raelene Su-Lin Pereira, Jonathan Lee Zhongwei and Mark Ortega (Rajah & Tann Singapore LLP) for DBS Bank Ltd, Habib Bank Limited and Oversea-Chinese Banking Corporation Limited (acting jointly as the Monitoring Committee); Chan Hock Keng, Ong Pei Chin, Monica Chong Wan Yee and Charisse Lau (WongPartnership LLP) for TT International Ltd.
Parties	:	The Royal Bank of Scotland NV (formerly known as ABN Amro Bank NV) — Commerzbank Aktiengesellschaft, Singapore Branch — Cooperatieve Centrale Raiffeisen-boerenleenbank BA (Trading as Rabobank International, Singapore Branch) — Oversea-Chinese Banking Corporation Limited — TT International Limited — nTan Corporate Advisory Pte Ltd — DBS Bank Ltd (acting jointly as the Monitoring Committee) — Habib Bank Limited (acting jointly as the Monitoring Committee) — Ho Lee Construction Pte Ltd

30 September 2015

Judgment reserved.

Sundaresh Menon CJ (with whom Chan Seng Onn J, Quentin Loh J and Vinodh Coomaraswamy J agreed):

Introduction

1 We have before us an application by nTan Corporate Advisory Pte Ltd (“nTan”) to set aside a previous decision of this court, ie, the Court of Appeal (“CA”). nTan alleges that that decision is defective for at least two reasons. One is that the CA acted without jurisdiction in making it, and the other is that it was a decision made in breach of the rules of natural justice. The decision in question was handed down in the broad context of a scheme of arrangement under s 210 of the Companies Act (Cap 50, 2006 Rev Ed) in respect of TT International Ltd (“the Company”). We shall refer to this scheme of arrangement as “the Scheme”. We should mention, at the outset, that nTan’s setting-aside application, apart from raising issues of jurisdiction and breach of natural justice, also gives rise to issues of res judicata. This will become apparent in due course.

2 In 2008, the Company encountered financial troubles. Seeking to extricate itself from its predicament, it appointed nTan as the independent financial advisor (“IFA”) to it and its subsidiaries. It was agreed that the fees payable to nTan for its services would include what was referred to as a value-added fee that would become payable only upon the occurrence of certain defined events, one of which was the agreement of the Company’s creditors to a scheme of arrangement and the approval of such a scheme by the court. In simple terms, the value-added fee payable to nTan (“the VAF”) was a percentage of that amount of the debt owed by the Company to its creditors which was “waived, written off, extinguished, forgiven or avoided” or converted into equity in the Company pursuant to the anticipated scheme of arrangement. Hence, the greater the value of the debt waived or otherwise rendered not payable as a result of that scheme of arrangement, the greater the amount of the VAF that nTan stood to receive. For this reason, the VAF was described as a success fee.

3 In due course, the Scheme was proposed and voted on by those of the Company’s creditors to whom it was to apply – we shall refer to these creditors as the “Scheme Creditors”. An attempt to have the Scheme approved by the court was unsuccessful, but following a re-vote, the CA gave its sanction to the Scheme on 13 October 2010. This was an event triggering nTan’s entitlement to the VAF under the terms of its agreement with the Company. But, to date, nTan still has not received the VAF. This is because, by a decision handed down on 27 September 2012, the CA held that: (a) nTan was not entitled to the full amount of the VAF; and (b) “the relevant parties” – namely, nTan/those officers of nTan who were named in the Scheme documents as the Scheme Manager (“the SM”), the Company and the Scheme Creditors constituting the Monitoring Committee which was to ensure (among other things) that the Scheme was implemented according to its terms (“the MC”) – were instead to agree on what the quantum of the VAF ought properly to be, failing which agreement, it would fall to the High Court to assess nTan’s fees. This is the CA decision that nTan seeks to set aside, and it is reported as The Royal Bank of Scotland NV (formerly known as ABN Amro Bank NV) and others v TT International Ltd and another appeal [2012] 4 SLR 1182. We shall refer to it as “the VAF Decision”.

Background facts: the Scheme

4 The Company was incorporated in Singapore in October 1984 as a private limited company and was subsequently listed on the Stock Exchange of Singapore in June 2000. It was, at the material time, the main distributor and licensee of the AKIRA brand of electronic products worldwide. As we have mentioned, the Company encountered financial troubles in 2008, which resulted in its creditors declaring events of default and either threatening to commence or actually commencing legal proceedings against it. It was against this backdrop that nTan was appointed as the IFA to the Company and its subsidiaries by an appointment letter dated 28 October 2008. That letter, read together with another letter from nTan to the Company dated 15 May 2009, stipulated that the fees payable to nTan for its services comprised the VAF in addition to charges for the time spent by its personnel.

5 The Scheme was eventually proposed as a solution to the Company’s financial woes. On 29 January 2009, the High Court granted the Company liberty to call a meeting of the Scheme Creditors in order for them to consider the Scheme and, if they thought fit, approve it. According to the Scheme documents, oversight of the Company’s implementation of the Scheme was to be entrusted to the SM, identified in the Scheme documents as “Mr Nicky Tan Ng Kuang, Mr Dan Yock Hian and/or Ms Lim Siew Soo”, all of whom were (as alluded to at [3] above) nTan personnel. Mr Nicky Tan Ng Kuang (“Mr Nicky Tan”) is and was at all material times the chief executive officer as well as the controlling shareholder of nTan.

6 The meeting of the Scheme Creditors was held on 16 October 2009. Under s 210(3) of the Companies Act, approval of the Scheme required not only that a majority in number of the voting Scheme Creditors sanction it, but also that this numerical majority represent “three-fourths in value” of the Scheme Creditors “present and voting either in person or by proxy at the meeting”. The latter requirement meant that the Scheme Creditors voting in favour of the Scheme had to hold at least 75% of the total value of the debt owed by the Company to those Scheme Creditors who were present and voting at the meeting. The value of each Scheme Creditor’s debt, and hence, the extent of each Scheme Creditor’s voting power at the meeting, was to be determined by the SM since, under the terms of the Scheme, the SM was to review and assess the proofs of debt submitted by the Scheme Creditors, and could admit or reject these either in whole or in part. Shortly after 5.00pm on the day of the meeting, after the Scheme Creditors had cast their votes, they were informed that the final outcome of the voting would be determined only after the SM had completed the review and assessment of the proofs of debt so as to ascertain whether the 75% threshold had been crossed.

7 Two months later, on 17 December 2009, the SM announced the voting results, having completed the adjudication of the proofs of debt in the meantime: the requisite majority of Scheme Creditors had voted in favour of the Scheme. The necessary “three-fourths in value” had been attained by a razor-thin margin. On the basis of the proofs of debt admitted by the SM, the Scheme Creditors that voted in favour of the Scheme held 75.06% of the total debt owed by the Company to those Scheme Creditors who were present and voting at the meeting. By way of contrast, if one were to take the pre-adjudication proofs of debt as submitted to the SM, the value of the votes in favour of the Scheme would be closer to 66%. This can be seen from the following table, which sets out a comparison of the proofs of debt submitted to and the proofs eventually admitted by the SM, as well as the respective proportions of the value of the votes for and against the Scheme:

	Submitted as at 16 October 2009	Admitted as at 17 December 2009
Total value of proofs voting, ie, less abstentions	$554.70m	$485.39m
Total value of votes in favour of the Scheme	$364.98m	$364.34m
And as a percentage of the total value of proofs	65.80%	75.06%
Total value of votes against the Scheme	$189.74m	$121.05m
And as a percentage of the total value of proofs	34.20%	24.94%

8 On the strength of this voting outcome, the court’s sanction of the Scheme was sought, and it was granted by Judith Prakash J in the High Court on 15 March 2010. However, a number of Scheme Creditors were dissatisfied with this and lodged Civil Appeals Nos 44 and 47 of 2010 (“CAs 44 and 47 of 2010”) against Prakash J’s decision. After hearing arguments on 18 August 2010, the CA, on 27 August 2010, set aside the approval of the Scheme because certain aspects of the voting procedure and the adjudication of some of the proofs of debt were found to be unsatisfactory (see The Royal Bank of Scotland NV (formerly known as ABN Amro Bank NV) and others v TT International Ltd and another appeal [2012] 2 SLR 213 (“The Royal Bank of Scotland v TT International”)). In setting aside the Scheme, the CA gave a number of directions, including a direction that a further meeting of the Scheme Creditors be held within four weeks. These directions are set out in Annexure I of The Royal Bank of Scotland v TT International.

9 The further meeting of the Scheme Creditors was duly held on 24 September 2010. On that occasion, the Scheme Creditors were, pursuant to the CA’s directions, divided into two classes for the purposes of voting on the Scheme. One class, consisting of the Company’s subsidiaries and related parties, voted unanimously in favour of the Scheme. In the other class, which was the general class of Scheme Creditors, a numerical majority representing 76.34% in value voted in favour of the Scheme.

10 Further hearings then took place before the CA on 5 and 13 October 2010. At the latter hearing, the CA issued brief grounds of decision (“the Brief Grounds”) declaring that the requisite majority of Scheme Creditors had indeed voted in favour of the Scheme at the meeting held on 24 September 2010 (see [5] of the Brief Grounds, which may be found in Annexure II of The Royal Bank of Scotland v TT International). The CA proceeded in the Brief Grounds to approve the Scheme, but subject to alterations which it considered it was empowered by s 210(4) of the Companies Act to make. One such alteration was in the composition of the MC (at [8(a)]). The CA directed that certain banks be made members of the MC; this was to give the MC a more balanced composition by including some Scheme Creditors who had not voted in favour of the Scheme.

11 The most important alteration for present purposes is the one at [8(j)] of the Brief Grounds (hereafter referred to simply as “[8(j)]”), where the CA directed as follows: “All professional costs (and disbursements) of the [SM’s] and the [Company’s] professional advisors incurred after 27 August 2010 shall be taxed by the High Court”. The significance of the date 27 August 2010 is that this was the date on which the CA set aside Prakash J’s decision to sanction the Scheme. We digress briefly to note that [8(j)] appears to have been included because at the hearing of CAs 44 and 47 of 2010, some of the Scheme Creditors had raised concerns over the lack of transparency as to the fees of the professionals involved in implementing the Scheme, together with concerns that these fees might be excessive.

Correspondence regarding the VAF

12 Throughout this time, neither the High Court nor the CA had been made aware of the VAF. The question of nTan’s entitlement to the VAF was first brought to the CA’s attention about a year and a half later by a letter dated 27 January 2012 addressed to the Supreme Court Registry (“the Registry”) from Rajah & Tann LLP (“Rajah & Tann”), the solicitors for the MC. The letter was copied to Allen & Gledhill LLP, who, in a later letter, indicated that they were acting for “Mr Nicky Tan, the [SM] of [the Company]”, as well as to WongPartnership LLP, the solicitors for the Company.

13 This letter triggered a series of correspondence between these three parties and the CA, which culminated in the VAF Decision. Given that nTan alleges that it was denied natural justice in the manner in which the CA arrived at the VAF Decision, and given that the CA arrived at the VAF Decision following this series of correspondence, we shall describe in some detail the contents of the various letters. In doing so, we shall use the names of the parties rather than the names of the law firms who represented them and sent the letters on their behalf.

The MC’s letter of 27 January 2012

14 The MC’s letter of 27 January 2012 raised two broad issues, both of which it said concerned [8(j)]. The first was the MC’s role in the taxation of the professional costs of the SM and the Company’s professional advisors. The MC reported that the Company was proposing that the taxation mandated by [8(j)] be waived so long as the Company agreed to the costs put forward by its professional advisors and the SM, and suggested that by this, the Company was attempting to circumvent the CA’s direction in [8(j)]. The MC said that it had declined the Company’s request as that would compromise the Scheme Creditors’ interests, and asked the CA to direct that any bill of costs submitted by any of the Company’s professional advisors (including the SM) be served on the MC, which would be permitted to instruct solicitors to participate in the taxation proceedings, with the resultant costs to be borne by the Company.

15 The second broad issue was the VAF. The MC alleged that it had only “recently” been apprised of the existence of the VAF, and suggested that the VAF should be subject to taxation under [8(j)]. Its contention was that the VAF had become payable only on 13 October 2010, ie, the day on which the CA sanctioned the Scheme, with the result that the VAF was a species of professional costs incurred after 27 August 2010 and so fell within the scope of [8(j)].

16 Leaving [8(j)] aside, the MC also expressed a more general concern that a success fee, the quantum of which depended on the value of debt restructured under the Scheme (as was the case for the VAF), would give rise to “a possible conflict of interest and duty on the part of nTan, in that it was nTan that adjudicated on the debts that were to be admitted for voting on and participating in the Scheme”. The MC then requested that the CA direct that the VAF be subject to taxation in accordance with [8(j)], and that the CA provide guidance as to how such taxation should be conducted.

Initial responses to the MC’s 27 January 2012 letter

17 On 30 January 2012, the Registry wrote to the SM and the Company with an invitation to respond to the MC’s letter of 27 January 2012. On 1 February 2012, the SM – identified as “Mr Nicky Tan” – wrote to the Registry. The SM’s letter was stated to be in response to the MC’s 27 January 2012 letter, but not the Registry’s 30 January 2012 letter. The letter said that the MC’s letter had made a number of “serious allegations” against various parties and, furthermore, contained some “factual inaccuracies”. The SM also complained that the MC had not disclosed relevant material facts. Given that the MC was asking the CA to make substantive directions in relation to the Scheme, the SM proposed that the MC put forward its request in a more appropriate manner by way of a formal application supported by affidavit, after which the SM and any other concerned party could file a reply affidavit.

18 On 3 February 2012, the Company wrote to the Registry. This letter addressed the first issue in the MC’s 27 January 2012 letter, but not the second. In other words, it did not deal with the issue of whether the VAF fell within [8(j)]. Rather, the Company’s sole concern in this letter was to explain why it wished to be able to waive the requirement under [8(j)] that its professional advisors’ costs be taxed, and to clarify that it had no objection to the MC reviewing the fees of its professional advisors, provided that the MC did not abuse this by stirring up “unnecessary disputes” regarding those fees.

19 On the morning of 7 February 2012, the SM sent another letter to the Registry. It stated as a preliminary matter that the SM had not received the Registry’s 30 January 2012 letter, and had known of its existence only when the Company extended him a copy. The SM then reiterated the request in his 1 February 2012 letter that the MC be directed to make its request for further directions from the CA by way of a formal application supported by affidavit. He also asked for more time to make a substantive and detailed response to the MC’s letter of 27 January 2012.

20 Next, the SM put forward a number of “additional points” for the CA’s consideration. These additional points addressed both of the issues raised in the MC’s 27 January 2012 letter. As to the first issue, which concerned the MC’s role in the taxation of the bills of costs of the Company’s professional advisors (including the SM), the SM rejected the MC’s insinuations that he had colluded with the Company to avoid taxation of his professional fees. He asserted that the Company had all along proposed that the MC should have a role to play in this process.

21 As to the second issue, which concerned the VAF, the SM rejected the MC’s allegation that it had only “recently” been informed of the VAF. The SM asserted that on 12 August 2011, the Company had informed the Scheme Creditors, including the MC, that it had earmarked $63m in its financial statements for meeting potential liabilities and other Scheme-related expenses “including, inter alia, nTan’s VAF”. It was accordingly suggested that the MC ought to have been aware of the existence of the VAF by 12 August 2011. The SM also said that the Company had provided the MC with further details relating to the VAF just over a month later, on 22 September 2011, and had informed the MC on that occasion that the VAF was payable upon successful sanction of the Scheme and that the Company would negotiate with nTan on the quantum of the VAF. The SM’s point was that the MC had in fact had at least five months since 22 September 2011 to plot its course in relation to the VAF.

22 The SM then went on to reiterate once more his request that the MC be directed to make a formal application supported by affidavit for the further directions which it sought from the CA. He added that, if the CA were not minded to direct so, he would seek an extension of time until 15 March 2012 to submit a detailed response to the MC’s 27 January 2012 letter. In making this request, the SM pointed out that no party would be prejudiced if more time were taken to set out all the facts and arguments properly. He also emphasised that it was important to consider the matter carefully because the MC’s request that (among other things) the VAF be taxed under [8(j)] would “have broad ranging impact and [raise] matters of public interest”, and would effectively “alter the agreement between nTan and the Company and the terms of the Scheme which ha[d] been approved by the requisite majority of the [Scheme Creditors] and sanctioned by the [CA]”.

The Registry’s letter of 7 February 2012 and the responses thereto

23 In the late afternoon on the same day (ie, 7 February 2012), the Registry sent the parties a letter on behalf of the CA. This letter referred to all the previous correspondence, including the letter sent that morning by the SM. It informed the parties that the CA was considering the issues raised and asked them to clarify certain matters relating to the VAF, namely:

(a) whether the VAF had been disclosed to any of the Scheme Creditors prior to the CA’s sanction of the Scheme on 13 October 2010, and if so, when, how and to whom disclosure had been made;

(b) whether the VAF had been disclosed at any meeting of the Scheme Creditors;

(d) whether the VAF had been disclosed to the Scheme Creditors on 12 August 2011, as the SM said it had been.

24 The Company was the first to respond, doing so on 10 February 2012. Its answers to the questions posed were:

(a) No – to the best of the Company’s knowledge, the VAF had not been disclosed to any of the Scheme Creditors prior to the CA’s sanction of the Scheme on 13 October 2010;

(b) No – the VAF had not been disclosed at the meetings of the Scheme Creditors held on 16 October 2009 and 24 September 2010;

(d) Yes – the VAF had been disclosed to the Scheme Creditors on 12 August 2011.

25 Both the SM and the MC responded on 14 February 2012. Turning, first, to the SM’s letter, his answers to the four questions posed were identical to those given by the Company. But, the SM did not confine his response to those matters. Instead, he went on to make two broad points. One was that the non-disclosure of the VAF in the context of the Scheme was not inconsistent with prevailing practice. The SM stated that, to his knowledge, companies in other schemes of arrangement likewise had not disclosed to their creditors the engagement letters or the fees of their professional advisors prior to the court’s hearing of the application to sanction the scheme in question. The SM’s other broad point was that, even before the CA sanctioned the Scheme on 13 October 2010, two of the Scheme Creditors had, in their submissions to the CA, articulated some concerns over the fees of the Company’s professional advisors and the lack of information disclosed by the Company in that regard, but the CA had not thought fit to pursue this line of inquiry very far.

26 A substantial part of the SM’s 14 February 2012 letter was taken up with expanding on the second of the two broad points just mentioned. He pointed out that:

(a) Ho Lee Construction Pte Ltd, a Scheme Creditor opposed to the Scheme, had complained in its submissions to the CA dated 1 October 2010 of a “lack of visibility in respect of nTan’s fees, including on a success formula”;

(b) Oversea-Chinese Banking Corporation Limited (“OCBC”), another Scheme Creditor opposed to the Scheme, had similarly complained in its written submissions to the CA dated 5 October 2010 that the Company had “not seen any need to justify the fees” paid to the SM and other professional advisors, and had expressed dismay at the impossibility of ascertaining whether the SM would “receive any success or contingency fee upon the sanction of the Scheme”;

(c) in its 5 October 2010 written submissions, OCBC had also argued (among other things) that, in view of the lack of transparency as to the SM’s fees, the office of SM ought not to be occupied by any person from nTan because “[g]rave doubt” had been cast on nTan’s “impartiality, independence, competence and professional judgment”; but

(d) significantly, the CA had not acceded to this last-mentioned aspect of OCBC’s written submissions (ie, the argument at sub-para (c) above), nor had the CA made any directions in response to it.

27 The SM concluded his 14 February 2012 letter by reiterating yet again his request that the MC be directed to make a formal application supported by affidavit for the further directions which it sought from the CA. He also reiterated his alternative request that he be granted an extension of time until 15 March 2012 to submit a detailed response to the allegations that the MC had made in its 27 January 2012 letter.

28 Moving, finally, to the MC’s response of 14 February 2012, its answers to the first three of the four questions posed by the CA were the same as those given by the Company and the SM. But, it departed from the Company and the SM on the fourth question, which was whether the VAF had been disclosed to the Scheme Creditors on 12 August 2011. The MC said that, to the best of its recollection, there had been no reference to or mention of the VAF on that day. The Scheme Creditors had indeed been told about the earmarking of $63m in the Company’s financial statements for meeting potential liabilities and other Scheme-related expenses, but the Company had then declined to answer a direct query regarding the professional fees due to the SM on the basis that those fees remained the subject of ongoing negotiations between the Company and the SM. The MC alleged that express disclosure of the VAF had taken place only on 22 September 2011.

29 The MC concluded its letter by expressing disagreement with nTan’s position that it should make a formal application supported by affidavit for the further directions which it sought from the CA. In this connection, the MC clarified that it was “merely seeking further guidance from the [CA] on the directions laid down in [the] Brief Grounds”, and was “not seeking a ruling from the [CA] on any factual allegations in relation to the fee arrangements between the Company and nTan and/or the [SM]”. According to the MC, the facts set out in its 27 January 2012 letter served only “to provide the background and context of [its] request for directions”. Nonetheless, the MC also indicated that, should the CA take the view that those facts were relevant to the directions sought in its 27 January 2012 letter, it would be happy to address the matters subsequently raised by the SM and the Company.

The Registry’s further letter of 20 February 2012

30 On 20 February 2012, the Registry wrote to the parties again. It listed four issues that the CA said had been raised for its consideration by the correspondence thus far, namely:

(a) whether any bill of costs to be taxed under [8(j)] should be served on the MC;

(b) whether the MC should be allowed to instruct solicitors to participate in the taxation proceedings;

(d) whether the VAF should be taxed under [8(j)].

The CA answered the first three issues in the affirmative, but emphasised, with regard to the third issue, that taxation could be waived only if the MC also consented.

31 As to the fourth issue, the CA requested the parties to address it on three issues of fact and four issues of law. The three issues of fact were:

(a) what the prevailing practice was for remuneration of scheme managers in Singapore;

(b) whether other scheme managers in Singapore, or in countries where similar statutory schemes existed, had ever adopted similar arrangements for success fees which would be paid over and above the scheme manager’s basic remuneration; and

32 The four issues of law were:

(a) whether the Company and/or the SM ought voluntarily to have disclosed the VAF to the Scheme Creditors before or at the meetings of the Scheme Creditors, and/or to the court;

(b) assuming that the Company and/or the SM had such an obligation to disclose, what the consequences of a failure by the Company and/or the SM to discharge that obligation ought to be;

(c) assuming that the Company and/or the SM had no such obligation to disclose, whether there should nonetheless be any legal constraint on the nature and quantum of the success fee arranged between a company and its scheme manager; and

(d) whether the court had inherent power to tax the VAF, and if not, whether the VAF fell within [8(j)].

