Soon Kok Tiang and others v DBS Bank Ltd and another matter

Lee Seiu Kin J:

[1] The year 2008 saw the world economy mired in what has been described as the worst financial crisis since the great depression of the 1930s and its aftershocks are still being felt today. Singaporean banks and financial institutions have been more fortunate than those in many other countries, but not all Singaporeans were untouched by the chaos that raged through the financial markets. Fears over the solvency of large and once-reputable banks, declining credit availability and damaged investor confidence caused the failure or acquisition of several major banking and finance institutions; Lehman Brothers Holdings Inc (“Lehman”) was one of these. On 15 September 2008, Lehman filed for Chapter 11 bankruptcy protection in the US. The filing marked the largest bankruptcy in US history; in Singapore, it caused many individuals who had invested in Lehman-linked structured investment instruments to lose their money.

[3] The HN5 were launched on 30 March 2007 and were intended to last a period of 5.5 years until 2012. Initially, the offering was open only to existing customers of the defendant on an “invitation only” basis. On 2 April 2007, the offering was made open to the public.

[4] At the time of the offering, information concerning the HN5 was available from two documents. The first was the base prospectus (“Base Prospectus”) – a term I shall use collectively to refer to the original prospectus dated 22 December 2005, the supplementary base prospectus dated 5 April 2006 which amended the original prospectus and a final version registered with the Monetary Authority of Singapore (“MAS”) on 27 December 2007. This Base Prospectus applied generally to govern the entire series of DBS High Note Programmes – from series 1 to 5. The HN5 was, as its name suggests, the fifth in the series. The second document was the pricing statement dated 29 March 2007 (“Pricing Statement”), which contained the specific terms and conditions relating to the HN5. During the launch, interested individuals were furnished with copies of the Base Prospectus, Pricing Statement, and an application form (“Application Form”). The Application Form was to be signed (in two places) and returned prior to 4.30pm on 30 April 2007 – the closing date and time of the HN5 offering.

[5] A total of 1,127 persons invested in the HN5, and these were issued to them by the defendant on 16 May 2007. Between 16 August 2007 and 18 August 2008, investors (including the plaintiffs) received five quarterly payments of accrued interest in respect of their investment in the HN5. Depending on the amount of their respective investment, each of the plaintiffs received payments that ranged between S$1,568.50 and S$25,096.00, and US$2,894.31 and US$12,404.22.

[6] A description of the HN5 was provided in the Pricing Statement issued by the defendant:

Buyers of the HN5 were promised high returns on their investment – a quarterly interest rate of either 5.00% or 6.50% per annum, depending on whether the investor had subscribed to the Singapore Dollar (“SGD”) or United States Dollar (“USD”) Tranche – until the specified maturity date in 2012. On the maturity date, they would also receive 100% of the principal amount they had invested, unless prior to that date, either:

[7] In order to explain what a credit event and constellation event consisted of, it is necessary to go into the details of the structure underlying the HN5. The funds raised from the sale of the HN5 were used to purchase another set of structured notes (“Reference Notes”) issued by Constellation Investment Limited (“Constellation”). The defendant was the holder of the Reference Notes. Constellation was a special purpose trust company established by the defendant in 2003 and incorporated under the laws of the Cayman Islands. Its primary objective was the issuing of various credit-linked and other structured notes to both retail and institutional investors. The defendant had previously used Constellation as a vehicle to issue 70 different structured notes to retail investors in Hong Kong between 2003 and 2007. The HN5 was part of a similar series issued in Singapore.

[8] Constellation used the funds it raised from the issue of the Reference Notes to invest in structured securities comprising of collateral debt obligations (“CDOs”) issued by a Cayman Islands-incorporated company, Zenesis SPC. These secured its obligations under the Reference Notes and generated monetary returns, which were used to pay the interest due to the investors in the HN5. As a result, the performance of the HN5 (ie their interest yield) was directly linked to the performance of the Reference Notes. In return, investors were exposed to the risk of Constellation’s own bankruptcy and the risk of any of the securities purchased by Constellation losing all their value. Both of these risks – were they to materialise – would amount to a Constellation Event. Such an event would trigger early redemption of the Reference Notes and this would, in turn, trigger the termination of the HN5.

[9] In addition to the above, however, the Reference Notes were also notionally linked to the credit performance of eight reference entities. The nature of this link was a “first-to-default” basis: ie, in the event that one of these reference entities failed to honour a specific financial obligation (“Reference Obligation”) that was identified in the Reference Notes, the Reference Notes would be redeemed, unwound and terminated. The HN5 would also suffer the same fate. It is this “default” by a reference entity that constitutes a credit event referred to in [6] above and it is an event that is entirely independent of Constellation’s own creditworthiness or even the value of the securities purchased by Constellation. In effect, buyers of the HN5 were risking their money on the following eventualities: that the defendant would remain solvent; that Constellation would remain solvent; that none of the securities that Constellation had bought with their money (the CDOs) would turn out to be worthless; and that none of the eight reference entities within the “first-to-default” basket would default in the 5.5-year life of the HN5. As it turned out, Lehman was one of the eight reference entities. At the time the Pricing Statement was issued, Lehman’s issuer credit ratings by Standard & Poor’s, Moody’s and Fitch stood at the confidence-inducing and secure-sounding levels of A+, A1 and A+ respectively – ratings that, with hindsight, can only be described as astounding in the light of events that transpired.