33 The CA directed the parties to respond within 14 days of their receipt of the Registry’s 20 February 2012 letter, meaning that the deadline was 5 March 2012. On 1 March 2012, the MC wrote to the Registry seeking an extension of time on the basis that novel points of law had been raised. The Registry replied on the same day informing the parties that the new deadline was 14 March 2012.

The parties’ responses to the Registry’s 20 February 2012 letter

The responses of the Company and the MC

34 In their responses filed on 14 March 2012 to the questions set out in the Registry’s 20 February 2012 letter, there was broad agreement between the Company and the MC on the factual issue of the prevailing practice for remuneration of scheme managers in Singapore and elsewhere. Both parties said that they were not aware of any specific schemes of arrangement in which the company and its scheme manager had agreed on a success fee to be paid over and above the scheme manager’s basic remuneration, which would consist of time costs and/or capped fees for agreed pieces or periods of work. As for the CA’s query on the amount of the VAF, the Company provided a figure of $15.2m, whereas the MC indicated a figure in the range of $15m to $30m.

35 Turning to the questions of law posed by the CA, the Company and the MC diverged on the first legal issue, namely, whether the Company and/or the SM ought voluntarily to have disclosed the VAF to the Scheme Creditors and/or to the court. The Company argued that there was no such obligation of disclosure because it was not the practice in Singapore for a company to include the details of its scheme manager’s fees in the scheme documents, and because, in this particular case, the Company understood from the SM that it was not to disclose to third parties the terms of its engagement with the SM without his consent.

36 The MC, on the other hand, contended that the Company and the SM did have an obligation to disclose the VAF. Two main reasons were furnished. The first had to do with the materiality of the information to be disclosed. The MC argued in this regard that the Scheme Creditors should have been given all information necessary for them to make a meaningful choice in voting on the Scheme, and that the existence of the VAF was one such piece of information as the large quantum of the VAF could have a material impact on the Company’s financial situation. The second reason had to do with an alleged position of conflict on the part of the SM and nTan. The MC argued that the VAF ensnared the SM and nTan in “two distinct levels of conflict”. One was that the SM had an incentive to admit the proofs of debt of those Scheme Creditors supporting the Scheme and reject the proofs of those opposing, so that the VAF would become payable; the other was that the SM had an incentive to admit as much debt as possible so as to increase the quantum of the VAF.

37 As to the second question of law, namely, what the consequences of a failure to disclose the VAF ought to be, the Company did not give any substantial answer. The MC, however, contended that the Company’s and the SM’s breaches of their disclosure obligations could have three possible consequences. First, the Scheme might be set aside on the ground that the Scheme Creditors’ consent to it had been obtained by fraud; second, the Company’s obligation to pay the VAF to nTan might be set aside by reason of the SM’s breach of a fiduciary duty owed by him to the Scheme Creditors such as would call for an account of profits, or alternatively, the Company’s liability to pay the VAF might be subordinated to the claims of the Scheme Creditors; and third, the SM might be removed and replaced.

38 In relation to the fourth question of law, viz, whether the court had inherent power to tax the VAF, and if not, whether the VAF fell within [8(j)], the MC considered that the CA did not have inherent power to tax the VAF, but maintained that the VAF did fall within [8(j)]. The Company, on the other hand, contended that the CA had inherent power to tax the VAF. It expressed no view on whether the VAF fell within [8(j)].

The response of the SM

39 The SM’s submissions in response to the questions set out in the Registry’s 20 February 2012 letter, which were likewise filed on 14 March 2012, spanned 155 paragraphs across 104 pages, with an annex that was five pages long. In these submissions, the SM did not launch at once into his answers to the aforesaid questions. Instead, he prefaced his answers with “four key points” which he said were “important and relevant” for the CA to have regard to.

40 The first key point was that the VAF was payable to nTan in its capacity as the Company’s IFA. The SM contended that a distinction should be drawn between nTan’s role as the Company’s IFA and its officers’ roles as the SM: while the SM’s role was primarily administrative in nature, the functions of the IFA were “wide-ranging and substantial” and had to be performed under “substantial time pressure” as well as in the context of “different classes of creditors making different demands and threats against the Company”. Seen in this light, the SM submitted, a success fee arrangement for the IFA’s work was fair and reasonable. It would even have practical benefits for the Company as the IFA would “dedicate its efforts towards achieving a successful outcome in the shortest practicable time” rather than towards “churn[ing]” work, ie, towards maximising the time spent working on the matter, as might be prone to be the case should remuneration be based only on the IFA’s time costs. More fundamentally, the SM contended, since the work performed by an IFA was done prior to a company’s entry into a scheme of arrangement, it “[stood] outside of the insolvency regime”. Thus, an IFA, unlike a liquidator or a judicial manager, was not an officer of the court, and the court therefore had no supervisory role over an IFA. If this were otherwise, said the SM, it would undermine the ability of financially-distressed companies to obtain specialist advice from IFAs.

41 The second key point raised by the SM was that the CA should have regard to the MC’s conduct in considering whether the VAF should be subject to taxation under [8(j)]. To begin with, the SM cautioned against conflating the question of taxation with the matter of his and the Company’s alleged failure to disclose the VAF to the Scheme Creditors; he argued that the MC itself, in its 14 February 2012 letter, had taken the position that its request for (among other things) a direction that the VAF be taxed under [8(j)] was separate from this alleged non-disclosure. In any event, the SM submitted, in so far as the MC was alleging a breach of an obligation to disclose, it was incumbent on the MC to make an application to the court explaining why there was such an obligation, given that the terms of the engagement between the Company and nTan were subject to a confidentiality clause.

42 The SM then pointed out that the MC had known all along that the SM was also the IFA, and that certain of the Scheme Creditors had at various points raised questions pertaining to nTan’s remuneration, including the specific question of whether nTan would receive a success fee – the SM suggested that this was “because the creditors were aware that such arrangements are commonplace in the financial and corporate sectors”. If the issue of nTan’s remuneration had truly been of concern, the SM contended, the MC could have sought disclosure by way of an application to the court prior to the sanction of the Scheme. Despite having had ample opportunities to do so, the MC had opted not to take that course. The SM’s point – although he did not say so in quite these terms – was that any alleged non-disclosure of the VAF did not impugn the validity of the Scheme Creditors’ approval of the Scheme because the issue of nTan’s remuneration had in truth been wholly immaterial to that approval, as shown by the fact that the Scheme Creditors had given their approval despite their awareness of the issue.

43 The third key point set out in the SM’s 14 March 2012 submissions was that the issues raised by the MC were in fact res judicata. The SM pointed out that OCBC had, in its written submissions to the CA dated 5 October 2010, complained that it was not known whether the SM would receive a success or contingency fee, and had argued (among other things) that the SM should be removed if such a fee were payable, but the CA’s only response to this had been to direct in [8(j)] that the professional costs incurred after 27 August 2010 by the SM and the Company’s other professional advisors be taxed. The argument implicit in this was that, since the CA had decided that the issue of nTan’s remuneration called for no other legal consequence apart from a direction that such remuneration be taxed, it was not open to the MC to argue now that some further direction should be made in relation to that same issue of nTan’s remuneration.

44 The SM’s fourth key point was that, although the MC had, in its 27 January 2012 letter, made an allegation of “a possible conflict of interest and duty” against nTan, it had thereafter declined to put this allegation on affidavit. The SM also emphasised that the MC seemed to allege only potential rather than actual conflict.

45 Having made these prefatory comments, the SM went on to answer the questions set out in the Registry’s letter of 20 February 2012. As to the first issue of fact concerning the prevailing industry practice in Singapore for remuneration of scheme managers, the SM distinguished between scheme managers and financial advisors. His position, in summary, was that the prevailing practice in respect of scheme managers was remuneration on a time costs basis only, whereas in respect of financial advisors, it was remuneration based on time costs plus a success fee. The SM highlighted that in five other schemes of arrangement in which Mr Nicky Tan had been the financial advisor of the company concerned, his fees as financial advisor had comprised both time costs and a success fee. In four of those five cases, Mr Nicky Tan had also adjudicated on the proofs of debt and chaired the meetings of the creditors, and in none of them had any disclosure been made to the creditors, prior to the sanction of the scheme, of the fees to be paid to the company’s professional advisors. The SM added that, “to the best of nTan’s knowledge”, it was not uncommon for other financial advisors to charge success fees.

46 The SM also asserted that Rajah & Tann, counsel for the MC, had acted as counsel for the company concerned in three of the five cases in which Mr Nicky Tan had been the financial advisor of a company under a scheme of arrangement, the implication being that Rajah & Tann must have known that there had been a success fee which had not been disclosed to the creditors in those instances. In fact, said the SM, two members of the MC (namely, OCBC and DBS Bank Ltd (“DBS”)) would also have been aware of “nTan’s standard practice of charging a [success fee] when acting as financial advisor”, given their involvement in those other schemes of arrangement.

47 On the second issue of fact, the SM submitted that in the UK, there was a “marked distinction” between financial advisors and insolvency officeholders in terms of the degree of curial supervision of their fees. While the courts could constrain the fees of insolvency officeholders such as liquidators and administrators, there was no similar restriction on the fees of financial advisors. He further submitted that the situation was much the same in Australia, Hong Kong and the United States. Hence, in all these jurisdictions, financial advisors could also charge success fees in addition to fees on a time costs basis. As to the third issue of fact, the SM submitted that the VAF would fall within the range of $28.4m to $31.8m. However, he added that he was discussing the matter with the Company with a view to arriving at a mutually agreed amount.

48 We turn now to the SM’s answers to the CA’s four questions of law. As to the first legal issue, the SM submitted that nTan did not have to make voluntary disclosure of its fee arrangements for work done as the Company’s IFA because its contract with the Company was a private and confidential one. He argued that there were no laws, regulations, industry guidelines or codes of practice which required such disclosure; nor was there any existing practice or precedent suggesting that there was any such requirement.

49 On the second question of law, the SM submitted that, even if there was an obligation to disclose the VAF which had been breached by him and the Company, no consequences should flow from this. This was because the Scheme Creditors had approved the Scheme despite knowing that OCBC had suggested, prior to the CA’s sanction of the Scheme, that there might be an undisclosed success fee payable to nTan. On this basis, it was contended that, even if there had been a breach of a disclosure obligation, that should not affect the Scheme Creditors’ approval of the Scheme.

50 As for the third question of law, the SM submitted that there ought not to be any legal constraint on the nature and quantum of the success fee arranged between a company and its professional advisor because such success fee arrangements were a matter of private contract between the parties concerned. If neither party made an issue of such an arrangement, it was not for the court to interfere. The SM also submitted that the court ought not to interfere because it would have a wider impact on the industry’s remuneration practices.

51 On the fourth and final legal issue, the SM submitted that the VAF did not fall within [8(j)] because the Company’s liability to pay the VAF had arisen on 19 April 2010 – that being the “Effective Date” of the Scheme as defined in the Scheme documents – which meant that the VAF was incurred before 27 August 2010 and therefore did not fall within [8(j)], which caught only professional costs incurred after that date. The SM also submitted that the CA did not have inherent power to tax the VAF because free market principles meant that the Company and nTan, as the Company’s IFA, should be able to contract on such terms as they might agree. The SM further argued that “the court would not have jurisdiction to tax the fees of the financial advisor, as such fees are outside the terms [of a scheme of arrangement]”. If an IFA’s fees were not ordinarily subject to taxation, it would not be correct to subject them to taxation just because the IFA happened to go on to assume the role of a scheme manager.

52 Having answered the questions posed by the CA, the SM proceeded to “provide some additional facts in response” to the MC’s allegation of “a possible conflict of interest and duty” on nTan’s part. He pointed out that, even though the CA had said in an earlier decision on the Scheme (viz, The Royal Bank of Scotland v TT International) that a scheme manager was “inherently in a position of conflict” (at [77]) because he would be remunerated for successfully resuscitating the company, there was nonetheless no objection to a scheme manager adjudicating on the proofs of debt and performing his other functions. If this was right, the SM submitted, it should make no difference that nTan was also receiving remuneration as the Company’s IFA. The SM then spent several paragraphs explaining how nTan, as the Company’s IFA and, later, as the SM (through those of its officers named at [5] above), had in fact acted against its own self-interest in the events leading up to the sanction of the Scheme.

Further correspondence after the parties’ submissions of 14 March 2012

53 The correspondence between the parties and the Registry did not end with the filing of the parties’ written submissions on 14 March 2012. About two weeks later, on 29 March 2012, the MC wrote to the Registry seeking to clarify what it alleged were factual inaccuracies in the SM’s submissions. It stated, among other things, that Rajah & Tann had not acted as counsel for the companies in question in the other schemes of arrangement in which Mr Nicky Tan had been involved, and that it had not been privy to the success fee arrangements in those schemes, much less had it advised on such arrangements.

54 The SM responded on 2 April 2012. He said that the only point he wished to make was that even in the other schemes of arrangement mentioned in his written submissions of 14 March 2012, the financial advisor’s success fee had not been disclosed in the scheme documents. He added that the question of whether or not Rajah & Tann knew about the VAF in this particular case (ie, the VAF as defined at [2] above) was not one which affected his submissions.

55 On 10 April 2012, the MC again wrote to the Registry. It clarified that its allegation of “a possible conflict of interest and duty” on nTan’s part was based on “simple undisputed facts”, including the following: (a) the quantum of the VAF was dependent on the value of debt restructured under the Scheme; (b) the higher the value of the supporting Scheme Creditors’ debt admitted by the SM, the more likely the Scheme would be approved so as to entitle nTan to the VAF; and (c) the VAF had not been disclosed to the Scheme Creditors before the Scheme was sanctioned by the CA. The MC concluded by emphasising that the only issue which it wanted resolved was that of taxation of the VAF, and that it did not intend to raise factual allegations which were irrelevant to this issue.

56 On 11 April 2012, the SM wrote to the Registry pointing out that the MC had not alleged any actual conflict of interest on nTan’s part. He also asked the CA to consider four points, namely:

(a) the MC had had the opportunity to raise issues relating to the terms of nTan’s engagement and its fees – an opportunity which it did in fact take – hence, these issues were res judicata;

(b) the MC did not dispute the SM’s submission that nTan had acted against its own self-interest in some instances in the events leading up to the sanction of the Scheme;

(c) the MC did not dispute the SM’s assertion that the prevailing industry practice was for a company under a scheme of arrangement not to disclose the success fees (if any) payable to its financial advisors; and

(d) the CA should be slow to rewrite the terms of the contractual arrangements freely negotiated between the Company and nTan.

There was no further correspondence relating to the VAF beyond this point.

The VAF Decision

57 On 27 September 2012, the CA released the VAF Decision. The upshot of it was that nTan was not entitled to the full amount of the VAF under its agreement with the Company. The CA first considered the content of the legal obligations owed to the Scheme Creditors by the Company and the SM. With regard to the Company, the CA held that the Scheme Creditors were “entitled to expect to receive accurate information which would allow them to make a holistic assessment” of the merits of the Scheme, and this meant that the Company was under a duty to disclose to the Scheme Creditors “all material information”, including information pertaining to the VAF (at [20]–[21] of the VAF Decision). The CA then held that the Company had breached this duty by failing to disclose the VAF (at [23]).

58 Next, the CA directed its attention to the SM. It affirmed the duties of the SM to act in good faith towards the Scheme Creditors and not to mislead them or suppress material information from them or even be a party to any attempt by the Company to do so (at [25] of the VAF Decision). Moreover, given the SM’s assumption of the “quasi-judicial” role of adjudicating on the admission and rejection of proofs of debt, the CA considered that he owed duties to be “objective, independent, fair and impartial” and not to place himself in “a position of conflict”. The CA considered that such a conflict had arisen in the present case, in that the quantum of the VAF which would accrue to nTan was dependent on the value of the debts adjudicated upon by Mr Nicky Tan, nTan’s “controlling shareholder” (at [26]). The CA rejected as irrelevant the distinction which the SM had sought to draw “between the VAF accruing to nTan in its role as a ‘financial advisor’ vis-à-vis that of a ‘scheme manager’” (at [27]).

59 Having found that both the Company and the SM had breached their legal duties to disclose to the Scheme Creditors and the court “all benefits accruing to the proposed SM (or his firm)” (at [29] of the VAF Decision), the CA turned to consider what the consequences of those breaches should be. It opined that, “[o]rdinarily, the Scheme should be set aside and put to a fresh vote because it might not have been approved by the [Scheme Creditors] if they had known about the VAF”. But, the CA concluded that it would not do so because such a course was “not practical … without causing more harm to the Company and [its] creditors” (at [33]).

60 Instead, the CA ordered (at [34] of the VAF Decision) that:

(a) the SM/nTan, the Company and the MC were to endeavour to agree on the amount of professional fees that ought properly to be awarded to nTan for its efforts in reviving the Company; and

(b) in the event that the parties were unable to reach agreement, nTan’s global fees would fall to be assessed by the High Court.

We shall refer to these orders as “the VAF Orders”. The CA went on to direct that an assessment of nTan’s fees would involve considering, first and foremost, the value contributed by nTan, while other factors to take into account would include the nature of nTan’s work, the time spent, the scope of work and the assistance in fact provided by nTan’s employees (at [35]).

nTan’s application to set aside the VAF Decision

61 Shortly after the VAF Decision was released, nTan filed two summonses, namely, Summons No 5682 of 2012 (“SUM 5682/2012”) and Summons No 6520 of 2012 (“SUM 6520/2012”). Both summonses contained the same prayers. The first prayer was for leave to intervene in CAs 44 and 47 of 2010 – as mentioned earlier, these were the appeals lodged by the Scheme Creditors opposed to the Scheme against Prakash J’s decision to sanction the Scheme. nTan had to apply for leave to intervene because it was not named as a party to the appeals. The second prayer was for the VAF Decision to be set aside in its entirety. Thereafter, the MC, through the agency of three of the Scheme Creditors (including OCBC and DBS), also sought leave to intervene in CAs 44 and 47 of 2010 for the purpose of opposing nTan’s application in SUM 5682/2012 and SUM 6520/2012 to set aside the VAF Decision. By agreement of the parties, nTan and the MC were permitted to intervene in the appeals, and so it is that nTan’s application to set aside the VAF Decision has come before us, opposed by the MC as well as the Company.

62 nTan’s application to set aside a decision of Singapore’s apex court is a rare one. Nevertheless, it is not unknown to the law, and indeed, it is common ground between the parties that the CA does have what is frequently called the “inherent jurisdiction” to set aside, in limited circumstances, a previous decision that it made. We should point out that the term “inherent jurisdiction” may not be entirely accurate, and that “inherent power” is the preferred language for the reasons given by this court in Re Nalpon Zero Geraldo Mario [2013] 3 SLR 258 (“Re Nalpon”) at [31]–[41]. For reasons that will shortly become evident, there is significance in this distinction.

63 It is also accepted by all the parties that one circumstance in which the CA may exercise this inherent power to set aside a previous decision that it made is where the manner in which it arrived at that previous decision constituted a breach of natural justice. The authority for this proposition is the decision of this court in Management Corporation Strata Title Plan No 301 v Lee Tat Development Pte Ltd [2011] 1 SLR 998 at [55]. We shall refer to that decision as “Lee Tat (No 2)” in order to distinguish it from an earlier CA decision in the same litigation, Lee Tat Development Pte Ltd v Management Corporation Strata Title Plan No 301 [2009] 1 SLR(R) 875, which we shall refer to as “Lee Tat (No 1)”. We shall make ample reference to both Lee Tat (No 1) and Lee Tat (No 2) later in this judgment.

nTan’s grounds for setting aside and the issues that arise

64 nTan’s setting-aside application in SUM 5682/2012 and SUM 6520/2012 is based on two main grounds. The first is that the VAF Decision was made in breach of the rules of natural justice, in that nTan was not given a proper opportunity to be heard on the VAF-related issues dealt with in that decision. The second is that the CA did not have jurisdiction to make the VAF Decision and/or the VAF Orders, the reason being that the CA’s statutorily-conferred jurisdiction over the Scheme did not extend to creditors of the Company who were not parties to the Scheme, such as nTan. Under the terms of the Scheme, nTan was expressly designated an “Excluded Creditor” whose debts would not be subject to the Scheme and would, instead, be paid by the Company as and when they fell due. This meant, so nTan contended, that the CA lacked jurisdiction to make findings and orders that would directly affect an agreement between nTan and the Company, which was what the CA did in the VAF Decision.

65 Apart from the two main grounds mentioned above, there are possibly two other distinct grounds for nTan’s setting-aside application. We say “possibly” because these two other grounds might instead be subsumed within the second of the two main grounds, viz, that of lack of jurisdiction. The first of these two possible grounds is that in the VAF Decision, the CA ought not to have dealt with issues relating to the non-disclosure of the VAF because those issues had previously been canvassed before the CA and were therefore res judicata. It is not entirely clear whether this res judicata ground is advanced as a subset of nTan’s jurisdiction ground, which was the suggestion made in oral submissions, or as an independent stand-alone ground, which was the position taken in written submissions. As for the other setting-aside ground which may possibly exist, it is that the CA had no power to make the VAF Orders. This contention was put forward under the broad heading of the jurisdiction ground, but it might be more accurate to see it as an alternative submission to the argument that the CA lacked jurisdiction to make the VAF Decision and/or the VAF Orders. We shall deal with both of these possible grounds for setting aside in the course of this judgment.

66 We heard nTan’s setting-aside application on 10 November 2014. The various grounds described above were fully argued before us by nTan, the MC and the Company, and we reserved judgment at the conclusion of the hearing. In the course of considering the matter, it occurred to us that there was a further pertinent issue that the parties had not addressed, which was, broadly, this: is nTan estopped by reason of res judicata from contending before us that the CA lacked jurisdiction and/or power to make the VAF Orders and/or hand down the VAF Decision? If so, nTan’s jurisdiction ground and any others that are either dependent on it or subsets of it would fail at the threshold regardless of any substantive merits. We invited the parties to tender further written submissions on this additional issue; we then decided to convene a further hearing for oral arguments to be made on this issue. The further hearing took place on 6 May 2015, and at its conclusion, we reserved judgment once more.

67 Given the shape of nTan’s contentions and the course that the present litigation has taken, the following issues arise for our decision:

(a) Is there any merit in nTan’s argument that the VAF Decision ought to be set aside on the ground that the issues dealt with therein are res judicata? This covers the first of the two possible setting-aside grounds described at [65] above.

(b) Did the manner in which the CA arrived at the VAF Decision constitute a breach of natural justice so as to warrant setting aside the VAF Decision and/or the VAF Orders? This, of course, covers the first main ground of nTan’s setting-aside application, which we have referred to at [64] above.

(c) Is nTan, by reason of res judicata, estopped from arguing that the CA did not have jurisdiction and/or power to hand down the VAF Decision and/or make the VAF Orders? This covers the additional point referred to at [66] above which we raised after the first hearing of nTan’s setting-aside application, although it also extends into what we have described at [64] above as the second main ground relied on by nTan.

(d) In the event that the preceding issue is answered in the negative, did the CA have jurisdiction and/or power to hand down the VAF Decision and/or make the VAF Orders? This covers what is left of nTan’s second main setting-aside ground at [64] above and also the second of the two possible setting-aside grounds described at [65] above.

We shall discuss these issues in the aforesaid sequence. If issue (c) is answered in the affirmative, it will not be necessary for us to decide issue (d).

Issue (a): nTan’s res judicata ground

68 The arguments made by nTan in respect of its res judicata ground are, in truth, very similar to those which it made in its written submissions to the CA dated 14 March 2012. nTan points now, as it did then, to the fact that the CA sanctioned the Scheme on 13 October 2010 in spite of concerns expressed by OCBC on 5 October 2010 over the possibility of an undisclosed success fee payable to the SM. nTan accordingly contends that the CA must have considered OCBC’s allegations of non-disclosure and lack of transparency, and must have determined, albeit not expressly, “that there [were] no such non-disclosure or transparency issues on the facts”. Hence, the MC should have been estopped by reason of res judicata from raising anew these issues of non-disclosure and transparency, as it sought to do in its letter of 27 January 2012, and therefore, the CA should not have considered those issues. Since it was the MC’s 27 January 2012 letter that led ultimately to the VAF Decision, nTan contends, the VAF Decision itself should never have come into existence.

69 In any event, argues nTan, the fact that OCBC expressed concerns over the possibility of an undisclosed success fee is sufficient to establish that, by 5 October 2010, it was aware of the existence of such a success fee, or at least alive to this possibility. Given this knowledge, nTan submits, it was incumbent on OCBC or any other disgruntled Scheme Creditor to pursue the success fee issue with all due vigour at that point in time. But, the issue was not pursued then even though it ought to have been, and thus, it was an abuse of process for the MC to pursue it belatedly in its 27 January 2012 letter.

70 In our judgment, we need not rule on the substantive merits of nTan’s res judicata ground. Specifically, it is not necessary for us to decide whether the MC was estopped from raising the issues relating to the VAF in its 27 January 2012 letter, or whether it was an abuse of process for the MC to have done so. Even if we were minded to accept, in nTan’s favour, that the MC was precluded from raising the VAF-related issues which it brought up in that letter and that, therefore, the VAF Decision should never have come into being, the critical and dispositive fact is that this was not the view taken by the CA in making the VAF Decision. The CA there clearly did engage with (among other issues) the VAF-related issues mentioned in the MC’s 27 January 2012 letter, and it is evident that it did not consider res judicata to be an obstacle to its doing so notwithstanding nTan’s protestations to the contrary in its written submissions of 14 March 2012.

71 Hence, taken at its highest, nTan’s res judicata ground amounts to no more than an allegation that the CA committed an error in failing to accede to its submissions on res judicata. But, the general rule is that a decision by a court of competent jurisdiction is, unless reversed on appeal, unimpeachable even if it might be wrong. There is sound policy behind this rule, namely, the public interest in the finality of litigation; finality requires an acceptance of the truth that the courts may sometimes decide wrongly (see Lee Tat (No 1) at [72]) and that even apex courts are not immune from this. Indeed, the whole point of the doctrine of res judicata is that erroneous decisions must nevertheless be given due effect, and in this connection, it is worth recounting what Lord Millett said in Mulkerrins v PricewaterhouseCoopers (a firm) [2003] 1 WLR 1937 at [10]:

As between the parties to a judicial decision, however, it does not matter whether the decision is right or wrong. … [Res judicata] is a form of estoppel which gives effect to the policy of the law that the parties to a judicial decision should not afterwards be allowed to relitigate the same question, even though the decision may be wrong. If it is wrong, it must be challenged by appeal or not at all. As between themselves, the parties are bound by the decision, and may neither relitigate the same cause of action nor reopen any issue which was an essential part of the decision. The doctrine comes into its own only when the decision is wrong; if it is right, it merely serves to save time and costs. [emphasis added]

Hence, even if the VAF Decision were afflicted by error, that, without more, would not afford an adequate basis to set it aside.

72 There remains the further argument that, if the MC were indeed precluded by res judicata from raising the VAF-related issues in its 27 January 2012 letter, this would mean that the CA did not have the jurisdiction to determine those issues. nTan appears to have put forward such an argument: in its written submissions, it contended that, in view of the application of res judicata, the CA was “functus” on the VAF-related issues raised by the MC and was “in that sense, without jurisdiction”. It followed, said nTan, that the VAF Decision had to be set aside as a “nullity”. So expressed, the res judicata ground is simply a subset of the jurisdiction ground, as counsel for nTan said in his oral submissions; accordingly, we leave the argument for consideration later in this judgment when we address the jurisdiction ground. We would add that nTan also argued that if the VAF Decision were indeed made without jurisdiction such that it was a nullity, that decision could not itself ever give rise to res judicata. We likewise address this point later.

Issue (b): nTan’s breach of natural justice ground

General observations

73 We turn now to nTan’s breach of natural justice ground, which is also its first main ground for seeking to set aside the VAF Decision. It is not disputed that the concept of natural justice consists of two broad rules that may be expressed as Latin maxims: one is nemo iudex in sua causa, which means “no one shall be a judge in his own cause”, and the other is audi alteram partem, which means “hear the other side”. It is also not disputed that the CA is bound to decide the matters before it in a manner which accords with these rules of natural justice. In the present case, nTan relies only on the audi alteram partem rule; for convenience, we shall refer to it as “the hearing rule”. In the administrative law and arbitration contexts, the hearing rule has been said to require that the parties be given notice of the allegations against them as well as a fair opportunity to be heard and to present their case: see the decisions of this court in Soh Beng Tee & Co Pte Ltd v Fairmount Development Pte Ltd [2007] 3 SLR(R) 86 at [43] and Kay Swee Pin v Singapore Island Country Club [2008] 2 SLR(R) 802 at [7]. We accept that these observations apply, at least conceptually, in the context of an application to set aside a CA decision.

74 However, in this connection, we should develop a point which we made in the context of a challenge to an arbitral decision founded on an alleged breach of natural justice in the arbitral proceedings. In AKN and another v ALC and others and other appeals [2015] 3 SLR 488 at [46]–[47], we repudiated the tendency, which we noticed was taking hold among counsel, to reframe what in substance might be errors of law or fact on the part of an arbitral tribunal as entailing an alleged breach of natural justice. Such a challenge should not be made lightly and, in our judgment, should never be made without counsel being satisfied that it can fairly be put that the decision under challenge was arrived at because of a material breach of one or more of the essential rules of natural justice. This is because such applications often entail a re-litigation of the merits, and so will impose an inevitable and considerable strain on limited judicial resources. But, if this is true in the context of arbitration, then it must apply with even greater force in the context of a decision of the CA, which is the apex court in this country. In our judgment, much more rather than less deference is due such a decision, not least because:

(a) the CA is not an inferior tribunal but the highest court in our land, and can be trusted, as a general rule, not to fail in according the parties procedural justice; and

(b) the CA, being itself the ultimate guardian of procedural justice (subject only to legislation), can be expected to act in a procedurally fair manner, and again, as a general rule, it can be taken to have determined that it has acted in a procedurally fair manner in arriving at its decisions.

In our judgment, it should follow that a challenge based on an alleged breach of natural justice should rarely, if ever, be made in respect of a decision of the CA. The considerations we have already articulated concerning the conservation of scarce judicial resources apply with even greater force in the context of rehearing a matter that has already been considered by the CA, as does the general interest in having finality in litigation.

nTan’s complaints

75 Having made these general observations, we turn to nTan’s breach of natural justice ground proper, which revolves around five specific complaints, as detailed in its written submissions. These are that:

(a) the MC was not directed to make a formal application supported by affidavit for the further directions which it sought from the CA, with the result that nTan did not know the “four corners” of the case against it;

(b) the MC had expressly emphasised in its letter of 14 February 2012 that it was “not seeking a ruling from the [CA] on any factual allegations in relation to the fee arrangements between the Company and nTan and/or the [SM]”, and in those circumstances, nTan could not have expected that the CA would make any findings of fact that were adverse to it in the VAF Decision;

(c) in the lead-up to the VAF Decision, there was no oral hearing to clarify disputed allegations of fact, and the parties were heard only through letters and written submissions;

(d) there was no cross-examination to test disputed facts; and

(e) there was no process of discovery such as would have allowed the parties to discover material supporting their factual allegations.

76 There is no doubt that the MC did not make a formal application supported by affidavit for the further directions which it sought from the CA, and that there was no discovery, cross-examination or oral hearing leading up to the VAF Decision. However, in our judgment, there is no single procedure mandated for resolving all the disputes that might possibly come before the courts. Disputes vary greatly in nature, and it would be wrong to mandate that a particular procedure must be followed in any given case. Sometimes, legislation, whether primary or secondary, prescribes the procedure to be followed in particular types of cases; in such situations, the courts are bound to comply. But, other times, nothing is specifically prescribed by law, and it is then unquestionably within the court’s discretion to determine how it will deal with the dispute before it in a manner which it thinks appropriate and fair. To put it in another way, when no particular procedure has been mandated or excluded by law, a court is not under any unqualified obligation to adopt any specific process or procedural tool available to it for resolving disputes, including the filing of affidavits, discovery, cross-examination and oral hearings. On the contrary, a court possesses a generous measure of discretion to decide which processes and procedural tools it will use or dispense with, and in this regard, it will take into account the nature of the issue it is being asked to decide.

77 There were parts of nTan’s written submissions which suggested that certain processes or procedural tools were categorically or at least generally required under certain conditions. We are unable to accept this. For example, nTan argued that “cross-examination should be permitted and/or ordered where there is conflicting evidence, where a person is accused on the basis of false statements of fact, and/or [where] a ‘grave finding of fact’ is sought against a person”. Such processes as discovery and cross-examination might often assist the courts in resolving factual disputes. But, it would be going much too far to say that their absence in the resolution of a given factual dispute must necessarily or even ordinarily mean that a breach of natural justice has thereby been occasioned. The real question in determining whether there has been a breach of the hearing rule is not what processes and procedural tools the court could have utilised but did not; rather, it is whether the party alleging that it has been prejudiced by an alleged breach of natural justice has in fact been denied a fair opportunity to be heard.

78 In our judgment, the real kernel of nTan’s breach of natural justice complaints may be distilled into two essential points. The first is that the procedure adopted by the CA for resolving the VAF-related issues left it unaware of the exact parameters or the “four corners” of the case against it, which meant that it was not afforded a fair opportunity to meet and refute that case. The second is that the CA made findings of fact in the VAF Decision which were adverse to nTan without going through a proper forensic process, and this deprived nTan of the opportunity to challenge adverse allegations of fact as well as to advance its own account of the facts. We address these two points in turn.

nTan’s knowledge of the case against it

79 On the first point, we do not accept nTan’s allegation that it lacked knowledge of the case against it. Although no formal application was made by the MC for the further directions which it sought from the CA, we are satisfied that the issues in dispute became sufficiently crystallised in the course of the exchange of correspondence between the parties and the Registry. From the very start, it would have been apparent that one key issue raised by the MC was whether the VAF should be taxed: in its 27 January 2012 letter, the MC requested the CA to “direct that the [VAF] be subject to taxation” under [8(j)]. That this was an issue the CA intended to resolve would have been clear by 20 February 2012 at the latest: the Registry’s letter of that date stated that four issues had been raised in the parties’ correspondence, the fourth of which was: “[w]hether a success-fee arrangement termed the Value-Added Fee (‘VAF’) between the Company and the [SM] should be subject to taxation under Para 8(j) [of the Brief Grounds]”.

80 We accept that, although the original issue raised by the MC was whether the VAF should be taxed under a specific paragraph of the Brief Grounds (ie, [8(j)]), the parameters of the CA’s inquiry subsequently broadened such that the focus came to be on the Company’s and the SM’s non-disclosure of the VAF to the Scheme Creditors before the Scheme was sanctioned and the consequences which should flow from that. But, this would have been clear to nTan as early as 7 February 2012, which was when the Registry sent the parties a letter containing the CA’s factual query as to (among other things) whether and, if so, when the VAF had been disclosed to the Scheme Creditors. It would have become clearer still upon the parties’ receipt of the Registry’s 20 February 2012 letter, in which the CA posed three questions of fact and four questions of law to the parties. Reading those questions, it could not have escaped the parties that the CA was concerned about the non-disclosure of the VAF and whether that constituted a breach of some duty; hence the first question of law, which was whether the Company and/or the SM ought voluntarily to have disclosed the VAF to the Scheme Creditors and/or to the court. It would also have been evident from the second question of law that the CA was contemplating the consequences that should flow should any such breach of duty be established, and that one of the consequences under consideration was taxation of the VAF, whether under [8(j)] or pursuant to the CA’s “inherent power”, hence the fourth question of law.

81 In our judgment, having read the Registry’s 20 February 2012 letter, nTan would have known that it was being considered:

(a) whether the Company and/or the SM had an obligation of some kind to disclose the VAF to the Scheme Creditors and/or to the court prior to the sanction of the Scheme;

(b) if such an obligation did exist, whether the Company and the SM had breached this obligation, since both the Company and the SM accepted, in their respective letters of 10 February and 14 February 2012, that the VAF had not been disclosed to either the Scheme Creditors or the court prior to the sanction of the Scheme;

(d) whether the court had the power to tax the VAF, either under [8(j)] or within the ambit of its “inherent power”.

82 nTan would have known that, unless it met and refuted the suggestions implicit in the above four questions, the CA might come to the view that those officers of nTan who were named in the Scheme documents as the SM (see [5] above) had breached a duty to disclose the VAF to the Scheme Creditors and the court, and might decide that the VAF should be taxed as a result. Hence, nTan would have known that it should seek to persuade the CA that there was no duty or obligation to disclose the VAF to the Scheme Creditors and the court; and/or that there had been no breach of any such duty or obligation of disclosure; and/or that any such breach should carry no consequences; and/or that in any event, the court had no power to tax the VAF.

83 Furthermore, this was not a case where the CA went off on a tangent and determined in the VAF Decision issues which were different from those that it had identified in the exchange of correspondence leading up to that decision. On the contrary, in the VAF Decision, the CA squarely addressed the issues and questions set out in the Registry’s 20 February 2012 letter. It held that:

(a) the Company and the SM did have an obligation to disclose the VAF to the Scheme Creditors and the court prior to the sanction of the Scheme;

(b) there had been a breach of this obligation;

(c) the ordinary consequence of such a breach was that the Scheme should be set aside, except that it was “not practical” to do so in the circumstances “without causing more harm to the Company and [its] creditors” (at [33]); and

(d) the VAF should be taxed – the word used in the VAF Decision was “assessed” (at [34]), but we consider this to be synonymous with “taxed” – in the event that the parties could not agree on an appropriate amount.

In our judgment, the issues and disputed matters that ultimately formed the basis of the VAF Decision were well within nTan’s cognisance from 20 February 2012 at the latest.

84 In summary, we do not think nTan can plausibly claim that because the MC did not make a formal application supported by affidavit for the further directions which it sought from the CA, it (nTan) was left ignorant of the allegations which it had to meet. The fact is that a substantial amount of correspondence was exchanged following the MC’s letter of 27 January 2012, and in the course of the correspondence, the CA expressly set out the issues that it wished the parties to deal with. In our judgment, that provided nTan with more than adequate notice of the case against it, and we are thus satisfied that there was no breach of natural justice on this score.

Opportunity to be heard on adverse allegations of fact

85 nTan’s other essential point in its breach of natural justice ground is that in the VAF Decision, the CA made a substantial number of findings of fact which were adverse to it without having given it a fair opportunity to be heard on those factual matters. Specifically, it was argued that the absence of discovery, cross-examination and an oral hearing meant that nTan did not have sufficient opportunity to prove the veracity of its own version of the facts or to challenge the facts raised against it. nTan strenuously argued that the adverse findings of fact made by the CA in the VAF Decision were incorrect, and that the CA might have come to a different view had nTan had the chance to put forward its position fully. We now summarise nTan’s attacks on the CA’s findings of fact in the VAF Decision.

86 First, nTan takes issue with a number of more general observations made by the CA on the remuneration of insolvency practitioners. In the VAF Decision, the CA said that insolvency practitioners’ fees came from an “unguarded pocket” (at [20]). nTan argues that the same cannot be said of the VAF because that gave nTan nothing more than an unsecured debt against the Company, meaning that nTan remained at risk of losing the VAF in the event of the Company’s financial collapse. The CA added that insolvency practitioners were given “almost carte blanche to determine (without rigorous oversight) their levels of remuneration even for the most mundane tasks”, such that there was “not infrequently a careless lack of transparency as to how fees are assessed”, leading to the perception that the fees were “sometimes arbitrarily fixed” and “not commensurate with either the efforts rendered or value contributed” (at [31]). nTan argues that these observations are untrue in its case because the VAF had been mutually agreed between it and the Company, and had not been arbitrarily fixed. Moreover, it contends, any lack of transparency was due not to it, but to the Company’s own decision not to disclose the VAF.

87 Second, nTan takes issue with the CA’s characterisation of the VAF in what it considers to be unfairly pejorative terms. The CA said that the VAF had been proposed and agreed to in “dire circumstances” (at [28] of the VAF Decision), and had a “material impact on the financial position of the Company and its ability to carry out the terms of the Scheme” (at [21]). nTan argues that it could not have taken advantage of the Company’s precarious financial situation to push through the VAF since the terms of the agreement between it and the Company had been extensively negotiated and the Company had had the benefit of sound legal advice before entering into that agreement; in addition, nTan argues that since the VAF gave rise to nothing more than an unsecured debt against the Company whereas the Scheme Creditors had a fixed as well as a floating charge over all of the Company’s assets, it is not true that the VAF had a material impact on the Company’s finances. In the VAF Decision, the CA also described the VAF as “an extraordinary amount that [would] leave many breathless”, and said that “the greater the pain endured by the [S]cheme [C]reditors, the greater the gain of the SM” (at [6]). nTan argues that, however great the quantum of the VAF, the fact is that the Company agreed to it; it further argues that it is wrong to think that it (nTan) would seek to bring detriment to the Scheme Creditors in order to benefit itself, pointing to the fact that the intent behind the Scheme was for the Scheme Creditors eventually to recover the entirety of their debts.

88 Third, nTan challenges the CA’s view that the Scheme Creditors lacked knowledge of the VAF prior to the sanction of the Scheme. In the VAF Decision, the CA said that: (a) it was only on 27 January 2012 that the VAF was brought directly to its attention (at [4]); (b) the MC became aware of the VAF only on 22 September 2011 (at [11]); and (c) the SM’s failure to disclose the VAF until 22 September 2011 meant that the Scheme Creditors had not given their “informed consent” to nTan’s continuing to act as the SM despite the position of conflict it was in (at [26]). But, nTan argues, the Scheme Creditors might have known of the VAF at an early stage, given that a number of them had repeatedly asked to see the terms of the agreement between nTan and the Company, and given that they would have known from the Company’s financial results announced on 27 May 2010 that $31m had been spent on restructuring expenses and professional fees. nTan also contends that the requisite majority of Scheme Creditors would have approved the Scheme whether or not they knew about the VAF, as evidenced by the fact that they ultimately approved the Scheme despite certain Scheme Creditors having raised concerns about the professional fees payable in respect of the Scheme. In addition, the CA suggested in the VAF Decision that nTan was able to avoid disclosing the VAF because it was “conveniently an excluded creditor” (at [20]); in response to this, nTan argues that there was nothing inherently sinister about its being an excluded creditor since many of the Company’s other professional advisors, including law firms, were also excluded creditors.

89 Fourth and finally, nTan takes issue with the CA’s finding that there was a conflict of interest on nTan’s part. The CA said that “a conflict had clearly arisen” (at [26] of the VAF Decision), and that nTan had “opted for covertness rather than transparency” (at [37]). nTan argues that the existence of the VAF did not necessarily mean that there was a conflict of interest since it could have earned the VAF through means other than securing the Scheme Creditors’ and the court’s approval of the Scheme; it adds that, in any event, it had in fact acted contrary to its own self-interest in the events leading up to the sanction of the Scheme – for instance, by advising the Company not to agree to a side deal which would have secured a more rapid passage of the Scheme.

90 The chief difficulty in nTan’s way is that, reading the VAF Decision in its entirety, few of the findings of fact which nTan contends were made by the CA, and which it seeks to impugn, appear to have been essential to the CA’s decision. In our judgment, the essential steps in the CA’s reasoning were as follows:

(a) As a matter of law, in schemes of arrangement, the scheme manager has a duty of good faith towards the company’s creditors and a duty not to act in conflict of interest. Moreover, both the company and the scheme manager have a duty to disclose material information to the creditors and the court.

(b) In the present case, the VAF could be considered material information which the Company and the SM had a duty to disclose.

(c) The Company and the SM had failed to disclose the VAF to the Scheme Creditors and the court, and had thereby breached their duty to disclose material information.

(d) The Company’s and the SM’s breaches of the duty of disclosure would ordinarily mean that the Scheme should be set aside and put to a fresh vote.

(e) But, given that it was “not practical” to set aside the Scheme “without causing more harm to the Company and [its] creditors” (at [33] of the VAF Decision), the appropriate solution would be to make the VAF Orders.

91 Given our understanding of the VAF Decision, we consider that that decision rests wholly on two key findings of fact. The first was that the Company and the SM had not disclosed the VAF to the Scheme Creditors and the court prior to the sanction of the Scheme, and the second was that the VAF was material information which should have been disclosed. In our judgment, nTan cannot claim that it was not given a fair opportunity to be heard by the CA on both these matters. The first finding of fact, that of the non-disclosure of the VAF, was in truth not so much a fact found as a fact undisputed: in the Registry’s letter of 7 February 2012, the CA expressly asked the parties whether the VAF had been disclosed to the Scheme Creditors and the court prior to the sanction of the Scheme, and nTan (as well as the Company) acknowledged that it had not.