[10] The trouble in the global financial markets began on 3 August 2007, when, as Lehman’s own former global head of quantitative equity strategies Matthew Rothman put it, “events that models only predicted would happen once in 10,000 years happened every day for three days.” Investors suffered a crisis of confidence in financial instruments underpinned by “sub-prime” debts and mortgages; the value of these and other investment instruments crashed; and the financial markets were thrown into complete turmoil. These events are too fresh and too deeply imprinted in the memory of most people to require further adumbration. Just over a year later, on 15 September 2008, Lehman filed a petition under Chapter 11 of the US Bankruptcy Code in the US Bankruptcy Court. This constituted an act of “default” by Lehman on one of the Reference Obligations – a US$1.23b subordinated note issued by Lehman and due for redemption in 2017 (the “Lehman Note”) – under the Reference Notes. The Reference Notes were terminated by Constellation and under the HN5 structure this in turn resulted in the termination of the HN5.

[11] On 19 September 2008, the defendant wrote to the HN5 investors enclosing the relevant notices of credit event (which were themselves dated 17 September 2008). The letters also notified investors of the consequences of Lehman’s Chapter 11 petition, namely that:

(a) The HN5 had been terminated.

(b) The defendant would, as the issuer of the HN5, be redeeming the investors’ investments in the HN5 at the credit event redemption amount (“CERA”) on the credit event redemption date.

(c) The investors would be informed of the CERA as soon as the information became available.

(d) The HN5 would have ceased to bear any interest from 16 August 2008.

[12] On 28 October 2008, the defendant wrote again to the investors. The letters informed them that the CERA for the HN5 had been calculated at zero. Thus, no amount would be due and payable to them on the credit event redemption date (which had been fixed on 3 November 2008). Each letter enclosed a Notice of CERA dated 27 October 2008 that explained how the CERA was calculated.

[13] The upshot of the matter was this: over 1,000 investors in the HN5 who had sunk in some S$103m saw their investments (and for many, their life savings) rendered worthless. According to an update on the resolution of conflicts relating to Lehman-linked structured investment instruments provided by MAS on 16 January 2009, approximately 200 investors who complained to the defendant about the HN5 received offers of full or partial settlement.The remaining 1,200, it appears, received nothing. Thus, it was perhaps not surprising that on 8 July 2009 the plaintiffs filed this originating summons in an attempt to claw back some of the money they had lost.

[14] The primary issue in these proceedings pertains to the calculation of CERA – the amount that investors in the HN5 were to receive upon the occurrence of a credit event. According to the plaintiffs, the Pricing Statement they received from the defendant contained four definitions of CERA. These are:

(a) The description at p 2 of the Pricing Statement, under the heading “Description of the Notes” (the “First CERA Description”):

(b) The description at p 4 of the Pricing Statement, under the heading “Description of the Notes” (the “Second CERA Description”):

(c) The description at p 14 of the Pricing Statement, under the heading “Summary of the Offering” (the “Third CERA Description”):

(d) The description at p 61 of the Pricing Statement, under the heading “Selected Information about the Reference Notes” (the “Fourth CERA Description”):

[15] Having identified four definitions of CERA in the Pricing Statement, the plaintiffs submit that:

(a) The amount which the plaintiffs would receive in the event of an early termination of the HN5 (ie the CERA) is a material term of the contract between the parties.

(b) This amount cannot be calculated based on the four definitions of CERA set out above as each of these are inconsistent with the other.

(c) This uncertainty as to a material term of the contract underlying the HN5 thereby renders it void for uncertainty.

[16] The defendant’s rebuttal to the above is made on several levels. Firstly, the defendant argues that the allegedly inconsistent CERA descriptions do not form part of the HN5 contract at all. Secondly, even if the four descriptions do form part of the HN5 contract, there can be no uncertainty about the calculation of the CERA, as:

(a) The Third CERA Description is expressly designated in the Pricing Statement as the prevailing one.

(b) The four CERA descriptions were never meant to be identical descriptions of the CERA and each serves a different purpose.

(c) The four CERA descriptions were, in any event, accurate, consistent and reconcilable (save for an obvious clerical error in the Fourth CERA Description).