92 As for the CA’s second key finding of fact – namely, the materiality of the VAF – which is arguably a finding of mixed law and fact, we are satisfied that the CA did afford nTan a fair opportunity to address this by posing to the parties the first question of law set out in the Registry’s 20 February 2012 letter. That question was: “Ought the Company and/or the [SM] [to] have voluntarily disclosed the VAF: i) to the [S]cheme [C]reditors before or at the [S]cheme meeting(s); and/or ii) to the court?” In our view, this question was unmistakably an invitation to nTan to make the argument that: (a) there was no obligation on those of its personnel who were named in the Scheme documents as the SM to disclose the VAF either at all or because the VAF would not have any material impact on the Company’s financial position and the ability of the Scheme Creditors to recover their debts; or (b) the Scheme Creditors had already been aware of the existence of the VAF when they voted on the Scheme; or (c) even if the Scheme Creditors had not been so aware, the disclosure of the VAF would not have made a difference to their votes on the Scheme. Hence, it is obvious to us that nTan had every chance to persuade the CA that the VAF did not fall within the category of material information which the SM and the Company had a duty to disclose. Indeed, it seems to us that in one way or another, nTan did put forward most, if not all, of these contentions in its substantive submissions filed on 14 March 2012 (see the summary at [39]–[52] above).

93 In our judgment, nTan’s real dissatisfaction with the VAF Decision springs primarily from its perception that the CA was effectively suggesting that it (and/or those of its officers named in the Scheme documents as the SM) had been guilty of some measure of bad faith or dishonesty – we use these terms loosely – without having taken proper steps to ascertain the factual truth of that suggestion. That is why nTan has taken pains to emphasise its discomfort at such statements as that it was “conveniently” an excluded creditor, or that “the greater the pain endured” by the Scheme Creditors, the greater its own gain. Were this truly the case, we might have had some sympathy for nTan’s position. After all, a professional organisation trades on its integrity. But, in our opinion, the VAF Decision, read holistically, does not and cannot reasonably be taken to suggest that either nTan or those of its personnel designated as the SM was dishonest or acted in bad faith. The essential steps in the CA’s reasoning which we have summarised earlier certainly do not include a finding to that effect. Moreover, on the issue of conflict of interest, we would point out that the CA said only that the SM (whom the CA identified (at [2] of the VAF Decision) as Mr Nicky Tan specifically), as the “controlling shareholder of nTan”, had found himself in “a position of conflict” (at [26] of the VAF Decision), and not that the SM had in fact acted against the interests of the Scheme Creditors. Also, the CA could have removed those personnel of nTan who were named in the Scheme documents as the SM from their office had it formed the impression that they had acted in bad faith or dishonestly; indeed, we would have expected that the CA would, almost inevitably, have done so if it had in fact formed that impression. But, as the CA did not do so, this is consistent only with the conclusion that it did not see any substantial want of probity in the discharge by the relevant nTan personnel of their duties as the SM. In these circumstances, given that no finding of dishonesty or bad faith was made against nTan either in its capacity as the Company’s IFA or in its capacity (through the designated personnel) as the SM, we are unable to agree that the particular forensic process which the CA adopted to determine the issues before it constituted a breach of natural justice.

94 In our judgment, therefore, notwithstanding the absence of discovery, cross-examination and an oral hearing, there was no breach of natural justice in the manner in which the CA arrived at the VAF Decision because nTan was given a fair opportunity to address the issues and refute the allegations that were essential to the CA’s rulings therein. We think it is material that nTan was able to put in a lengthy set of submissions on 14 March 2012 which not only answered the three questions of fact and four questions of law that the CA had posed by way of the Registry’s 20 February 2012 letter, but also brought up a number of issues that nTan considered relevant, including, for instance, the contention that res judicata prevented the MC from raising the non-disclosure of the VAF. The fact that nTan was able to do so buttresses the conclusion that it was sufficiently aware of the allegations which it had to rebut and had a fair opportunity to be heard on those allegations.

95 We are therefore satisfied that nTan’s breach of natural justice ground does not afford a basis for setting aside the VAF Decision.

Issue (c): Whether nTan is estopped from advancing the jurisdiction ground

96 What remains of nTan’s case, then, is the jurisdiction ground – this is the contention that the VAF Decision should be set aside because the CA had no jurisdiction to hand it down. As we have mentioned above, there is, first, the threshold issue of whether nTan is estopped by reason of res judicata from advancing that contention at all. This is because it may be argued that the CA implicitly but necessarily decided, in handing down the VAF Decision, that it did have jurisdiction to make that decision; alternatively, it may be argued that it would be an abuse of process for nTan to rely on the jurisdiction ground now, given that it could and should have done so at an earlier stage.

97 Thus far, we have used the terms “res judicata” and “jurisdiction” without defining or explaining them. We have now reached a point in this judgment where we must elaborate on these matters. Hence, we turn now to examine the meaning and content of these concepts – beginning with res judicata, and then going on to jurisdiction.

The concept of res judicata

98 As Lord Sumption put it in the UK Supreme Court’s decision in Virgin Atlantic Airways Ltd v Zodiac Seats UK Ltd (formerly Contour Aerospace Ltd) [2014] AC 160 (“Virgin Atlantic”) at [17], res judicata is a “portmanteau term which is used to describe a number of different legal principles with different juridical origins”. For present purposes, we need concern ourselves with only three such “distinct but interrelated” principles, namely: (a) cause of action estoppel; (b) issue estoppel; and (c) what is known as either the “extended” doctrine of res judicata or the defence of “abuse of process”: see the decision of Sundaresh Menon JC in Goh Nellie v Goh Lian Teck and others [2007] 1 SLR(R) 453 (“Goh Nellie”) at [17]–[25]. Broadly speaking, these three principles operate to preclude litigants from making arguments that were previously rejected by a court or tribunal, or that they ought to have advanced on an earlier occasion. Underlying all three principles is the policy that litigants should not be twice vexed in the same matter, and that the public interest requires finality in litigation.

99 Cause of action estoppel is “that which prevents a party from asserting or denying, as against the other party, the existence of a particular cause of action, the non-existence or existence of which has been determined by a court of competent jurisdiction in previous litigation between the same parties”: per Diplock LJ (as he then was) in Thoday v Thoday [1964] P 181 at 197. Notwithstanding the use of the phrase “cause of action”, it is worth clarifying that the leading text on res judicata states that this type of estoppel can arise “whenever a substantive claim is granted or refused even if the claimant has no cause of action in the traditional sense”: see K R Handley, Spencer Bower and Handley: Res Judicata (LexisNexis, 4th Ed, 2009) at para 7.03. In our judgment, in determining whether cause of action estoppel applies, the central inquiry will be directed at whether the later action is in substance a direct attack on and seeks to reverse, other than by way of a permitted appeal, an earlier decision made in relation to a disputed matter between the same parties.

100 Issue estoppel is that which arises “when a court of competent jurisdiction has determined some question of fact or law, either in the course of the same litigation (for example, as a preliminary point) or in other litigation which raises the same point between the same parties”: per Lord Hoffmann in Watt (formerly Carter) and others v Ahsan [2008] 1 AC 696 (“Ahsan”) at [31]. It is thus of wider application than cause of action estoppel, which might perhaps be thought of as a specific instance of issue estoppel in cases where the claim in the later proceedings is identical to that which was determined in earlier proceedings between the same parties.

101 Strictly speaking, cause of action estoppel and issue estoppel apply only in situations where a litigant seeks to re-argue points which have already been the subject of a previous judicial decision in earlier proceedings between the same parties. There is then the litigant who seeks to argue points which were not previously determined by a court or tribunal because they were not brought to the attention of the court or tribunal in the earlier proceedings even though they ought properly to have been raised and argued then. The law recognises that in this latter situation, the litigant – in the absence of “special circumstances” – should also not be permitted to argue those points in order that there might be finality in litigation, and this is where the “extended” doctrine of res judicata comes in. The foundational authority for this is Henderson v Henderson (1843) 3 Hare 100; 67 ER 313 (“Henderson”), where Sir James Wigram V-C said (3 Hare 100 at 114–115; 67 ER 313 at 319):

… [W]here a given matter becomes the subject of litigation in, and of adjudication by, a Court of competent jurisdiction, the Court requires the parties to that litigation to bring forward their whole case, and will not (except under special circumstances) permit the same parties to open the same subject of litigation in respect of matter which might have been brought forward as part of the subject in contest, but which was not brought forward, only because they have, from negligence, inadvertence, or even accident, omitted part of their case. The plea of res judicata applies, except in special cases, not only to points upon which the Court was actually required by the parties to form an opinion and pronounce a judgment, but to every point which properly belonged to the subject of litigation, and which [the] parties, exercising reasonable diligence, might have brought forward at the time. [emphasis added]

102 The “extended” doctrine of res judicata is encapsulated in essence in this passage from Henderson. It is called the “extended” doctrine simply because it extends cause of action estoppel and issue estoppel beyond cases where the point sought to be argued in later proceedings had actually and already been decided by a court in earlier proceedings between the same parties – what we shall call “strict” cause of action or issue estoppel – to cases where the point was not previously decided because it was not raised in the earlier proceedings even though it could and should have been raised in those proceedings. As mentioned earlier, this “extended” doctrine has come also to be known by the name “abuse of process”; and at first glance, that might be confusing because res judicata and abuse of process are “juridically very different”, the former being a “rule of substantive law” and the latter, “a concept which informs the exercise of the court’s procedural powers”, as Lord Sumption notes in Virgin Atlantic (at [25]). However, as Lord Sumption proceeds to explain, res judicata and abuse of process are “overlapping” concepts “with the common underlying purpose of limiting abusive and duplicative litigation”, such that there is no difficulty in conceiving of the “extended” forms of cause of action and issue estoppel as being “concerned with abuse of process” while simultaneously being “part of the law of res judicata”.

103 It is important to determine precisely which of the three res judicata principles – cause of action estoppel, issue estoppel or the “extended” doctrine of res judicata – applies on the facts of a given case because they call for different approaches. With cause of action estoppel, “the bar is absolute in relation to all points decided unless fraud or collusion is alleged”: see the opinion of Lord Keith of Kinkel in the House of Lords decision of Arnold and others v National Westminster Bank plc [1991] 2 AC 93 (“Arnold”) at 104D–104E. In contrast, Lord Keith considered that issue estoppel was a less rigid principle, in that there might be an exception to it “in the special circumstance that there has become available to a party further material relevant to the correct determination of a point involved in the earlier proceedings”, provided that the further material in question “could not by reasonable diligence have been adduced in those [earlier] proceedings” (at 109A–109B). We shall discuss this exception to issue estoppel, which we shall refer to as “the Arnold exception”, in greater detail later; but for now, we should highlight that the Arnold exception, as originally formulated, was a narrow one, meaning that issue estoppel remained a rather robust and rigorous principle even if it was somewhat less so than cause of action estoppel.

104 This may be contrasted with the higher degree of flexibility available to the courts when faced with the “extended” forms of cause of action and issue estoppel. Lord Bingham of Cornhill, in the House of Lords decision of Johnson v Gore Wood & Co (a firm) [2002] 2 AC 1 (at 31D), saw the question of whether a litigant should be estopped from taking a point that could have been raised in earlier proceedings between the same parties not as a “dogmatic” inquiry, but rather, as a “broad, merits-based judgment which takes account of the public and private interests involved and also takes account of all the facts of the case”. This idea was echoed in Goh Nellie at [53], where it was noted that “a court should determine whether there is an abuse of process by looking at all the circumstances of the case”, including whether there is fresh evidence that might warrant re-litigation or whether there are bona fide reasons why a matter was not raised in the earlier proceedings. In this regard, the court is not to “adopt an inflexible or unyielding attitude” (see likewise Goh Nellie at [53]).

105 It is thus evident that there is a marked difference in the analytical approach to strict cause of action and issue estoppel on the one hand and the “extended” forms of those principles on the other. For this reason, where there exists the possibility that a litigant is estopped from raising a matter by reason of res judicata, it is necessary to ascertain whether or not that matter was previously decided – and, in cases where the matter was not previously decided because it was not raised in earlier proceedings between the same parties, whether or not that matter could reasonably have been raised for decision in those earlier proceedings – in order to determine what the relevant principle of res judicata is. As Menon JC said in Goh Nellie at [41]:

… [W]here the issue has in fact been directly covered by the earlier decision, it will be caught either by cause of action estoppel or issue estoppel. As one moves further away from what was directly covered by the earlier decision, then the relevant doctrine becomes the defence of abuse of process rather than issue estoppel.

The concept of jurisdiction and nTan’s jurisdiction ground

Jurisdiction as the court’s authority to hear and determine a dispute

106 Moving to the concept of jurisdiction, this refers to the court’s “authority, however derived, to hear and determine a dispute that is brought before it” [emphasis added]: see Re Nalpon at [13] and [31], citing with approval the decision of Chan Sek Keong J (as he then was) in Muhd Munir v Noor Hidah [1990] 2 SLR(R) 348 at [19]. But, the word “jurisdiction” has also been used to mean other things, and in particular, it is sometimes treated as a synonym of “power”, which Chan J defined in the same passage as the court’s “capacity to give effect to its determination by making or granting the orders or reliefs sought by the successful party in the dispute” [emphasis added]. This is especially apparent when we speak of “inherent jurisdiction”; as we have already noted at [62] above with reference to Re Nalpon at [31]–[41], that in fact means “inherent power”.

107 In the present case, nTan has been careful to indicate that it is first challenging the CA’s jurisdiction in the sense of its authority to hear and determine the dispute which arose in the exchange of correspondence between the parties and the Registry starting with the MC’s letter of 27 January 2012. But, it has also put forward the alternative submission that, even if the CA did have authority to hear and determine that dispute, it lacked the power to make the VAF Orders. We shall come to this alternative submission in due course.

The merits of nTan’s jurisdiction ground

108 Addressing, first, nTan’s challenge to the CA’s authority, we understand its argument to proceed in the following way. The CA, being a creature of statute, only has such authority to hear and determine disputes as statute confers on it. It cannot go beyond the confines of an appeal to determine issues which do not arise within that appeal; nor can it determine the rights and liabilities of persons, especially non-parties, which do not arise for consideration in the appeal. In the present case, CAs 44 and 47 of 2010 were appeals against Prakash J’s decision to sanction the Scheme, and thus pertained strictly to the sanction of the Scheme. The agreement between the Company and nTan – a non-party to the Scheme – on the VAF payable to the latter was a separate matter, and was not part of the subject matter before the CA in those appeals. In short, the CA had authority to hear and determine only a dispute as to whether the Scheme should be sanctioned, and hence, had no authority to hear and determine a dispute relating to a private agreement between the Company and nTan, a creditor which was not even a party to the Scheme.

109 We are willing to assume that, in relation to schemes of arrangement, the High Court has authority to hear and determine only disputes pertaining to the sanction of the scheme before it, and, further, that this would also be the extent of the CA’s jurisdiction. We accept that this authority does not, as a general rule, extend to hearing and determining disputes relating to agreements between the company and any of its creditors, eg, a dispute as to the interpretation of the terms of such an agreement or as to whether there has been a breach of such an agreement. So, for instance, if a company has a contractual dispute with one of its creditors – whether or not that creditor is a party to the scheme – regarding the quality of goods supplied to the company by that creditor, the court’s authority to hear and determine the question of whether to sanction the scheme does not give it authority to hear and determine that contractual dispute.

110 But, we do not think it follows from this that the court, when determining whether to sanction a scheme of arrangement, cannot hear and inquire into matters pertaining to an agreement between the company and a non-party to the scheme if and to the extent it is satisfied that those matters may have a bearing on whether or not the scheme should be sanctioned. It is well-established that, in deciding whether to sanction a scheme of arrangement, the court does not act as a mere rubber-stamp. Its sanction is “not a mere formality”; instead, it must be satisfied not only that the procedural requirements for approval of the scheme by the company’s creditors have been satisfied, but also that the scheme is, in substance, a reasonable one: see our decision in The Oriental Insurance Co Ltd v Reliance National Asia Re Pte Ltd [2008] 3 SLR(R) 121 at [43]. Since the court is concerned in this context with the substance of the scheme, it must follow that it may inquire, for example, into the reasonableness of excluding a particular creditor from the scheme even if the requisite majority of the other creditors have approved the scheme on those terms. The excluded creditor in such a case may not be a party to the scheme, but the debt accruing to that creditor under its agreement with the company could be affected by the court’s decision, for example, to make it a condition of the scheme that that debt should also be subject to the terms of the scheme.

111 In our judgment, in such a case, it cannot reasonably be argued that the court lacks the authority to hear and inquire into such matters since the court is ultimately concerned with whether it will sanction the scheme and may, in that context, have regard to any matter that could reasonably have a bearing on that decision. From these general observations, we move now to the facts of this case. In the course of oral arguments, nTan’s counsel, Mr Edwin Tong SC (“Mr Tong”), accepted that it would have been open to the CA in CAs 44 and 47 of 2010 to refuse to sanction the Scheme unless nTan was brought in as a Scheme Creditor; this must be correct because s 210(4) of the Companies Act provides that the court may grant approval to a scheme subject to such conditions or alterations as it thinks just. In these circumstances, it must have been within the CA’s authority to inquire into the facts and circumstances pertaining to the professional fees payable to nTan, including, in particular, whether the VAF ought to have been disclosed to the court and/or the Scheme Creditors before sanction of the Scheme was sought, and what the remedies should be if such disclosure ought to have been but was not in fact made.

112 The truth is that a host of matters and issues may fairly be considered to be relevant to the ultimate question of whether a scheme of arrangement is to be sanctioned, and if the court has the authority to hear a dispute as to the ultimate question, it must have the authority to hear disputes concerning matters and issues that may potentially be relevant to or have a bearing on that question. Furthermore, in our judgment, that authority neither ceases nor diminishes after the court has sanctioned a scheme; should new facts emerge which cast doubt on the validity of the scheme’s sanction, the court has the authority to determine whether those new facts should have any effect on the scheme. The court may well decide that there should be no change to the status quo or that the new facts are best dealt with by way of a fresh set of proceedings, but the fact remains that the court has the authority to make that decision in the first place.

113 What happened in the present case was that the CA set out to hear and consider a complaint as to the non-disclosure of the VAF and the consequences, if any, that this should have on the Scheme, which it had earlier sanctioned. There is no doubt at all, in our minds, that the CA could have rescinded its sanction of the Scheme on the basis of the non-disclosure of the VAF, and that it could have removed the relevant personnel of nTan from the office of SM if it concluded that their breach of the duty to disclose was such as to undermine their suitability for that office. This power to rescind the sanction of the Scheme on this ground and/or to replace the SM is predicated upon the CA having the authority to hear and inquire into the fact and the consequences of the non-disclosure of the VAF. The fact that the dispute relating to such non-disclosure involved an agreement between the Company and nTan, a creditor excluded from the Scheme, does not detract from the characterisation of the ultimate dispute or question before the CA as being one concerning the continuing validity of its sanction of the Scheme in the light of the non-disclosure of the VAF.

114 Likewise, the genesis of the dispute over the non-disclosure of the VAF does not detract from the foregoing characterisation of the ultimate question before the CA. There is no doubt that this dispute began as a dispute over whether the VAF should be taxed. But, as we have already said (at [80] above), the CA subsequently broadened the inquiry as it corresponded with the parties by letter. That is the fact which matters. Thereafter, having determined in the VAF Decision that the non-disclosure of the VAF did undermine the continuing validity of the sanction of the Scheme, the CA decided that it nonetheless would not set aside the Scheme, but would instead make the VAF Orders. It may be that, as nTan argues, the CA could not have made those orders, but that is an argument which goes only to the CA’s power to fashion remedies and not to its authority to hear the dispute in the first place.

115 nTan also argues that the CA lacked jurisdiction to hear and determine the dispute relating to the non-disclosure of the VAF because the issues pertinent to that dispute were res judicata. This is the point which we have alluded to at [65] and [72] above. In our judgment, this argument too is without merit and we can dispose of it shortly. Res judicata operates against the litigants, and not against the court: it bars the litigants from raising an issue or advancing a contention, and if a party persuades the court that a matter is caught by the doctrine, the court would grant an order giving effect to this; but, the doctrine does not have any effect on the court’s authority to hear the dispute before it and, having heard it, to determine whether or not to uphold the argument that the matter in question is foreclosed by res judicata. Of course, the court might be mistaken in its assessment of that argument, but that does not convert what might well be an erroneous decision into one which was made without jurisdiction. In the present case, nTan says that the CA would have been “functus” if the VAF-related issues dealt with in the VAF Decision were res judicata. We understand this to mean that if those issues ought to have been canvassed earlier but were not, the CA would not even have been able to consider those issues subsequently. In the way in which this argument was put to us, we reject it for several reasons. First, and to reiterate, an argument that a matter is res judicata cannot displace the court’s authority to hear that matter and determine whether it is indeed caught by res judicata. Second, in the present case, it cannot be disputed that the VAF-related issues dealt with in the VAF Decision had not in fact been decided at any time prior to the making of that decision because it was common ground that the VAF had not been disclosed until after the CA issued the Brief Grounds on 13 October 2010, and it was the chain of correspondence triggered off after the MC learnt of the VAF that led to the VAF Decision. Hence, nTan’s argument that the VAF-related issues dealt with in that decision were res judicata was at best founded on the “extended” doctrine of res judicata, and as we have already noted, this is a more flexible doctrine.

116 Hence, we disagree with nTan’s argument that the CA lacked authority to hear and determine the dispute over the non-disclosure of the VAF that arose in the exchange of correspondence between the parties and the Registry starting with the MC’s letter of 27 January 2012. This is sufficient to dispose of the primary thrust of nTan’s jurisdiction argument. However, one point was raised which, although not material, ought to be addressed. This is the point alluded to at [72] above, namely, whether a decision made by a court without jurisdiction can give rise to res judicata. This point is not material given the view we have taken that the CA did not act without jurisdiction (or authority) in hearing the dispute relating to the non-disclosure of the VAF and in determining that dispute as it did in the VAF Decision. But, since the point was taken by nTan on the premise that the CA acted without jurisdiction, we comment on it as follows.

Whether a decision made by a court without jurisdiction can give rise to res judicata

117 The point in question was advanced by nTan relying on the decision of this court in Koh Zhan Quan Tony v Public Prosecutor and another motion [2006] 2 SLR(R) 830 (“Koh Tony”). The accused persons in that case were charged with murder, but were convicted by the trial judge of the lesser offence of robbery with hurt. The Prosecution appealed by way of Criminal Appeal No 2 of 2005 (“CCA 2/2005”) and the CA allowed the appeal, substituting the robbery convictions with murder convictions and sentencing the accused persons to suffer death. Thereafter, the accused persons filed criminal motions arguing that the CA did not have the jurisdiction to hear the Prosecution’s appeal in CCA 2/2005 because the Prosecution could only appeal against the acquittal of an accused by a trial judge, which was not the case there since the accused persons had been convicted by the trial judge of robbery with hurt.

118 The Prosecution took a preliminary objection to the criminal motions, contending that the court could not entertain those motions because it was functus officio (at [11]). The CA rejected that objection on the ground that it was not being asked in the motions to rule on the substantive merits of CCA 2/2005; instead, it was being asked to rule on whether it could even have considered the substantive merits of that appeal at all when the matter went before it. If the Prosecution’s objection were upheld, it would mean that “this court would not even be able to consider whether or not it was acting in excess of its jurisdiction in the first instance” [emphasis in original] (at [14]). The CA went on to observe that if it had in fact acted in excess of its jurisdiction in hearing and ruling on the substantive merits of CCA 2/2005, its decision in that appeal would have been a nullity for want of jurisdiction, and in that situation, if it were to decline to hear the criminal motions, this could have the effect of “sanctioning a nullity” [emphasis in original] (see, likewise, [14] of Koh Tony). Considerations of “logic, common sense as well as of justice and fairness” suggested that this could not be so (at [14]).

119 The CA acknowledged (at [19] of Koh Tony) that the jurisdictional arguments which the accused persons were making in the criminal motions “ought, ideally, to have been raised and considered during the hearing of the appeal in [CCA 2/2005] as a preliminary point of law” [emphasis in original]. But, the court proceeded to say that there was “clear legal authority” for the view that “a party cannot be estopped from arguing that the earlier proceedings were conducted in excess of the jurisdiction of the court concerned” [emphasis in original] (at [19]). The issue of jurisdiction was special because it possessed “a fundamental quality of the highest order”, and the doctrine of estoppel “cannot be pleaded in order to ‘cure’ a decision by a court without jurisdiction – such a decision being, essentially, a nullity” [emphasis in original] (at [20]).

120 nTan argues on this basis that Koh Tony stands for the proposition that a decision can never give rise to res judicata where the court concerned had no authority or jurisdiction to hear the dispute and make the decision in the first place. As we have already observed, in the present case, even if the CA was mistaken in holding in the VAF Decision that: (a) it could order the VAF to be taxed in the context of considering whether to vitiate the sanction earlier granted for the Scheme; and (b) the VAF-related issues dealt with in that decision were not res judicata, these were not matters which went to the jurisdiction (or authority) of the CA to hear and decide the dispute arising from the non-disclosure of the VAF. In that sense, as we have already said at [116] above, whether or not a decision made by a court without jurisdiction can give rise to res judicata is not material.

121 Having said that, we would caution against the expansive reading of Koh Tony canvassed by nTan. Leaving aside for the moment the fact that Koh Tony was a criminal case and that this might thus limit its applicability to civil cases, Koh Tony was, at best, a case of “extended” and not strict cause of action or issue estoppel as it was a case in which the court had not previously decided the jurisdictional point which the accused persons were seeking to make. It is therefore not authority for the broad proposition that a decision made by a court without jurisdiction can never give rise to res judicata in its strict form. On the contrary, in civil cases at least, it is uncontroversial that if a court has expressly decided – perhaps as a preliminary point – that it does have authority to hear and determine a dispute, that decision does give rise to cause of action estoppel or issue estoppel even if the court is mistaken and the true position is that the court does not in fact have such authority. Mr Tong accepted as much. The case of Ahsan (cited at [100] above) is a useful illustration of that proposition in the context of issue estoppel, and it is no less so merely because it concerned the decision of an administrative tribunal rather than that of a court.

122 The applicant in Ahsan (“Ahsan”) was a councillor of the Labour Party. Certain allegations were made against Ahsan and other councillors, but Ahsan was cleared of any misconduct following an inquiry by the Labour Party. Despite that, he was not subsequently chosen by the Labour Party to stand as a candidate for election in forthcoming local government elections. He responded to this by making a complaint to an employment tribunal alleging that the Labour Party had discriminated against him on racial grounds, contrary to s 12(1) of the Race Relations Act 1976 (c 74) (UK) (“the 1976 Race Relations Act”), which provided that it was unlawful for a “qualifying body” – viz, a body “which can confer an authorisation or qualification which is needed for, or facilitates, engagement in a particular profession or trade” – to discriminate against a person on racial grounds. In resisting Ahsan’s complaint, the Labour Party argued that it was not such a “qualifying body”.

123 The employment tribunal decided as a preliminary point that the Labour Party was a “qualifying body”. In the meantime, Ahsan had instituted a second complaint against the Labour Party in relation to a subsequent event, this time alleging victimisation; he was later to put forward a third complaint, this one also based on racial discrimination. The question of whether the Labour Party was a “qualifying body” was common to all three complaints. Returning to the preliminary point decided by the employment tribunal, the Labour Party appealed to the Employment Appeal Tribunal (“the EAT”), but its appeal was dismissed. The EAT gave the Labour Party leave to appeal to the English Court of Appeal, but this was not taken up by the Labour Party, and all three of Ahsan’s complaints went on to be heard by the employment tribunal on the substantive merits.

124 In the meantime, in unconnected proceedings, the Labour Party argued the very same point in relation to s 12(1) of the 1976 Race Relations Act – viz, that it was not a “qualifying body” – and this went up to the English Court of Appeal, which held in favour of the Labour Party. The Labour Party brought this development before the employment tribunal in the proceedings concerning Ahsan and invited the tribunal to dismiss Ahsan’s applications on the basis that it did not have jurisdiction to proceed with the case. The employment tribunal declined that invitation. The Labour Party appealed to the EAT, which allowed the appeal in part. The Labour Party then appealed to the English Court of Appeal, which allowed the Labour Party’s appeal in its entirety. Thereafter, Ahsan appealed to the House of Lords, which unanimously allowed his appeal.

125 Lord Hoffmann delivered a judgment with which all the members of the court agreed. He confirmed that the Labour Party was not a “qualifying body” and that the EAT, in answering the preliminary question, had therefore erred (at [28]). The question was whether that erroneous ruling nonetheless bound the Labour Party in the proceedings relating to Ahsan. Lord Hoffmann noted that the majority of the English Court of Appeal had held that the EAT’s ruling was not binding because it involved an error as to the employment tribunal’s jurisdiction, and it was thought that “a party could not be estopped by conduct from disputing a tribunal’s jurisdiction” (at [29]). But, said Lord Hoffmann, this was a case in which the employment tribunal at first instance had reached an actual decision that it had jurisdiction, and that made all the difference (at [30]). In those circumstances, the employment tribunal’s decision was binding upon the Labour Party and Ahsan by reason of issue estoppel even though it was wrong (at [33]), and there was no reason to invoke the Arnold exception (at [34]). We shall discuss the Arnold exception later; but, what is significant for present purposes is the House of Lords’ holding that, even though the employment tribunal and the EAT had, strictly speaking, decided the preliminary question only in relation to Ahsan’s first complaint, their ruling on that question was binding in relation to his second and third complaints as well.

126 It is thus evident that, in civil cases at least, an erroneous decision by a court that it has jurisdiction gives rise to strict cause of action or issue estoppel. Admittedly, the situation may be less clear where the question of the court’s jurisdiction is not the subject of any previous decision, such that the relevant principles of res judicata are no longer strict cause of action or issue estoppel but the “extended” forms of those doctrines. This is where it may be contended that Koh Tony stands for the proposition – at least in relation to criminal proceedings – that, if the court has not previously made a decision as to its jurisdiction, a party may never be estopped from subsequently challenging that court’s jurisdiction. We accept that the CA in Koh Tony expressed itself in rather categorical and all-encompassing terms in saying that a party “cannot be estopped” [emphasis in original] from attacking the jurisdiction of a court because any decision by a court without jurisdiction would be a “nullity” [emphasis in original] (at [19]–[20]). In our view, however, notwithstanding the apparent generality of these words, they must still be considered within the factual context of Koh Tony. This is because, speaking generally, a court’s reasoning is directed at the factual context before it and the particular circumstances of the dispute it is being called upon to resolve.

127 The relevant context of Koh Tony is that it was a criminal case, and one that was quite literally a matter of life and death. Criminal cases and civil cases may involve different considerations. Among the more obvious of these is that in criminal cases, what is at stake is very often an individual’s liberty, and, sometimes, even his life. There is also the fact that a criminal conviction marks an individual, potentially for life, with the accompanying moral stigma. Given the gravity of the potential consequences and the public interest in ensuring that the exercise of the State’s coercive powers in the realm of criminal justice is beyond reproach, we doubt that any court in the world would uphold a wrongful conviction – let alone a conviction handed down by a court without jurisdiction – on the ground of finality alone. These are not concerns in civil cases. A similar appreciation for this difference between civil cases and criminal cases was demonstrated in a different context by Chao Hick Tin JA in Soh Meiyun v Public Prosecutor [2014] 3 SLR 299, where he considered (at [13]–[15]) that an appellate court’s readiness to admit additional evidence on appeal in a criminal case would not necessarily be constrained to the same degree as in a civil case.

128 Another difference between criminal cases and civil cases was articulated by Lord Hailsham of St Marylebone in the House of Lords decision of Director of Public Prosecutions v Humphrys [1977] AC 1 at 32G–33B, namely, that the parties in civil cases are “on an equal footing”, but in criminal cases, the accused “requires to be protected against oppression by the executive”. Moreover, in criminal cases, the State “has interests and duties which are not simply those of a civil litigant”. Because the parties in criminal proceedings are not equally situated, criminal cases may require an approach that allows more flexibility in appropriate circumstances in order to correct the imbalance; the logic of res judicata, however, is that it operates equally against all parties. Hence, res judicata may be less apt in the criminal context, and correspondingly, any generosity that may be extended in the criminal context towards an accused who wishes to reopen a decision against him by taking a belated jurisdictional objection, such as was shown by the CA in Koh Tony, would not be apt in the civil context, where greater stringency would be called for and where considerably more weight will be placed on the interest of finality in litigation.

129 For these reasons, we doubt whether it is indeed the legal position that “extended” cause of action and issue estoppel can never apply in civil cases to bar a litigant from raising a belated jurisdictional objection. In the circumstances of this case, for the reasons set out at [116] above, we need not come to a firm conclusion on this. But, in principle, we do not see why a belated attempt in a civil case to attack the jurisdiction of a court or tribunal cannot be an “abuse of process”, with the result that the litigant attempting to make such an attack would be estopped from doing so. We reiterate that Koh Tony should be viewed in its criminal context – and, specifically, a capital one at that – and we see no reason to read it in so sweeping a manner that it necessarily excludes the possibility of the “extended” doctrine of res judicata applying to preclude a jurisdictional objection outside that context.

Whether nTan is estopped from advancing its alternative argument that the CA lacked the power to make the VAF Orders

130 We commenced this part of our analysis by referring (at [96] above) to the threshold question of whether nTan was estopped from challenging the CA’s finding that it had jurisdiction, in the sense of authority, to hear the dispute over the non-disclosure of the VAF and to decide that dispute in the way which it did in making the VAF Decision. In the event, we did not reach this question because we were satisfied, for the reasons given above, that the CA did have such authority and that nTan was mistaken in contending otherwise. Although it was not necessary for us to do so, we also touched on the further argument by nTan that a decision made by a court without jurisdiction could not give rise to res judicata. What is left is nTan’s alternative submission, namely, that even if the CA did have authority to hear and determine the dispute arising from the non-disclosure of the VAF, it did not have the power to make the VAF Orders. We shall turn to the merits of this contention in the next (and final) section of this judgment (see [194]–[195] below), but the question that falls for consideration here is whether nTan is estopped by any of the res judicata principles from making this alternative submission. In this regard, there is, first, a preliminary point that should be dealt with, even though it was not raised as an impediment to the possibility of the relevant aspect of res judicata being applicable in this case.

131 We have noted above at [61] that nTan became a party to CAs 44 and 47 of 2010 in the formal sense only after it applied for leave to intervene following the issuance of the VAF Decision on 27 September 2012. The MC also applied shortly thereafter for leave to intervene in order to oppose nTan’s application to set aside the VAF Decision. (As we have also mentioned at [61] above, nTan’s and the MC’s respective applications for leave to intervene were both, in the event, unopposed.) Both nTan and the MC were therefore not formally parties to CAs 44 and 47 of 2010 at the time the Brief Grounds were issued on 13 October 2010, and likewise at the time the VAF Decision was released on 27 September 2012. However, it was the matters raised by the MC after the Brief Grounds were released – starting with its letter of 27 January 2012 – which triggered the chain of correspondence that culminated in the VAF Decision. The key issue which emerged from that chain of correspondence was directed squarely at nTan and the VAF that it wished to enforce. The MC squarely challenged nTan’s entitlement to the VAF, essentially on the ground that the VAF had not been disclosed before the Scheme was sanctioned, and the CA ruled on the merits of that matter in the VAF Decision. Hence, we are satisfied that, as a matter of substance, nTan and the MC were both party to the dispute arising from the non-disclosure of the VAF even though they were not parties (in the formal sense) to CAs 44 and 47 of 2010 at the time that dispute arose. In considering the applicability of res judicata, it is substance, rather than form, which matters. We thus find that nTan and the MC were each capable of being caught by the applicable aspect of res judicata in relation to the CA’s rulings in the VAF Decision.

132 Turning to the real complaint that nTan raises with regard to the CA’s alleged lack of power to make the VAF Orders, in the VAF Decision, the CA did not state in express terms that it was deciding that it had the power to order that (among other things) the VAF be taxed. But, in our judgment, having regard to the development of the correspondence between the parties and the Registry beginning with the MC’s letter of 27 January 2012, the inevitable conclusion is that the CA did indeed decide just that. The reason we say this is that in the Registry’s letter of 20 February 2012, the fourth question of law which the CA posed to the parties was: “Does the Court have the inherent power to tax the VAF? If not, does it fall within Para 8(j) [of the Brief Grounds]?” It is thus evident that the CA addressed its mind to the question of its power to tax the VAF. The parties then argued the point in their respective written submissions, with nTan contending that the CA had no such power because the VAF had been freely negotiated and agreed on between it and the Company, and because the VAF, being a fee due to it as the Company’s IFA, stood outside the terms of the Scheme. Having regard to the fact that the CA did consider the question of its power to tax the VAF and that arguments in support of a negative answer to this question were put before it, and taking into account, further, the fact that the CA eventually decided to order that the VAF be taxed (or, in its own words, “assessed”) if the SM/nTan, the Company and the MC failed to agree on an appropriate sum, the necessary and inescapable conclusion, in our judgment, is that the CA did decide that it had the power to make the VAF Orders.

133 It follows that the CA’s decision to make the VAF Orders could give rise to some form of res judicata against nTan, but it is necessary to first establish which particular form of res judicata applies. In our judgment, the relevant principle in this context is cause of action estoppel, even though the dispute before the CA did not involve a cause of action in the traditional or orthodox sense. This is because the MC, in its correspondence with the Registry starting with its 27 January 2012 letter, in effect sought an order that was directed at denying nTan’s right to enforce or recover the VAF agreed to by the Company, save to the extent that this was permitted by the court. The CA arrived at a decision on this very matter in the VAF Decision, of which the VAF Orders form a part. nTan now seeks to set aside the VAF Decision and, as a corollary, the VAF Orders; and to that end, it has mounted a direct assault on the VAF Decision. Since nTan’s setting-aside application involves a direct assault on the VAF Decision and the VAF Orders, the subject matter of its application is necessarily identical to the subject matter of the proceedings which culminated in the VAF Decision and the VAF Orders. It is this identity of subject matter that makes the applicable aspect of res judicata in this case cause of action estoppel rather than issue estoppel. We are therefore satisfied that nTan’s argument as to the CA’s lack of power to make the VAF Orders is caught by cause of action estoppel.

134 According to Lord Keith in Arnold, cause of action estoppel constitutes an absolute bar to re-litigation in relation to all the points or matters actually decided, except in cases of fraud or collusion. But, nTan brought to our attention an English case that appears to suggest otherwise, namely, R (on the application of East Hertfordshire District Council) v First Secretary of State [2007] EWHC 834 (Admin), a decision of Sullivan J (as he then was) in the English High Court. That case concerned a local planning authority’s application for judicial review of the decision of an Inspector of the Planning Inspectorate quashing an enforcement notice issued by the authority against the owners of a bed and breakfast (“the B&B”). The authority had issued the enforcement notice alleging a breach of planning control in respect of the B&B.

135 The planning authority had in fact issued two separate enforcement notices in respect of the B&B, both of which were then quashed by Inspectors. The first notice was quashed by the first Inspector because the authority had failed to submit relevant information in support of its allegation of breach of planning control. The first Inspector, in quashing the notice, said that his decision did not prevent the authority from issuing a second enforcement notice in respect of the same allegation. That was what the authority did. But, the second notice was quashed by a second Inspector on the basis that the relevant legislation did not permit the issuance of a further notice after a previous one had been quashed. The planning authority subsequently applied for judicial review of the second Inspector’s decision, and that was the application before Sullivan J.

136 The question of cause of action estoppel arose because it was arguable that the first Inspector’s decision to quash the first enforcement notice gave rise to such an estoppel, which would have meant that the planning authority was, before the second Inspector and Sullivan J, estopped from challenging the quashing of the second enforcement notice. But, Sullivan J held that there was no cause of action estoppel because the first enforcement notice had been quashed not on substantive grounds – ie, the first Inspector had not made any finding that there was no breach of planning control – but on the ground that there had not been enough information for the first Inspector to make a decision either way. The true position was, thus, that the first Inspector had been “unable to adjudicate upon an issue because of a complete lack of information”, and cause of action estoppel could not arise in those circumstances (at [19]).

137 Having held that cause of action estoppel did not apply, it was not strictly necessary for Sullivan J to go further. Nevertheless, he proceeded to consider what the position would be if cause of action estoppel did apply, and he addressed the question of whether the Arnold exception could be invoked. It was in this context that Sullivan J held that the Arnold exception was not limited to issue estoppel, but applied also to cause of action estoppel (at [21]); he seemed, moreover, to reject the contention that the Arnold exception applied only to cause of action estoppel in its “extended” or “abuse of process” form but not in its strict form (at [22]). He acknowledged that, in any event, the case before him was one of “extended” and not strict cause of action estoppel (at [23]). Sullivan J then found that the circumstances of the case before him were sufficiently “special” to warrant the application of the Arnold exception. Such “special circumstances” comprised: (a) the fact that there had been no positive decision that the planning authority’s allegation of breach of planning control was unmeritorious; (b) the public interest at stake in ensuring that planning control was adhered to; and (c) the fact that there would be no injustice to the owners of the B&B as they retained the right to persuade an Inspector that there was in fact no breach of planning control.

138 In so far as Sullivan J appears to have suggested that the Arnold exception could be invoked even in the context of strict cause of action estoppel, we do not, with respect, accept his view. This is not merely because his view might be considered obiter dicta twice over in that: (a) he had already decided that cause of action estoppel did not apply to bar the planning authority from challenging the quashing of the second enforcement notice because the decision to quash the first enforcement notice was in effect a “non-decision” on the grounds of inadequate information; and (b) he considered that the case before him concerned cause of action estoppel in its “extended” rather than its strict form. Rather, it is because Sullivan J purported to find support in Arnold for his view when we do not see any basis for this. Sullivan J quoted (at [22]) the passage in which Lord Keith remarks generally that, since estoppel is meant to “work justice” between the parties, it is open to the courts to “recognise that in special circumstances inflexible application of it may have the opposite result” (see Arnold at 109B–109C). This seems to have been Sullivan J’s basis for saying that even strict cause of action estoppel was subject to the Arnold exception. But, in our judgment, Lord Keith’s general remarks must be read subject to his more specific exposition that the only exception to strict cause of action estoppel is “fraud or collusion” which would “justify setting aside the earlier judgment” (see Arnold at 104D–104E). Sullivan J’s view, with great respect, was therefore based on shaky ground, and, in our judgment, it should not be accepted in our jurisdiction.

139 We thus affirm Lord Keith’s statement in Arnold that where strict cause of action estoppel applies, “the bar is absolute” save where fraud or collusion is alleged (at 104D–104E); we would add that the absolute nature of the bar was reiterated by the UK Supreme Court in Regina (Coke-Wallis) v Institute of Chartered Accountants in England and Wales [2011] 2 AC 146 (at [27]). Returning to the facts of the present case, since the CA, in making the VAF Orders, actually decided that it had the power to make those orders, and given that no fraud or collusion is alleged, the bar is absolute and nTan is estopped from contending otherwise now. Therefore, nTan’s argument that the CA did not have the power to make the VAF Orders does not furnish a basis for setting aside those orders.

Observations on the Arnold exception

140 Given that we have held that nTan is precluded by cause of action estoppel from taking issue with the CA’s power to make the VAF Orders, it is not necessary for us to consider the principle of issue estoppel and the precise limits of the Arnold exception. But, we nonetheless state our views on this because full arguments were made by the parties, and because we wish to emphasise that the courts should be exceedingly slow to permit the re-litigation of decided issues in civil cases in view of the importance of finality in civil litigation. We begin by considering the decision in Arnold itself.

The decision in Arnold

141 Arnold concerned a 32-year lease with an initial yearly rent of £800,000. The rent was subject to review every five years. The rent from each review date was to be whichever was the greater of: (a) the rent payable during the last preceding rent period; and (b) the “fair market rent” as at the review date. The “fair market rent” was defined as the rent at which the leased premises “might reasonably be expected to be let at the relevant review date in the open market … without a premium with vacant possession and subject to the [same] provisions” as those which applied to the 32-year lease. In this way, each rent review was to be carried out partly by reference to a hypothetical lease for the unexpired residue of the term. If the parties could not agree on the “fair market rent”, this was to be determined by arbitration.

142 When the time came for the first rent review in 1983, a dispute arose between the landlords and the tenants on an issue of construction. This issue was whether the “fair market rent” should be fixed on the basis: (a) that the hypothetical lease contained the same provision for five-year rent reviews as the actual lease, in which case, the “fair market rent” would be £1,003,000 per annum; or (b) that the hypothetical lease contained no such provision for rent reviews, in which event, the “fair market rent” would be £1,209,000 per annum. The arbitrator held in favour of the first construction, but on appeal, Walton J decided that the second construction was the correct one. The tenants sought leave to appeal to the English Court of Appeal, but Walton J declined to grant them the certificate which they needed to take the matter further and there was no further recourse against his refusal to do so. Subsequent to Walton J’s decision, there came a number of cases in the English High Court and the English Court of Appeal in which the same issue of construction arose. In all those cases, the courts, contrary to Walton J, held in favour of the first construction.