Thirdly, the defendant claims that the plaintiffs are estopped from relying on the inconsistency as ground for voiding the HN5 contract; and its fourth and final point is that voiding a contract (let alone one that is fully performed) for uncertainty is a measure of last resort and one the court should not take if there are other devices it may deploy to save the contract.

[17] Based on the parties’ submissions, the crucial issues in this case are:

(a) Whether the HN5 contract between the plaintiffs and defendant consists of the whole of, or only a part of, the Base Prospectus and Pricing Statement.

(b) Whether the four definitions of CERA provided in the HN5 contract are inconsistent with one another.

(c) Whether that inconsistency necessitates the HN5 contract being declared void for uncertainty.

[18] The first issue is the identification of the HN5 contract and its terms. It is the plaintiffs’ case that the entirety of the Base Prospectus and Pricing Statement collectively form the contractual documents of the HN5. Thus, everything found in the two documents (in particular the four alleged definitions of CERA) make up contract’s terms and conditions. In contrast, the defendant claims that only parts of the Base Prospectus and Pricing Statement make up the HN5 contract. If the defendant is right, then the outcome of its submission is that the four alleged definitions of CERA will not be found within the contract at all. The question, therefore, is what terms were the HN5 contracts made and agreed upon?

[19] In order to give an answer to that question, I must consider what constituted the relevant offer and acceptance in this case. Although the launch of the HN5 was billed as an “offering” by the defendant and the period between 30 March 2007 and 30 April 2007 termed the “Offer Period”,this does not, of itself, render the launch (and the documents handed out during that launch) an “offer” at law. Clause 7 of appendix A (terms and conditions and procedures for application, acceptance and cancellation) to the Pricing Statement (“Appendix A”) grants the defendant the “right to reject or to accept” any application without the need to give reasons. The launch of the HN5 was, therefore, an invitation to treat and it was the individual investor who, by completing, signing and submitting the Application Form, made the offer to the defendant to buy into the HN5. The contract came into being with the defendant’s acceptance of such offer. Thus, one must look to the terms in the Application Form to determine whether the Base Prospectus and Pricing Statement were incorporated in part or in their entirety as part of the terms of the HN5 contract.

[20] The Application Form contains five sections on its front and a set of terms and conditions on the reverse side. Those five sections are entitled:

Payout/Maturity Instruction;

(b) Section B: Agreement;

(c) Section C: the HN5 Risk Acceptance;

(d) Section D: Non-US Person Declaration; and

(e) Section E: Client’s Acknowledgment.

[21] The crucial section is Section B. This section is signed by each applicant and reads as follows:

[22] The two sides disagree on the meaning of the emphasised words. The defendant’s position is it incorporates only those sections of the Pricing Statement which have the words “Terms and Conditions” in its title. These are, namely, the section under the heading “Terms and Conditions of the Notes” and Appendix A, which is titled “Terms and Conditions and Procedures for Application, Acceptance and Cancellation.” The following statement is found in the section “Terms and Conditions of the Notes” in the Pricing Statement:

[23] Thus, according to the defendant, Section B of the Application Form incorporates only the following as part of the HN5 contract:

(a) The terms and conditions on the reverse of the Application Form;

(b) The following parts of the Pricing Statement, namely:

(i) the section under the heading “Terms and Conditions of the Notes” found in the Base Prospectus;

(ii) the conditions supplement that were attached to the HN5 issue; and

(c) Appendix A of the Pricing Statement, which sets out the “Terms and Conditions of the Procedures for Application, Acceptance and Cancellation”.

[24] The plaintiffs’ position is that the phrase “Terms and Conditions set out in DBS High Notes 5 Pricing Statement in particular Appendix A” should be construed as incorporating the entire Pricing Statement into the terms and conditions of the HN5 contract. This is because there is nothing in the phrase to indicate that only those sections of the Pricing Statement with the words “Terms and Conditions” in their title are so incorporated – the phrase may just as plausibly mean that the terms and conditions of the HN5 contract are “set out” by the entire Pricing Statement.

[25] The parties’ dispute, therefore, concerns the interpretation of the phrase “Terms and Conditions set out in DBS High Notes 5 Pricing Statement”. In Zurich Insurance (Singapore) Pte Ltd v B-Gold Interior Design & Construction Pte Ltd [2008] 3 SLR(R) 1029 (“Zurich Insurance”) the Court of Appeal endorsed, at [114], the contextual approach set out by Lord Hoffmann in Investors Compensation Scheme Ltd v West Bromwich Building Society [1998] 1 WLR 896 (“Investors Compensation”) at 912-913, (see also [56] of Zurich Insurance):

[26] Pursuant to this approach, ambiguity is not a pre-requisite for admission of relevant extrinsic material showing the context of the contract to determine if the language takes on another meaning instead of the plain meaning of the words (Zurich Insurance, [25] supra at [130]):

Thus, in order to determine the meaning of the phrase “Terms and Conditions set out in DBS High Notes 5 Pricing Statement”, the court may have recourse to anything that would have formed part of the parties’ background knowledge at the time of the contract. Even if the words to be interpreted do not, on their face, appear to be ambiguous, the background context should be considered in order to determine the existence of any latent ambiguity.