143 On the strength of these subsequent cases, the tenants sought to reopen Walton J’s decision on the issue of construction when the second rent review came around in 1988. They brought an action seeking, among other things, a determination of the true construction of the rent review clause. The landlords applied to strike out the action on the ground that the tenants were barred by issue estoppel from bringing the action. The landlords’ striking-out application was dismissed at first instance, and the House of Lords unanimously affirmed the result.

144 The leading judgment in the House of Lords was that of Lord Keith, with whom the other members of the court agreed. Lord Keith kept in mind two sets of distinctions. The first was the distinction between cause of action estoppel and issue estoppel, and the second was the distinction between a case in which a previous court had actually decided the point sought to be raised subsequently and a case in which a previous court had not decided the point because it had not been raised then. This second distinction corresponds to that between strict cause of action and issue estoppel on the one hand and the “extended” form of those doctrines on the other.

145 Arnold was a case of strict issue estoppel. This was because the tenants were seeking to reopen an issue that had been expressly decided by Walton J, which was whether the hypothetical lease contained the same provision for rent review as that in the actual lease. It was not a case of strict cause of action estoppel because it involved a different rent review from the first rent review in 1983 that had given rise to Walton J’s decision. It was also a case characterised as involving a litigant seeking to bring forward further relevant material which he could not by reasonable diligence have adduced in the earlier proceedings. That further material consisted of the subsequent judicial decisions which took a different view from Walton J on the issue that the tenants were seeking to reopen, namely, the correct construction of the term “fair market rent”.

146 Lord Keith began his discussion by noting the absolute nature of the bar created by strict cause of action estoppel (at 104D–104E). This was then contrasted with the “extended” form of that doctrine: where a point was not decided by a previous court because it had not been raised, Lord Keith said, cause of action estoppel “may not apply in its full rigour” (at 105B). In this regard, he referred to Henderson, which had recognised that exceptions might be made where “special circumstances” were present. Lord Keith then moved on to issue estoppel. He held that the less rigorous Henderson approach to “extended” cause of action estoppel was also applicable to “extended” issue estoppel, meaning that there were “special circumstances” under which the “extended” form of issue estoppel would not operate (at 107C–107D).

147 From “extended” issue estoppel, Lord Keith turned finally to strict issue estoppel. He considered that there was good reason to treat strict cause of action estoppel and strict issue estoppel differently, the reason being that in the former, “the subject matter of the two proceedings [is] identical”, whereas in the latter, “the subject matter is different” (at 108G). Hence, Lord Keith considered that strict issue estoppel need not be as absolute a bar as strict cause of action estoppel. Once this was accepted, said Lord Keith, “it is hard to perceive any logical distinction between a point which was previously raised and decided and one which might have been but was not” in the context of proceedings in which a litigant sought to bring forward further material which could not by reasonable diligence have been adduced in earlier proceedings between the same parties (at 108H). Given that the further material was not available to the litigant at the time of the earlier proceedings, Lord Keith stated, it was “hard to see why there should be a different result according to whether [the litigant] decided not to take the point, thinking it hopeless, or argued it faintly without any real hope of success” (at 108H–109A). The crux of Lord Keith’s point is that when an issue has been decided without the benefit of particular evidence or arguments, and it was reasonable in the circumstances for those matters not to have been raised in the earlier proceedings, the case for the application of res judicata, at least in relation to strict issue estoppel, may be less compelling.

148 Lord Keith in fact concluded that there should be an exception to issue estoppel “in the special circumstance that there has become available to a party further material relevant to the correct determination of a point involved in the earlier proceedings”, provided that this further material “could not by reasonable diligence have been adduced in those [earlier] proceedings”. This exception applied “whether or not that point was specifically raised and decided” (at 109B), meaning that both the strict and the “extended” forms of issue estoppel were subject to the exception. We reiterate that Lord Keith was speaking of a narrow category of strict issue estoppel cases where, although the broad issue had in fact been decided in earlier proceedings between the same parties, a particular argument or a particular piece of evidence that could have been (and, in Arnold’s case, was) decisive had not been raised or adduced then, and the failure to raise that argument or adduce that evidence was an excusable one because that argument or evidence would not have been available at the time of the earlier proceedings despite the exercise of reasonable diligence.

149 Attending to the facts of Arnold, Lord Keith opined that the subsequent judicial decisions disapproving of Walton J’s view could “not entirely inappositely be described as a change in the law” (at 109C). The necessary implication is that Lord Keith thought that this change in the law could constitute “further material” for the purposes of the exception to issue estoppel that we have outlined above. But, over and above this, it was necessary to establish whether there were “special circumstances” to warrant an exception to issue estoppel being made. Lord Keith concluded that the case did present “special circumstances”, such that the tenants should be permitted to reopen a previously-decided point, for the following reasons:

(a) the tenants had no right of appeal against the decision of Walton J, who had erroneously refused to grant a certificate that the case included a question of law of general public importance, which certificate was necessary before the tenants could file an appeal (at 110D–110E);

(b) it would be “most unjust” that the tenants should be faced with a succession of rent reviews over a period of more than 20 years, all proceeding upon a construction of the lease which was highly unfavourable to them and generally regarded as wrong (at 110E);

(c) the landlords would “most unfairly” be receiving a very much higher rent than what they would be entitled to on a correct construction of the lease (at 110F); and

(d) the public interest in seeing an end to litigation was of little weight given that there would, in any event, have to be arbitration at each rent review if the parties could not agree on the “fair market rent” (at 110F–110G).

150 It will be evident from what we have said, as well as from the unique circumstances in Arnold and the reasoning of Lord Keith, that the exception to issue estoppel created by that case was conceived as a very narrow one requiring at least the existence of further material which could have had a significant bearing on the earlier decision, and which was not and could not by reasonable diligence have been adduced in the earlier proceedings, such as a change in the law. It might also be said that the Arnold exception, as framed in its original narrow manner, requires the absence of an effective right of appeal – indeed, Menon JC in Goh Nellie (at [44]) considered this to be the “decisive factor” in Arnold – which would exclude the application of the exception in cases where the litigant had an opportunity to appeal but did not avail himself of it (see [162]–[163] below). Certainly, there must be shown to be a sufficient degree of injustice so as to outweigh the public interest in the finality of litigation, such as the fact that the litigant would have to live with the consequences of an erroneous decision for a prolonged future period of time if an exception to issue estoppel were not made.

151 Furthermore, the Arnold exception, as originally framed, would affect only future or prospective rights, and would not permit a litigant to reach into the past and undo the effects of a previous decision. Before Arnold reached the House of Lords, Mann LJ in the English Court of Appeal (see Arnold and others v National Westminster Bank plc [1990] 1 Ch 573) indicated that he was willing to make an exception to issue estoppel only because there was “no attempt to re-open the rent review for 1983 to 1988” (at 600F), which was the rent review giving rise to Walton J’s decision. In other words, there was no attempt by the tenants to reverse the effect of Walton J’s decision, in that they did not claim the excess rent which they had paid pursuant to Walton J’s erroneous construction of the term “fair market rent”. This was, of course, correct because Walton J’s decision was governed by cause of action estoppel and, hence, was incapable of being reopened and re-litigated. And the Arnold exception, as originally conceived, would not have permitted the tenants to do that.

The development of Arnold in the UK

152 As we have observed, at one level, Arnold can be seen as creating a very narrow exception to issue estoppel circumscribed by the peculiar circumstances of that case. But, on another level, it could possibly be seen as laying down the potentially wide principle that since one of the purposes of issue estoppel is to work justice between the parties, the courts may decline to apply that aspect of the estoppel doctrine in “special circumstances” whenever the inflexible application of that doctrine would lead to injustice instead (see Arnold at 109B–109C). Indeed, the UK cases that came after Arnold have approached it from both perspectives.

153 We turn first to three UK cases which adopted the narrow view of the Arnold exception. On this view, the applicability of the exception requires, as a threshold factor, the emergence of new or further material that could not by reasonable diligence have been adduced in the earlier proceedings which are said to give rise to issue estoppel. That material might consist of new facts or evidence, or a change in the law. In these three cases, the courts refused to permit the invocation of the Arnold exception on the basis that this condition of new and previously-unavailable material had not been met. We describe these cases briefly below.

154 The first is the English Court of Appeal decision of Curling and others v Securicor Limited and another [2001] EWCA Civ 358 (“Curling v Securicor”), which concerned a dispute between a company called Securicor and a number of its former employees. Securicor ran immigration detention centres at Heathrow Airport and Gatwick Airport until the Home Office terminated its services and awarded the tender to a company called Group 4. All but one of the claimant former employees were employed by Group 4 in the same jobs as those in which Securicor had previously employed them. Subsequently, the former employees brought a claim against Securicor – and, in the alternative, against Group 4 – for statutory redundancy payments. At first instance, the employment tribunal which heard the matter dismissed the claim, but on appeal, the EAT (as defined at [123] above) reversed that decision.

155 The relevant point for the purposes of issue estoppel was that at first instance, the employment tribunal found that as between Securicor and Group 4, there had been no “transfer of undertaking” as defined in certain employment regulations, which meant that Securicor had not transferred to Group 4 the duties and liabilities owed to the former employees. The EAT did not interfere with that finding. Securicor applied to the EAT for permission to appeal to the English Court of Appeal, but this was declined. It was open to Securicor thereafter to seek such permission from the English Court of Appeal itself, but it did not do so. Instead, it duly paid the former employees the statutory redundancy payments.

156 The former employees subsequently commenced further proceedings against Securicor for supplementary contractual payments pursuant to an agreement between their trade union and Securicor, again with an alternative claim against Group 4. Securicor’s defence was that there had been a “transfer of undertaking” to Group 4. Both the former employees and Group 4 argued that the raising of this defence was barred by issue estoppel. Securicor’s response was that the EAT’s decision had been overruled by an English Court of Appeal case which came after it (viz, Dines v Initial Health Care Services Ltd [1994] IRLR 336 (“Dines”)), and that this was a change in the law which attracted the Arnold exception. The English Court of Appeal in Curling v Securicor rejected this argument. It held (at [33]) that Dines had turned on a different point and had not overruled the EAT’s decision whether expressly or by implication, and hence, there had been no change in the law; on this basis, there was no question of the Arnold exception arising.

157 Similarly, Henderson J, in the English High Court decision of Littlewoods Retail Ltd and others v Revenue and Customs Commissioners [2014] STC 1761 (“Littlewoods”), rejected an argument that there had been a change in the law such as could give rise to the Arnold exception. That case involved a series of disputes between Littlewoods and the revenue commission in relation to value-added tax (“VAT”). By a decision handed down in October 2001, the English Court of Appeal held that Littlewoods had overpaid certain amounts of VAT and that the revenue commission was liable to repay those principal amounts. The revenue commission did not seek leave to appeal to the House of Lords. Subsequently, in October 2004, the revenue commission conceded a similar liability to repay in respect of other amounts of overpaid VAT.

158 What ensued was further litigation on the question of the revenue commission’s liability to pay compound interest on all the overpaid VAT. In the course of this litigation, the revenue commission sought in March 2013 to amend its defence in order to argue that, contrary to the English Court of Appeal’s October 2001 decision and its own concession of liability in 2004, the VAT paid by Littlewoods had not been overpaid after all. Littlewoods cried foul, asserting issue estoppel. The revenue commission’s riposte was that the reasoning in a 2012 decision of the European Court of Justice, Grattan Plc v Revenue and Customs Commissioners [2013] STC 502 (“Grattan”), led to the conclusion that the VAT paid by Littlewoods had not in fact been overpaid, and that this development in the law brought the matter within the Arnold exception. The revenue commission was permitted to amend its defence, but without prejudice to Littlewoods’ right to raise arguments on issue estoppel at the trial before Henderson J. That was what Littlewoods did.

159 Henderson J did not consider Grattan to have had “anything like the radical impact” on the overpayment issue that the revenue commission was alleging. He held that the real relevance of Grattan was merely that it had “belatedly suggested” to the revenue commission certain arguments which “could perfectly well have been raised as grounds for seeking leave to appeal to the House of Lords” from the English Court of Appeal’s October 2001 decision. Grattan could not be said to have shown that the English Court of Appeal’s October 2001 decision was wrong, nor that “the relevant legal principles had changed in any major respects” (at [157]–[158]). Hence, Henderson J concluded that the circumstances of the case did not “come within measurable distance of the kind of change in the law which may, exceptionally, prevent an issue estoppel from operating”. However, he did accept that “the balance might arguably have come down in [the revenue commission’s] favour” if Grattan had indeed been “a ground-breaking decision which clearly demonstrated for the first time” that the VAT in question had not been overpaid (at [168]).

160 The third case to mention in this connection is JSC BTA Bank v Ablyazov [2013] EWHC 3691 (Ch). There, Henderson J heard the claimant bank’s application for summary judgment against the defendant (“Mr A”). Prior to this, the bank had brought contempt of court proceedings against Mr A, and in those proceedings, the latter had been found to be the true owner of a company called Bubris. This finding was not disturbed on appeal. The bank thereafter instituted civil proceedings against Mr A, out of which arose the summary judgment application before Henderson J. Mr A contended in his defence that he was not the beneficial owner of Bubris.

161 Henderson J held that the finding in the contempt proceedings as to Mr A’s beneficial ownership of Bubris gave rise to issue estoppel in the subsequent civil proceedings. He considered whether Mr A might avail himself of the Arnold exception and concluded that he could not. This was because Mr A had not been able to point to any further material that might cast doubt on the correctness of the finding in the contempt proceedings; moreover, the first instance court in the contempt proceedings had expressly considered the question of whether there was sufficient material for it to adjudicate safely on the question of who the beneficial owner of Bubris was and had answered this question in the affirmative (at [51]).

162 A further aspect of the narrow view of the Arnold exception which has been underscored by the cases is the unavailability of a right of appeal (or any further right of appeal) from the decision said to give rise to the estoppel that is sought to be avoided. In Arnold itself, there was no right of appeal because Walton J declined to grant the tenants the certificate needed to take the case to a higher court. As we have already noted, Menon JC in Goh Nellie opined that the “absence of an effective right of appeal” was “the decisive factor in Arnold” (at [44]). In a similar vein, in the Hong Kong case of Ho Kin Man & ors v Commissioner of Police [2013] 1 HKC 13, Deputy Judge Saunders felt able to say that it was “plain that had there been a right of appeal” in Arnold “which had not been exercised”, the House of Lords “would not have permitted the issue to be reopened as an exception to issue estoppel” (at 25I–26A).

163 In two of the three cases discussed above, it was thought that the litigant’s failure to pursue an avenue of appeal open to him weighed heavily against the application of the Arnold exception even if that failure was not determinative. In Curling v Securicor, the English Court of Appeal considered that the facts before it were “not nearly so compelling” as those in Arnold, partly because Securicor, “when refused permission to appeal by the EAT, did not apply to [the English Court of Appeal] for permission” (at [35]). In Littlewoods, Henderson J took the view that the revenue commission’s decision not to seek leave to appeal against the English Court of Appeal’s October 2001 decision on overpayment of VAT “t[old] strongly against permitting the issues decided in that case to be re-opened” (at [168]).

164 As against this, there is the wide perspective that considers Arnold to be a precedent for making exceptions to issue estoppel more generally in order to avoid “injustice”. This perspective appears to have been adopted in no less than two decisions of the UK’s highest court, albeit not in any unequivocal terms; these decisions are Ahsan (at [34]) and Virgin Atlantic (at [22], specifically proposition (3) therein). A less ambiguous endorsement of the wide perspective can be found in the English Court of Appeal decision of Olympic Airlines SA (in special liquidation) v ACG Acquisition XX LLC [2014] EWCA Civ 821. There, the court declined to read Arnold “as deciding that the special circumstances found to exist in that case are the only type of circumstances that can qualify as justifying an exception to an issue estoppel” (at [68]), citing a practical concern that Lord Reid had raised in Carl Zeiss Stiftung v Rayner & Keeler Ltd and others [1967] AC 853 at 917C–917E, namely:

Suppose the first case is one of trifling importance but it involves for one party proof of facts which would be expensive and troublesome; and that party can see the possibility that the same point may arise if his opponent later raises a much more important claim. What is he to do? The second case may never be brought. Must he go to great trouble and expense to forestall a possible plea of issue estoppel if the second case is brought? This does not arise in cause of action estoppel: if the cause of action is important, he will incur the expense: if it is not, he will take the chance of winning on some other point. It seems to me that there is room for a good deal more thought before we settle the limits of issue estoppel.

165 There is also Yukos Capital Sarl v OJSC Rosneft Oil Co (No 2) [2014] QB 458, in which the English Court of Appeal seemed to think that there was a larger “discretionary in special circumstances” exception to issue estoppel – which could fairly be termed an “amorphous” exception – of which Arnold was but an example (at [159]). In that case, the claimant (“Yukos”) obtained a number of arbitral awards in Russia against the defendant (“Rosneft”). Yukos applied in the Netherlands for leave to enforce those awards. Between the time Yukos put in its application and the time the Dutch courts adjudicated upon it, the Russian courts set aside the arbitral awards. But, the Amsterdam appeal court nevertheless gave Yukos leave to enforce the awards, holding that the Russian setting-aside decisions should not be recognised in the Netherlands because it was likely that the Russian courts, in making those decisions, had not acted impartially and independently, but had instead been guided by the interests of the Russian state and instructed by the Russian executive.

166 Subsequently, Yukos sought to enforce the arbitral awards in London. Rosneft resisted this on the basis that the awards had been set aside by the Russian courts. In response, Yukos argued that, by reason of the decision of the Amsterdam appeal court, Rosneft was estopped from contending that the Russian setting-aside decisions were not or were not likely to be the result of a partial and dependent judicial process, so as to be contrary to natural justice. The English Court of Appeal held that the Dutch decision did not give rise to issue estoppel. But, it went on to opine that, if issue estoppel had indeed arisen, it would have invoked the “discretionary in special circumstances” exception on grounds of international comity – since the matter in question concerned the propriety of Russian decisions, it would be “an abdication of responsibility on the part of the English court” to accept without question the opinion of the Dutch courts that those decisions had not been arrived at impartially and independently (at [160]).

The treatment of Arnold in Singapore in the Lee Tat cases

167 In Singapore, the Arnold exception was considered in some depth by the CA in Lee Tat (No 1) and Lee Tat (No 2). As will become clear, these decisions took the wide perspective of the Arnold exception. Both decisions stemmed from protracted litigation between the management corporation of a condominium known as Grange Heights (“the Grange Heights MC”) and Lee Tat Development Pte Ltd (“Lee Tat”). Grange Heights sits on Lot 687, which is an amalgamation of Lot 111-34 and Lot 561. Lee Tat owns two neighbouring pieces of land, Lots 111-32 and 111-33. Adjoining Lots 111-32, 111-33 and 111-34 is Lot 111-31 (“the Servient Tenement”), which is a long and narrow strip of land serving as a private road between the three lots and Grange Road. By certain conveyances in 1919, a right of way over the Servient Tenement was granted in favour of Lots 111-32, 111-33 and 111-34, all of which may be regarded as dominant tenements. A useful sketch depicting pictorially these plots and the surrounding land may be found in Lee Tat (No 1) at 944.

168 In essence, the litigation between the Grange Heights MC and Lee Tat centred on whether the residents of and visitors to Grange Heights (collectively, “Grange Heights residents and visitors”) were entitled to use the private road on the Servient Tenement as a right of way despite the fact that Grange Heights sat not only on a dominant tenement (namely, Lot 111-34), but also on Lot 561, which was not a dominant tenement in that it had not been granted a right of way over the Servient Tenement. Had Grange Heights sat only on Lot 111-34, there would be no doubt that Grange Heights residents and visitors would, by reason of the right of way over the Servient Tenement, be entitled to use the private road. To complicate matters, Lee Tat became the owner of the Servient Tenement at some point during the years in which the litigation took place.

169 For present purposes, the saga may be taken as having begun in 1989. At that time, Lee Tat was not the owner of the Servient Tenement yet. Lee Tat nonetheless erected an iron gate at one end of the Servient Tenement and a fence at the other, thereby preventing Grange Heights residents and visitors from using the private road on the Servient Tenement. The Grange Heights MC responded by instituting proceedings against Lee Tat seeking an injunction and damages (“the 1989 proceedings”). Punch Coomaraswamy J found in favour of the Grange Heights MC in a decision reported as Management Corporation Strata Title Plan No 301 v Lee Tat Development Pte Ltd [1990] 2 SLR(R) 634. He held that, even though Lot 111-34 had been amalgamated with Lot 561, which was not a dominant tenement, the amalgamation was “only for the purposes of survey and issue of documents of title” and did not “destroy or extinguish the right of way” enjoyed by Lot 111-34, which instead remained intact: at [8]. Therefore, he held, Grange Heights residents and visitors still had a right of way over the Servient Tenement. Lee Tat appealed, but its appeal was unanimously dismissed by the CA (“the 1992 CA”) in a decision reported as Lee Tat Development Pte Ltd v Management Corporation Strata Title Plan No 301 [1992] 3 SLR(R) 1.

170 In 1997, Lee Tat became the owner of the Servient Tenement. In 2004, the Grange Heights MC returned to court seeking a declaration that it was entitled to repair and/or maintain the private road which was its right of way over the Servient Tenement. Lee Tat responded with an application of its own seeking declarations that: (a) the easement granted in favour of Lot 111-34 was not intended to be made appurtenant to Lot 561; (b) the amalgamation of Lot 111-34 and Lot 561 did not result in the conferment of any easement in favour of Lot 561; and (c) the right of way over the Servient Tenement was not to be used for access to and from Grange Heights. Lee Tat also sought an injunction prohibiting Grange Heights residents and visitors from using the private road running through the Servient Tenement.

171 With the consent of the parties, it was ordered that Lee Tat’s application be heard first. At first instance, Woo Bih Li J dismissed the application on the basis that Lee Tat’s claims were barred by issue estoppel (see Lee Tat Development Pte Ltd v Management Corporation Srata Title Plan No 301 [2004] 4 SLR(R) 828). On appeal, Woo J’s decision was upheld, likewise on the ground of issue estoppel, by a majority of the CA consisting of Yong Pung How CJ and Belinda Ang Saw Ean J, with Chao Hick Tin JA dissenting. We shall refer to this constitution of the CA as “the 2005 CA”, and to the proceedings before Woo J at first instance as well as before the 2005 CA on appeal as, collectively, “the 2005 proceedings”. The 2005 CA’s decision is reported as Lee Tat Development Pte Ltd v Management Corporation Strata Title Plan No 301 [2005] 3 SLR(R) 157, and we shall refer to it as “the 2005 Lee Tat CA Decision”.

172 Before the 2005 CA, it was argued on behalf of Lee Tat that issue estoppel did not operate because Lee Tat was suing in the 2005 proceedings as the owner of the Servient Tenement, whereas it had sued only as the owner of two of the dominant tenements (namely, Lots 111-32 and 111-33) in the 1989 proceedings. That, of course, was not the only argument which Lee Tat made; persuading the 2005 CA that issue estoppel did not operate would have meant only that Lee Tat was not precluded from advancing submissions on the substantive question of whether Grange Heights residents and visitors were entitled to use the right of way over the Servient Tenement. Lee Tat would also have to advance submissions on that substantive question, and that is what it did. In arguing that Grange Heights residents and visitors were not entitled to use the right of way, Lee Tat contended that any entitlement which Grange Heights residents and visitors might once have had was extinguished when it (Lee Tat) became the owner of the Servient Tenement; and in this connection, it relied heavily on the case of Harris v Flower (1904) 74 LJ Ch 127, which held that a right of way granted over a servient tenement for the benefit of a dominant tenement could not be used for the benefit of a non-dominant tenement. The majority in the 2005 CA held that the fact that Lee Tat had become the owner of the Servient Tenement made no difference to the rights of Grange Heights residents and visitors over the Servient Tenement, those rights having been decided in the 1989 proceedings: at [16] of the 2005 Lee Tat CA Decision. What this meant was that issue estoppel applied; and in so far as Lee Tat had relied on Harris v Flower to make its substantive argument that Grange Heights residents and visitors did not have a right of way over the Servient Tenement, the majority did not decide whether Lee Tat’s substantive argument was correct. Instead, the majority’s response was simply that Coomaraswamy J and the 1992 CA after him had not followed the principle laid down in Harris v Flower, and that was the end of the matter because, for the purposes of issue estoppel, it was irrelevant whether or not Coomaraswamy J’s and the 1992 CA’s decisions not to follow Harris v Flower were correct: at [25].

173 With Lee Tat’s application thus disposed of, the focus of the litigation turned to the Grange Heights MC’s application for a declaration that it was entitled to repair and/or maintain the private road on the Servient Tenement. Woo J granted the declarations sought, whereupon Lee Tat appealed. The CA (“the 2008 CA”), which was differently constituted from the 2005 CA, unanimously allowed the appeal in Lee Tat (No 1).

174 In Lee Tat (No 1), the 2008 CA characterised the issue before it (“the Main Issue”) as being whether Grange Heights residents and visitors could use the Servient Tenement to cross between Lot 561 and Grange Road through Lot 111-34 (at [11]). This was the same issue which Lee Tat had sought to litigate in the 2005 proceedings, and it was the issue which Lee Tat was said to be estopped from reopening. The 2008 CA considered that this issue had not actually been decided by either Coomaraswamy J or the 1992 CA. For this and other reasons (at [51]–[55]), the 2008 CA held that when the majority in the 2005 CA ruled that issue estoppel precluded Lee Tat from litigating the Main Issue, it was in error.

175 In other words, according to the 2008 CA, the true position was that Lee Tat was not and had never been estopped from litigating the Main Issue. But, the difficulty was that the majority in the 2005 CA had decided, however erroneously, that Lee Tat was estopped from litigating that issue – a difficulty the 2008 CA acknowledged in Lee Tat (No 1) at [16]. The decision of the majority in the 2005 CA itself gave rise to issue estoppel; in sum, Lee Tat was estopped from arguing that it was not estopped from litigating the Main Issue, even though the correct view (as the 2008 CA saw it) was that it was not so estopped.

176 The 2008 CA thus turned to the Arnold exception. Having opined that the exception should be recognised by the Singapore courts (at [78]), it held that the following circumstances constituted such “grave injustice” that it warranted the application of the exception (at [80]–[81]):

(a) neither Coomaraswamy J nor the 1992 CA after him had decided the Main Issue, meaning that the majority in the 2005 CA erred in deciding that that issue had been decided by them;

(b) the majority in the 2005 CA had held that issue estoppel applied even though not all the elements of that form of estoppel had been satisfied;

(c) the decision of the majority in the 2005 CA was not likely to result in finality as there would probably be continuing litigation between Lee Tat and the Grange Heights MC, eg, as to whether the private road on the Servient Tenement could be used for vehicular as well as pedestrian traffic;

(d) the majority in the 2005 CA had “perpetuated an erroneous legal position” in that their holding was contrary to the established law on easements laid down in Harris v Flower;

(e) the Grange Heights MC had been guilty of disingenuous behaviour in that one of the arguments which it raised in the 2005 proceedings was inconsistent with an argument which it had put forward in the 1989 proceedings;

(f) Grange Heights residents and visitors would not be inconvenienced in not being able to use the right of way over the Servient Tenement since they could enter and exit Grange Heights at other locations; and

(g) the erroneous decision of the majority in the 2005 CA had prevented Lee Tat from making optimal use of the Servient Tenement which it owned.

177 The 2008 CA went on to say (at [82]–[83] of Lee Tat (No 1)) that there was a more “fundamental” objection to applying issue estoppel, which was that the Main Issue had “never been decided on the merits” and thus could not “possibly be the subject of an estoppel”. Coomaraswamy J, the 1992 CA after him and the majority in the 2005 CA had not decided the Main Issue on the merits; instead, the 2005 Lee Tat CA Decision had “merely ruled on whether or not the Main Issue had been adjudicated” in the 1989 proceedings. The 2008 CA held that where the issue which was said to be the subject matter of issue estoppel had not in fact been decided on the merits, “this factor alone entails that issue estoppel should not apply”; it “may be considered as another category of ‘special circumstances’ … for the purposes of the Arnold exception”, but at the same time “can and should form a separate and independent basis for not applying the doctrine of issue estoppel”.

178 The litigation between the Grange Heights MC and Lee Tat did not end with Lee Tat (No 1). Approximately seven months after that decision was handed down, the Grange Heights MC applied by way of Summons No 3446 of 2009 (“SUM 3446/2009”) to set aside the 2008 CA’s decision in Lee Tat (No 1) on the basis that that decision had been arrived at in breach of natural justice, in that the Grange Heights MC had not had the opportunity to address the 2008 CA at all on the Arnold exception. As SUM 3446/2009 was premised on the assumption that the CA was not functus officio and could therefore reopen and reconsider the 2008 CA’s earlier decision in Lee Tat (No 1), the Registry directed the Grange Heights MC to file a fresh action to determine this jurisdictional point as a preliminary issue. That action took the form of Originating Summons No 875 of 2009 (“OS 875/2009”), in which the Grange Heights MC sought declarations to the effect that: (a) the CA, as the final appellate court in Singapore, had “the jurisdiction and power to reopen and set aside an earlier decision of its own and to reconstitute itself to rehear and/or reconsider the matters arising therefrom”; and (b) the CA therefore had “the jurisdiction and power to grant the reliefs sought by the [Grange Heights MC] in [SUM 3446/2009]” (see Lee Tat (No 2) at [27]). At first instance, OS 875/2009 was dismissed by the High Court. On appeal by the Grange Heights MC, for reasons which we need not delve into, the CA upheld, in Lee Tat (No 2), the High Court’s decision to dismiss OS 875/2009 as “an exercise in futility” (at [67]). What is relevant for present purposes is that in dismissing the Grange Heights MC’s appeal, the CA made some comments in Lee Tat (No 2) on Arnold and the 2008 CA’s interpretation of Arnold in Lee Tat (No 1).

179 The CA in Lee Tat (No 2) expressly rejected a submission by the Grange Heights MC that the Arnold exception was a narrow one confined to cases involving the emergence of new or further material, including a change in the law. It held, instead, that the exception was “intended to serve the wider interests of justice”, with the “overriding consideration” being “to work justice and not injustice” (at [90]). Given the concern with justice underlying Arnold, the CA opined, “there does not seem to be any reason to confine the application of the Arnold exception to the narrow ambit espoused by the [Grange Heights] MC” (at [94]).

180 The CA proceeded to explain that Lee Tat (No 1) was not a case in which the Arnold exception was applied solely because of an “egregious error” on the part of the majority in the 2005 CA. Rather, that “egregious error” was but the starting point of the inquiry into whether there existed “special circumstances” so as to warrant the application of the Arnold exception (at [95]). In reiterating that the facts sufficed to found “special circumstances”, the CA emphasised in particular what it considered to be the disingenuous conduct of the Grange Heights MC in making submissions in the 2005 proceedings which contradicted other submissions that it had earlier made in the 1989 proceedings (at [98]–[99]). Finally, in order to avoid any doubt, the CA expressed its view that the Arnold exception was capable of applying to a decision of an apex court (at [100]).

Our views on the proper ambit of the Arnold exception

181 In summary, in the UK, the cases have adopted both the narrow and the wide perspectives of the Arnold exception, but in Singapore, it is the wide perspective that has taken hold, according to Lee Tat (No 2). In our judgment, notwithstanding the cases which have subscribed to the wide perspective, a close reading of Arnold suggests that the House of Lords did not think it was creating an open-ended and nebulous exception to strict issue estoppel. Later courts promulgating the wide perspective have seized on Lord Keith’s references to the broad notion of “justice”, such as when he said that one of the purposes of issue estoppel was to “work justice between the parties” (at 109B) or where he asked rhetorically whether “considerations of justice” required that a party should be shut out from re-litigating an issue in respect of which a previous court had made a “very egregious mistake” (at 109F). But, as we cautioned earlier in relation to Koh Tony, the apparent generality of Lord Keith’s words must be read in context.

182 In carving out the exception to issue estoppel in Arnold, Lord Keith proceeded tentatively and carefully: he noted that there appeared to be “no decided case where issue estoppel has been held not to apply by reason that in the later proceedings a party has brought forward further relevant material which he could not by reasonable diligence have adduced in the earlier [proceedings]” (at 108E). It was only because there was “an impressive array of dicta of high authority in favour of the possibility of this” (at 108E–108F) that Lord Keith felt able to say that there “may” be an exception to issue estoppel in that special circumstance (at 109A–109B) – and even then, his use of the word “may” indicates that he perceived the availability of further material which could not reasonably have been adduced in the earlier proceedings as a necessary but insufficient condition for the operation of the exception. Thereafter, Lord Keith said, the court had to consider whether the further material was “confined to matters of fact” or could encompass “a change in the law” (at 109C), and the fact that he thought the court had to consider that point demonstrates that the exception to issue estoppel in his mind was entirely limited to cases where there was “further material relevant to the correct determination of a point involved in the earlier proceedings” (at 109A). Lord Keith’s subsequent comments concerning “justice” have to be read in this light.

183 In our judgment, and with respect, the CA in Lee Tat (No 2) was not correct in taking the view that “Lord Keith’s primary consideration in formulating the Arnold exception … was that the exception was intended to serve the wider interests of justice, with the ‘overriding consideration in mind’ … being ‘to work justice and not injustice’” (at [90]). We think, on the contrary, that Lord Keith had in mind a much more circumscribed exception, with the emergence of further material that could not reasonably have been adduced in the earlier proceedings being the threshold condition which has to be met before the inquiry into the justice of the case is even engaged. Notably, in Arnold, Lord Keith turned to assess the injustice of preventing the tenants from re-litigating the issue at hand only after he was satisfied that this threshold condition had been met.

184 Of course, whatever Arnold might have decided, it is within the prerogative of this court to lay down a broader and more open-ended exception to issue estoppel that is based solely on considerations of justice. But, in our view, there are good reasons not to do so. In the civil sphere, the public interest in the finality of litigation is an important and compelling one that must be safeguarded closely. Any application to reopen an issue which has previously been decided is likely to take up substantial time and resources as it obliges the court to peruse the relevant records all over again. A wide formulation of the Arnold exception runs the real risk of encouraging disgruntled litigants to try their luck repeatedly, a phenomenon that we have already seen manifested in applications made to this court to reopen or reconsider decisions already made. In a real sense, the CA runs the risk of becoming a court of endless further appeals between the same parties ventilating the same issues but having recourse to arguments not made in the earlier proceedings, and this was never in the contemplation of the statute which created this court. It is not enough to say that the near-absolute bar created by strict cause of action estoppel should stymie most applications to reopen decisions; it often does not take too much creativity to advance such an application in the guise of fresh proceedings pertaining to a different subject matter such that the relevant res judicata doctrine is issue estoppel rather than cause of action estoppel. Nor is the argument for adopting the wide perspective of the Arnold exception advanced by placing emphasis on the need “to work justice and not injustice”. Delivering justice is, of course, the mission of every self-respecting court; but the incantation of the need “to work justice and not injustice”, emotive though these words may be, can significantly undermine the interest of finality if what is meant by them is that an attempt to reopen an erroneous decision can be pursued indefinitely. At one level, any decision that is wrong may be said, at least potentially, to work injustice. But, does that mean that any erroneous decision may be reopened and reconsidered regardless of whether all available avenues of appeal have been exhausted? Moreover, a dissatisfied litigant will often be prone to harbour a lingering sense that it has been dealt an injustice. But, does that mean that such a litigant should be allowed to repeatedly seek reconsideration of the decision against it in its pursuit of what it perceives to be the “just” outcome? And if, one day, that litigant succeeds in having the decision against it overturned because a differently constituted court takes a different view of the law, should the unsuccessful party in those proceedings – viz, the erstwhile victor in the earlier proceedings – be afforded the same opportunities to expunge the bitter taste of injustice that it will now no doubt experience?

185 This segues into our next point, which is that, in the final analysis, it is not merely in the court’s interest to relieve the strain on its limited resources; it is also in the public interest, as more resources can then be devoted to litigants with bona fide and non-vexatious claims. Moreover, the parties to a litigation themselves have an interest in the finality of the proceedings between them as that is what enables them to put the litigation behind them as a chapter concluded and carry on with their lives. This is so even for the litigant who is unhappy with a decision and seeks to reopen it; in this regard, we draw attention to the comments made by Lee Seiu Kin J, albeit in a different context (namely, in the context of an application to block a vexatious litigant from instituting proceedings without the court’s leave), where he said (see Attorney-General v Mah Kiat Seng [2013] 4 SLR 788 at [24]) that it was “detrimental, objectively speaking”, for a litigant to apply over and over again to reopen a decision because “maintaining the proceedings similarly diverts [the litigant’s] own time and resources from more productive endeavours”.

186 In our judgment, an approach allowing exceptions to issue estoppel only in rigidly-demarcated categories of cases would better achieve the objective of minimising the wastage of the court’s time and resources, which is in the interests of the court, the public and the parties themselves, while at the same time allowing some room for intervention in truly exceptional cases. It is not enough to say that the vast majority of applications to reopen previously-decided matters would founder even on a broad and open-ended approach to the Arnold exception. This is because such applications place a strain on judicial resources so long as they remain pending before the courts; what is needed is to discourage the making of such applications in the first place.

187 We are unable to see an alternative course to take. It may be urged that we can define a length of time beyond which the parties would be precluded from challenging, in subsequent proceedings, a matter that has previously been decided between them or a maximum number of times that a party can challenge such a matter, but it is impossible to set out such limits in advance in a non-arbitrary way. It may also be argued that a wide approach to the Arnold exception would be adequately balanced by the High Court’s power under s 74 of the Supreme Court of Judicature Act (Cap 322, 2007 Rev Ed) to order that vexatious litigants not be permitted to institute legal proceedings without the court’s leave, but that requires an application by the Attorney-General to be brought. In our judgment, the CA, the highest court in this land, ought to guard the judicial process – which includes the efficacy of that process – and one way in which it can do so is to place strict limits on the Arnold exception to issue estoppel.

188 We turn now to the question of what the limits of the Arnold exception should be. In our judgment, it is not sufficient to focus on the availability of new or further material that could not with reasonable diligence have been adduced in the earlier proceedings. The real difficulty in Arnold was that there was a patently wrong decision by Walton J which threatened to continue to guide future decisions affecting the parties concerned – in that case, future decisions as to the rent payable upon future rent reviews. In effect, the future rights of the parties were going to be determined based on a past pronouncement of the law that was wrong. That is what marks Arnold as exceptional. It is antithetical to the rule of law that the future rights of the parties (for instance, under a contract), which are meant to be worked out against the backdrop or in the shadow of the law, should take place against a backdrop or shadow that everybody by that stage knows to be wrong. In this sense, the Arnold exception is forward-looking – it does not look into the past to undo rights that have accrued pursuant to the erroneous decision, just as in Arnold itself, the tenants did not seek to claw back the excess rent they had paid pursuant to Walton J’s erroneous construction of the rent review clause.

189 As we have already mentioned, Lord Keith’s formulation of the Arnold exception premises the exception on the availability of further material that could not reasonably have been adduced in the earlier proceedings. In our judgment, the substance of Lord Keith’s “further material” requirement lies not so much in the availability of further material in itself, as in the availability of some point that was not taken or argued before the previous court and that could not reasonably have been taken or argued at that point in time. In Arnold, the point not previously taken was the subsequent English Court of Appeal decisions which were contrary to Walton J’s construction of the rent review clause, and it was reasonable for the tenants not to have cited those decisions in the proceedings before Walton J because those decisions had not come into existence yet at that time. This test also takes care of Lord Reid’s practical concern alluded to at [164] above (namely, the concern that a party may choose in the earlier proceedings not to contest a particular issue in view of the cost and expense of doing so, given that it is not certain whether there will be any later proceedings involving that same issue), in that it will be open to a party to show that his failure to contest a particular issue at an earlier point in time was a reasonable one.

190 In the circumstances, and in summary, we respectfully depart from this court’s view in Lee Tat (No 2) that the Arnold exception is a wide and nebulous one with loosely-defined boundaries based on the wide interests of justice. Instead, we hold that the following cumulative requirements must be met before the Arnold exception can apply to enable a litigant to avoid being estopped from reopening an issue that was previously the subject of a decision:

(a) the decision said to give rise to issue estoppel must directly affect the future determination of the rights of the litigants;

(b) the decision must be shown to be clearly wrong;

(c) the error in the decision must be shown to have stemmed from the fact that some point of fact or law relevant to the decision was not taken or argued before the court which made that decision and could not reasonably have been taken or argued on that occasion;

(d) there can be no attempt to claw back rights that have accrued pursuant to the erroneous decision or to otherwise undo the effects of that decision; and

(e) it must be shown that great injustice would result if the litigant in question were estopped from putting forward the particular point which is said to be the subject of issue estoppel – in this regard, if the litigant failed to take advantage of an avenue of appeal that was available to him, it will usually not be possible for him to show that the requisite injustice nevertheless exists.

191 It follows that, in our judgment, Lee Tat (No 1) was, with respect, wrongly decided. It was held by the 2008 CA in that case that, since the majority in the 2005 CA had not decided the Main Issue of whether Grange Heights residents and visitors had a right of way over the Servient Tenement on the merits, the 2005 Lee Tat CA Decision could not give rise to issue estoppel on that issue (at [82]). But, in our view, that sidesteps impermissibly the true issue estoppel created by the decision of the majority in the 2005 CA: the majority held that Lee Tat was estopped by the earlier decisions of Coomaraswamy J and the 1992 CA from arguing that Grange Heights residents and visitors did not enjoy a right of way over the Servient Tenement. Even if the majority in the 2005 CA might have erred in so ruling, that error did not affect the position that Lee Tat remained estopped from arguing in later proceedings that Grange Heights residents and visitors did not enjoy such a right of way. Moreover, in the proceedings in Lee Tat (No 1), there was no point relevant to the question of issue estoppel that had not been taken or argued before the 2005 CA, and hence, there was no room for the application of the Arnold exception.

192 This course which we are taking in departing from Lee Tat (No 1) and Lee Tat (No 2) is not an inconceivable one. Slightly over two decades ago, this court took the view that it “should not hold itself bound by any previous decisions of its own” where adherence to any particular prior decision “would cause injustice in a particular case or constrain the development of the law in conformity with the circumstances of Singapore”, although it also emphasised that this power to depart from prior decisions would be “exercised sparingly”: see the Practice Statement (Judicial Precedent) issued by the CA on 11 July 1994 (reported at [1994] 2 SLR 689). This statement was made in the context of the abolition of appeals from local decisions to the UK Privy Council, and so was directed primarily at decisions made pursuant to or in the shadow of English law; but, in our view, it reflects a more general acknowledgment that our apex court may on occasion have to change course from a turn that it previously made. This is one such occasion, and while it may seem ironic that we are overruling a past decision of this court in the name of finality, taking the long-term view, we think that it would minimise unnecessary strain on the court’s resources and give litigants greater assurance that they can – indeed must – order their affairs on the strength of judicial decisions relating to the matters which they are disputing.

193 In short, and to reiterate, we hold that the Arnold exception is a narrow one premised on the conditions which we have summarised earlier at [190] above. If, in that situation, the application of issue estoppel would cause great injustice, the court may lift the estoppel going forward and in so far as it does not reopen or otherwise affect rights that have accrued pursuant to the earlier decision which gives rise to the estoppel. There is no further exception to strict issue estoppel beyond this.

Issue (d): Jurisdiction and power

194 We have held at [139] above that nTan is estopped from contending that the CA had no power to make the VAF Orders. However, as full arguments were advanced on whether the CA did have the power to make these orders, we venture to say that nTan’s case is not without merit. The essence of nTan’s contentions on this point is this: if a creditor of a company is expressly excluded from a scheme of arrangement such that its debt is not subject to the terms of the scheme, the court has no power to make that creditor’s debt subject to the scheme without putting the scheme to a re-vote, as that would amount to a rewriting of the substance of the scheme by the court. The court cannot rewrite the scheme in this manner because it would transform the scheme into something other than that which the requisite majority of the company’s creditors approved. Instead, the company must seek its creditors’ fresh approval for the scheme, and if it is sought to bring the hitherto-excluded creditor within the terms of the scheme, that creditor must be able to vote on the scheme.

195 On the facts of this case, the CA, having found that there had been a breach of the duty to disclose material information to the Scheme Creditors and the court before sanction of the Scheme was granted, could have set aside the Scheme and left it to the Scheme Creditors to approve a new scheme under which nTan would no longer be an “Excluded Creditor”, such that the VAF would be subject to the terms of this new scheme. There is no question that the CA had the power to do that. But, argues nTan, the CA had no power to order unilaterally that the VAF be taxed if the SM/nTan, the MC and the Company were unable to agree on an appropriate sum. This would be to alter the Scheme such that it would no longer be the scheme which the Scheme Creditors had approved. As we have said, there is some force in this argument, but, in our judgment, the fact that the CA may have been wrong in deciding that it had the power to make the VAF Orders does not afford us a basis for reopening the VAF Decision, of which the VAF Orders form a part. In the final analysis, the public interest in the finality of litigation must prevail.

Conclusion

196 We summarise our decision as follows:

(a) There is no merit in nTan’s contention that the VAF Decision should be set aside on the basis that the VAF-related issues which the CA decided therein were res judicata. This is because, even if those issues were indeed res judicata and the CA erred in failing to take that view, an error of that nature offers no basis to set aside the VAF Decision.

(b) The VAF Decision was not reached in breach of natural justice because we are satisfied, having regard to the correspondence between the Registry and the parties leading up to the VAF Decision, that nTan was fully aware of the case against it and was given a fair opportunity to be heard on the matters of fact and law that were essential to the CA’s conclusions in the VAF Decision.

(c) The CA did not lack authority to hear and determine the dispute which was the subject matter of the VAF Decision because that dispute is properly to be characterised as one relating to the continuing validity of the CA’s prior sanction of the Scheme in view of the non-disclosure of the VAF before sanction was granted, meaning that nTan’s jurisdiction argument fails.

(d) Even though there is some force in nTan’s argument that the CA lacked the power to make the VAF Orders, given that the CA, in making those orders, expressly decided that it did have such power, nTan is estopped by reason of cause of action estoppel from arguing again now that the CA lacked that power.

197 In the result, we dismiss nTan’s application in SUM 5682/2012 and SUM 6520/2012 to set aside the VAF Decision. We shall hear the parties on the question of costs and any consequential orders which may have to be made.

Andrew Phang Boon Leong JA (concurring):

Introduction

198 I have had the opportunity to read in draft the judgment delivered by the learned Chief Justice on behalf of this court. I agree with the judgment of the Chief Justice; but, as I was on the panel of this court in Koh Zhan Quan Tony v Public Prosecutor and another motion [2006] 2 SLR(R) 830, Lee Tat Development Pte Ltd v Management Corporation Strata Title Plan No 301 [2009] 1 SLR(R) 875 and Management Corporation Strata Title Plan No 301 v Lee Tat Development Pte Ltd [2011] 1 SLR 998, and as those decisions have been commented on extensively by the Chief Justice and, in some respects, have been overruled by the present judgment of this court, I think it is apposite if I add some further (albeit brief) observations with respect to each of the aforementioned decisions. In doing so, I shall, unless otherwise stated, use the same abbreviations as those used in the Chief Justice’s judgment.

The decision in Koh Tony

199 Turning, first, to Koh Tony, I delivered the judgment of this court in that case. I therefore feel able to state without reservation that I entirely agree with the learned Chief Justice’s views (above at [127]–[129]) generally, and specifically, his observation that Koh Tony “should be viewed in its criminal context – and, specifically, a capital one at that” (above at [129]). Indeed, although the Chief Justice did not think it was necessary to “come to a firm conclusion” as to whether “extended” cause of action and issue estoppel could be invoked in a civil context to estop a litigant from making a belated attempt to attack the jurisdiction of a court, I agree with him that, “in principle”, this possibility exists, and that it would be going too far to hold that, in a civil case, the “extended” forms of cause of action as well as issue estoppel can never apply to bar a belated jurisdictional objection (see, likewise, [129] above of the Chief Justice’s judgment).

200 I would only add the further observation that, in Koh Tony itself, in addition to that case being of a criminal nature and involving a capital offence, the jurisdictional objection was a direct one, inasmuch as it centred on the interpretation of s 44(3) of the Supreme Court of Judicature Act (Cap 322, 1999 Rev Ed) (“the SCJA”) (the current equivalent of that provision is now to be found in s 374(3) of the Criminal Procedure Code (Cap 68, 2012 Rev Ed)). This particular provision related – in no uncertain terms – to the criminal jurisdiction of this court. Indeed, it is significant to note that s 44 of the SCJA was, at the time Koh Tony was decided, to be found in Pt V of the SCJA, whose heading then read as follows: “Criminal Jurisdiction of Court of Appeal” [emphasis added]. Further, the heading for s 44 itself at that time read as follows: “Jurisdiction to hear and determine criminal appeals” [emphasis added]. Indeed, in Koh Tony, this court observed as follows (at [29]):

It is important to emphasise that the present decision is not a carte blanche for this court to review its previous decisions when it is truly functus officio. In particular, this court has neither the jurisdiction nor power to review the substantive merits of the case; indeed, this much is clear from the existing case law. In contrast, in the present proceedings, we are concerned with s 44(3) of the SCJA – that particular provision being, in both substance and form, concerned (as the heading of s 44 itself confirms) with the criminal jurisdiction of this court. For the avoidance of doubt, the decision in the present case is confined to the precise question of whether this court has the jurisdiction and power to consider if the earlier court had the jurisdiction to entertain the appeal before it [ie, the Prosecution’s appeal in CCA 2/2005]. In the circumstances, we would respond to this question in the affirmative for the reasons set out above. [emphasis in original]

201 In a similar vein, towards the end of its judgment in Koh Tony, this court observed thus (at [58]):

In summary, we hold that this court had the jurisdiction and power to inquire into whether it had the jurisdiction to entertain the [Prosecution’s] appeal in the earlier proceedings in Criminal Appeal No 2 of 2005. However, we reiterate that this holding is confined to the precise issue before us (see, in particular, [29] above) and should not be taken as giving an even limited licence, let alone a carte blanche, for applications inviting this court to review its previous decisions in situations where it is functus officio. A clear situation of this would be when the substantive merits of the case have already been reviewed and ruled upon by this court.

202 Indeed, in Koh Tony, this court also observed that it would view dimly any attempt to dress up what were in effect arguments on the substantive merits of an earlier decision in ersatz jurisdictional guise. Referring to its affirmative answer to “the precise question” [emphasis in original] posed in the passage from Koh Tony (at [29]) quoted at [200] above, this court commented (at [31]):

The only possible problem that might arise is if similar challenges to this court’s jurisdiction are mounted too liberally and (worse still) vexatiously. We do not, however, consider this to be a serious objection when viewed from a practical perspective. Firstly, we do not envisage that there would be many situations where issues relating to the jurisdiction of this court would be raised. Secondly, as we have already alluded to above, even the possibility that this court might have exceeded its jurisdiction is a serious matter – and all the more so in the context of criminal proceedings. Thirdly, and following from the previous point, if the issue of jurisdiction raised is considered to have been raised frivolously or vexatiously or when the issue has become moot, sanctions can be imposed on the lawyers concerned, including (but not limited to) the making of an appropriate order as to costs. In the interests of justice, however, we would state that the conduct of a case would have to be really beyond the pale in order to be considered to constitute frivolous or vexatious conduct.

203 In summary, Koh Tony therefore ought not to be given an expansive reading, but ought, instead, to be read in the light of the precise issue which was before this court in that particular case – which clearly involved the interpretation of a statutory provision relating to the criminal jurisdiction of this court. In the language used by the Chief Justice at [106]–[107] above of his judgment, the contention advanced by the accused persons in Koh Tony in support of their criminal motions was that the CA did not have the authority to hear the Prosecution’s appeal in CCA 2/2005 because, so the accused persons argued, the Prosecution could only appeal against the acquittal of an accused person (see [117] above). Moreover, as importantly (and to reiterate a point which I made earlier), as the learned Chief Justice has pointed out, Koh Tony was also a criminal case involving a capital offence. In such circumstances, the court is faced with a different matrix when considering how it should approach the raising at a late stage of a point which could and should have been but was not raised earlier.

204 In the civil context, a situation similar to that which obtained in Koh Tony would, in my view, be (in the very nature of things) even rarer than in the criminal context (or even virtually non-existent) – especially given that the established case law on the civil jurisdiction of the courts is already very well-established (see generally, for example, Jeffrey Pinsler, Singapore Court Practice (LexisNexis, 2014) at vol II, pp 383–497 and, by the same author, Principles of Civil Procedure (Academy Publishing, 2013) at ch 24).

The decisions in Lee Tat (No 1) and Lee Tat (No 2) viewed in the context of the Arnold exception

205 I turn now to Lee Tat (No 1) and Lee Tat (No 2) and consider them specifically in relation to the Arnold exception. It is important to observe, at the outset, that Arnold was not in fact cited by any of the parties to the 2008 CA in Lee Tat (No 1). Indeed, as a result of this, the Grange Heights MC initiated further proceedings (as summarised at [178] above of the learned Chief Justice’s judgment), culminating in the appeal in Lee Tat (No 2).

206 The fact that Arnold was considered as well as elaborated upon by the 2008 CA in Lee Tat (No 1) is significant in the context of the present application by nTan to set aside the VAF Decision because the 2008 CA did not, in Lee Tat (No 1), have the benefit of submissions from counsel which raised the very pertinent issues concerning the scope of the Arnold exception that we have had in the present setting-aside application – particularly in relation to the narrow and the wide perspectives of that particular exception.

207 I should state at the outset that I endorse the view that the 2008 CA, in Lee Tat (No 1), did not act in breach of natural justice when it considered, analysed and applied the Arnold exception without input from the parties themselves; I also endorse the reasons given by this court in Lee Tat (No 2) for arriving at that conclusion.

208 That having been said, the question remains as to whether the scope of the Arnold exception was in fact considered in Lee Tat (No 2). In my judgment, it is significant to first note that Lee Tat (No 2), as alluded to above, was concerned primarily with the issue of whether or not the 2008 CA in Lee Tat (No 1) had acted in breach of natural justice in considering, analysing and applying the Arnold exception without input from the parties. For this purpose, in Lee Tat (No 2), it was not necessary for this court to undertake a detailed consideration of the scope of that exception. I have deliberately emphasised the word “detailed” in the preceding sentence. Let me elaborate.

209 It cannot, in fairness, be said that in Lee Tat (No 2), this court did not consider the scope of the Arnold exception at all. This is clear from the judgment itself. However, it equally cannot, in my respectful view, be said that in Lee Tat (No 2), this court considered the Arnold exception in as sufficiently nuanced a manner as has been done in the present setting-aside application by nTan. This is not surprising because a close reading of the judgment in Lee Tat (No 2) itself will reveal that the CA in that appeal was concerned with:

(a) first, confirming, as a matter of principle, that it (the CA) had the inherent jurisdiction: (i) to reopen and rehear an issue which it had earlier decided if it were shown that the earlier decision had been arrived at in breach of natural justice; and (ii) in such circumstances, to possibly set aside the whole or part of its earlier decision on that particular issue; and

(b) secondly, determining whether (on the premise that the CA had the necessary jurisdiction to reopen and rehear an issue which it had earlier decided) it ought to grant the declarations which the Grange Heights MC sought in OS 875/2009 (these declarations are set out at [27] of Lee Tat (No 2), and are also summarised at [178] above of the Chief Justice’s judgment).

210 In so far as the second point was concerned, the CA in Lee Tat (No 2) held (at [67]) that it was “pointless” to set aside the 2008 CA’s decision in Lee Tat (No 1), and therefore, it would “serve no useful purpose” to grant the declarations sought by the Grange Heights MC in OS 875/2009. In particular, the court observed as follows (likewise at [67] of Lee Tat No (2)):

In our view, it is pointless to set aside the 2008 CA’s decision [ie, Lee Tat (No 1)] and grant the MC [ie, the Grange Heights MC as defined at [167] above of the Chief Justice’s judgment] a hearing on the Arnold exception (viz, the issue which the MC claims it was not heard on in the 2008 appeal) because the rulings of the 2008 CA as set out at sub-paras (a) and (c) of [66] above entail that even if the ruling at sub-para (b) thereof were discounted, the outcome of the 2008 appeal would still be the same. In this regard, it bears reiteration that the MC has conceded before us that the ruling at sub-para (a) of the preceding paragraph [ie, [66] of Lee Tat (No 2)] is correct; ie, the MC now accepts that the [majority in the 2005 CA] erred in ruling in the [2005 Lee Tat CA Decision] that the Harris v Flower issue [of whether Grange Heights residents and visitors were entitled as against Lee Tat, the owner of the Servient Tenement, to use the right of way over the Servient Tenement] was res judicata (see [44] above). We also note that the MC has not argued that the Arnold exception is applicable to the rulings set out at sub-paras (a) and (c) of the preceding paragraph. It follows that these two rulings are res judicata as between the MC and Lee Tat. Indeed, we would go further to state that if Lee Tat had been permitted to raise the Harris v Flower issue in the [appeal before the 2005 CA] (which, as just mentioned, the MC now accepts Lee Tat should have been allowed to do) and if the [majority in the 2005 CA] had ruled on the merits of that issue, they (and also Chao JA) would undoubtedly have decided (as the 2008 CA did in [Lee Tat (No 1)]) that the Harris v Flower principle … prevented [Grange Heights residents and visitors] from using the [r]ight of [w]ay [over the Servient Tenement] for access between Grange Road and Lot 561. Accordingly, the Judge’s decision to dismiss [OS 875/2009] as an exercise in futility is essentially correct. The power of the court to grant a declaration is discretionary, and where the court feels that a declaration will serve no useful purpose, no declaration will be granted (see Latham Scott v Credit Suisse First Boston [2000] 2 SLR(R) 30 at [74]).

211 The CA thus dismissed the appeal in Lee Tat (No 2) on the basis that even if arguments on Arnold had been raised before the 2008 CA, the latter would have made the same decision, and hence, it would “serve no useful purpose” to grant the Grange Heights MC the declarations which it sought in OS 875/2009. In keeping with this, the CA in Lee Tat (No 2) then immediately proceeded to observe (at [68]) that “[w]hat we have stated above is sufficient to dispose of this appeal” [emphasis added]. Whilst it then went on to make further observations on the Grange Heights MC’s substantive arguments to the effect that it (the Grange Heights MC) had not been heard on the Arnold exception in the proceedings before the 2008 CA, these observations were, strictly speaking, obiter dicta. More importantly, whilst counsel for the Grange Heights MC in the proceedings in Lee Tat (No 2) did argue (at [34]) that:

… [T]here were strong policy reasons for not recognising an “egregious error” exception to res judicata in Singapore as such an exception would undermine certainty and confidence in the legal system: nothing could be more undesirable than to permit a litigant to re-litigate a decision made against it by the CA, which decision had every appearance of finality, in the hope that a differently constituted bench might be persuaded to take a view which its predecessor had rejected …

the above (general) argument by the Grange Heights MC’s counsel (which this court has, in principle, accepted in the present setting-aside application by nTan) was not underpinned by the (specific) case law reflecting, respectively, the narrow and the wide perspectives of the Arnold exception, which, in my view, are pivotal to any consideration of the scope of this exception. It is true that the Grange Heights MC’s counsel in Lee Tat (No 2) did (at [35]) characterise the Arnold exception as constituting:

… a narrow exception that was applicable only when there was “new material, whether new factual evidence or a subsequent change in the law, such as to amount to ‘special circumstances’ to avoid the usual doctrine of finality in litigation” … in the larger interest of justice. [emphasis added in bold italics]

However, it is important to reiterate that there appears to be no indication that this particular argument by counsel as to the “narrowness” of the Arnold exception was in fact premised on specific case law and the corresponding specific arguments which have been referred to in the Chief Justice’s judgment in the present case.

212 In fairness, I should mention that the case law which deals with the narrow as well as the wide perspectives of the Arnold exception (as analysed above at [152]-[166] of the Chief Justice’s judgment) was – with the exception of two cases (namely, Curling v Securicor and Ahsan) – developed after the decision in Lee Tat (No 2) itself as well as after the decision in Lee Tat (No 1) (it should also be noted that neither Curling v Securicor nor Ahsan was, in any event, referred to in Lee Tat (No 2) or Lee Tat (No 1)). However, there was also nothing to prevent submissions by counsel as to how Lord Keith’s words in Arnold should, as a matter of general principle (ie, without reference to the relevant case law), be interpreted. In this last-mentioned regard, as (if not more) importantly, there is no indication that arguments relating to the need to read Lord Keith’s words in Arnold in their particular context were made and considered in Lee Tat (No 2) itself, and even if such arguments were made, that they were understood in the manner which this court has done in the present case (as to which, see the Chief Justice’s judgment above at [141]-[151] and [181]-[193]). Indeed, that this appears to be the case is buttressed by the fact that the CA itself, in responding in Lee Tat (No 2) to the Grange Heights MC’s submissions which I have briefly alluded to above, adopted (at [95]-[99]) what was in essence the same analysis as that which the 2008 CA had earlier adopted in Lee Tat (No 1), which analysis, it will be recalled, was arrived at without the benefit of any input from the parties - with the possible exception that the CA in Lee Tat (No 2) noted that counsel for the Grange Heights MC had by then conceded that the majority in the 2005 CA had, in the 2005 Lee Tat CA Decision, been wrong in law. With respect, this last-mentioned point (ie, the concession by the Grange Heights MC’s counsel in Lee Tat (No 2) that the decision of the majority in the 2005 CA was wrong in law) was, at best, a neutral one since an error of law on the merits is not generally a ground for reopening a decision simply because the decision concerned is, ex hypothesi, res judicata (see, in a similar vein, [71] above of the Chief Justice’s judgment).

213 To summarise, it appears that this court was not, in Lee Tat (No 2), fully apprised of the specific arguments relating to how Lord Keith’s words in Arnold should be interpreted – ie, whether a narrow or a wide perspective of the learned law lord’s words ought to be adopted. Assuming that these specific arguments were indeed not (or not fully) available to the CA in Lee Tat (No 2), this court can, in the present setting-aside application by nTan, consider these very same arguments. Indeed, it would, in my view, be remiss if this court does not consider these arguments – given the broad policy implications (or, rather, consequences) which the learned Chief Justice has outlined above. In this regard, it should also be noted that in the present case, this court specifically invited submissions from the Company, the SM/nTan and the MC on the issue of how Lord Keith’s words in Arnold should be interpreted.

Concluding remarks

214 I would conclude with a few observations as to why the concept of an “egregious error” mentioned in Arnold and referred to in Lee Tat (No 1) as well as Lee Tat (No 2) ought, as has been observed by the learned Chief Justice at [181]–[187] above of his judgment, to be interpreted in a narrow manner (according to the principles summarised by the Chief Justice above at [190]). First, I agree with the reasons set out by the Chief Justice in his judgment for adopting a narrow approach in this regard. More specifically, it is clear that the House of Lords in Arnold did not refer to the concept of an “egregious error” as a stand-alone legal principle in and of itself. And even if the House is nevertheless construed as making such a reference (which would in fact constitute the wide perspective of the Arnold exception referred to above), this would, in my respectful view, be a very dangerous and undesirable approach for the Singapore courts to adopt, simply because of the vagueness of the concept of an “egregious error”. Indeed, an “egregious error” which is not jurisdictional in nature would, ex hypothesi, be an error of law on the merits of a case. As I have mentioned above (at [204]), an error which is jurisdictional is likely to be rare – or even virtually non-existent – in the civil context. To permit the reopening of a decision on the basis of an “egregious error” of law relating to the merits of a case would be to undermine the very raison d’etre of the doctrine of res judicata itself. As the learned Chief Justice has already pointed out (at [184] above of his judgment):

… A wide formulation of the Arnold exception runs the real risk of encouraging disgruntled litigants to try their luck repeatedly, a phenomenon that we have already seen manifested in applications made to this court to reopen or reconsider decisions already made. In a real sense, the CA runs the risk of becoming a court of endless further appeals between the same parties ventilating the same issues but having recourse to arguments not made in the earlier proceedings, and this was never in the contemplation of the statute which created this court. … [emphasis added in bold italics]

215 In the same paragraph of his judgment (ie, [184] above), the Chief Justice added (and I agree):

… It is not enough to say that the near-absolute bar created by strict cause of action estoppel should stymie most applications to reopen decisions; it often does not take too much creativity to advance such an application in the guise of fresh proceedings pertaining to a different subject matter such that the relevant res judicata doctrine is issue estoppel rather than cause of action estoppel. …

Indeed, it must not be forgotten that the need for finality in judicial decisions is itself a no less important aspect of the overall concept of “justice”. As the learned Chief Justice aptly observed in his judgment (above at [185]):

… [I]t is not merely in the court’s interest to relieve the strain on its limited resources; it is also in the public interest, as more resources can then be devoted to litigants with bona fide and non-vexatious claims. Moreover, the parties to a litigation themselves have an interest in the finality of the proceedings between them as that is what enables them to put the litigation behind them as a chapter concluded and carry on with their lives. This is so even for the litigant who is unhappy with a decision and seeks to reopen it; in this regard, we draw attention to the comments made by Lee Seiu Kin J, albeit in a different context (namely, in the context of an application to block a vexatious litigant from instituting proceedings without the court’s leave), where he said (see Attorney-General v Mah Kiat Seng [2013] 4 SLR 788 at [24]) that it was “detrimental, objectively speaking”, for a litigant to apply over and over again to reopen a decision because “maintaining the proceedings similarly diverts [the litigant’s] own time and resources from more productive endeavours”.

216 What, then, about the argument from injustice if a narrow view of the Arnold exception is applied in cases where a particular rule or principle of law which is wrong is embodied in a decision of this court? The fact of the matter is that – as this very decision demonstrates – this court is empowered, pursuant to the Practice Statement which it issued on 11 July 1994 (see Practice Statement (Judicial Precedent) [1994] 2 SLR 689), to overrule its previous decisions. At this juncture, I would like to observe that the argument from error (and consequent injustice) ought not to be overstated. For example, the House of Lords (now the UK Supreme Court) has had occasion to overrule only a small handful of its decisions since its own Practice Statement on departing from judicial precedents (viz, Practice Statement (Judicial Precedent) [1966] 1 WLR 1234) was issued almost five decades ago (reference may also be made to Louis Blom-Cooper, “1966 and All That: The Story of the Practice Statement” (and the literature as well as the cases cited therein) in ch 9 of The Judicial House of Lords: 1876-2009 (Louis Blom-Cooper, Brice Dickson & Gavin Drewry eds) (Oxford University Press, 2009)). Indeed, this very fact buttresses the point made in the present case to the effect that the House of Lords in Arnold could not possibly have been intending to adopt a wide perspective of the exception articulated by Lord Keith. Returning to the argument from injustice, it is, of course, also open to our Parliament to overrule a decision of this court should it see a need to do so. But, such overruling as alluded to in the present paragraph would be prospective in effect, in the sense that it would not undo the outcome of a final decision already made. This is ultimately the effect of the doctrine of res judicata, which places an emphasis on finality in litigation; and in the final analysis, the effect of our decision today is to reiterate the importance of that doctrine and to affirm that it will apply, save in the exceptional circumstances outlined by the Chief Justice at [190] above of his judgment.

The Royal Bank of Scotland NV (formerly known as ABN Amro Bank NV) and others v TT International Ltd (nTan Corporate Advisory Pte Ltd and others, other parties) and another appeal [2015] SGCA 50

The Royal Bank of Scotland NV (formerly known as ABN Amro Bank NV) and others v TT International Ltd (nTan Corporate Advisory Pte Ltd and others, other parties) and another appeal
[2015] SGCA 50