[27] The only documents put in evidence constituting the background to the contract are the Base Prospectus, Pricing Statement, and Application Form. It is clear that the Base Prospectus and the Pricing Statement are documents “which would have affected the way in which the language of the [HN5 contracts] would have been understood by a reasonable man”: Investors Compensation, supra at [25]. Both documents were given to potential investors along with the Application Form and they were clearly intended to be read in conjunction with the latter; indeed, Section E of the Application Form requires an investor to make the application “pursuant to the Pricing Statement and Base Prospectus” and asks for explicit acknowledgement of the same. The information found in the Base Prospectus and Pricing Statement, therefore, clearly constitutes the background knowledge made available by the defendant to the plaintiffs at the time of formation of the HN5 contracts.

[28] While it is not difficult to say that the Base Prospectus forms part of the relevant background knowledge to the HN5 contract, I do face some difficulties deciding which version of the Base Prospectus to rely on. As mentioned at [4] above, there were three different documents issued under the label “Base Prospectus”: the original base prospectus dated 22 December 2005; the supplementary base prospectus dated 5 April 2006 amending the former; and a final version registered with MAS on 27 December 2007. The defendant relies on a paragraph from the Base Prospectus dated 22 December 2005 to demonstrate that the HN5 contract does not embody the entire Base Prospectus and Pricing Statement. This is found in the section entitled “Risk Factors” and states that:

[29] However, a check of the final, updated base prospectus of 27 December 2007 reveals a significant change in the wording of the same paragraph:

This second version of the paragraph appears to support the plaintiffs’ interpretation of the “Terms and Conditions set out in DBS High Notes 5 Pricing Statement” phrase from the Application Form, since there can be no doubt that the paragraph implies that the “full terms and conditions of [the HN5]” may be found by reading the applicable Issue Document – namely, the (entire) Pricing Statement.

[30] Unfortunately, it appears that neither party to this case noticed the inconsistency between the two versions of the base prospectus. As mentioned above, the defendant relies on the first prospectus dated 22 December 2005; the plaintiffs, in contrast, have tendered to this court the updated version of 27 December 2007. I was given no explanation for the difference, nor any account of how or when it came about. The affidavit of Mr Soon Kok Tiang filed on behalf of the plaintiffs does not mention which base prospectus the investors received at the time of the launch of the HN5; nor does it state which version they had in their possession when they submitted their Application Forms. Instead, it merely asserts that the 2005 base prospectus “as updated by the [2007 base prospectus]” sets out the terms and conditions of the HN5. In contrast, Ms Debbie Lam Thuan Meng’s (“Ms Lam”) affidavit filed on behalf of the defendant implies that it was the 2005 version of the base prospectus that was issued to investors in the period between 30 March 2007 and 30 April 2007. She did indicate, however, that this 2005 version had, even then, already been amended by virtue of the supplementary base prospectus dated 5 April 2006. The information to be found in the supplementary base prospectus was therefore, according to Ms Lam, part of the basis upon which the HN5 was launched. Nevertheless, it is not clear when the change giving rise to the inconsistency pointed out above (see [29] above) took place. In other words, it is not certain whether the 2007 version of the base prospectus merely consolidated the 2005 original base prospectus and any amendments the 2006 supplementary base prospectus introduced or whether it had, itself, introduced changes that were not to be found in the two.

[31] Despite the paucity of information concerning the history behind the different versions of the Base Prospectus, I am inclined to place greater reliance on that version tendered by the plaintiffs. The fact that the plaintiffs have (through Mr Soon) produced this particular version of the base prospectus leads one to believe that this must have been the version that they received and relied upon when applying for the HN5. They would not, after all, have had free access to the different versions of the base prospectus. What they possess, they would have obtained from the defendant, whose lawyers were responsible for drafting the various documents. I accept, therefore, that the relevant paragraph from the Base Prospectus would have told investors that the terms and conditions of the HN5 could be found in the section entitled “Terms and Conditions of the Notes” in the Base Prospectus and in “the applicable Issue Document” – namely, the entire Pricing Statement. This, therefore, formed part of the background context against which the HN5 contract was concluded.

[32] However, any reliance that I place on the paragraph from the 2007 base prospectus (see [29] above) must be weighed alongside any other conclusions that I might draw from the general context provided by the three documents. Were it not for the reasons stated at [34]-[35] below, looking at the documents as a whole I would have been slow to find that everything in the Base Prospectus and Pricing Statement amounts to a contractual term or condition. A quick glance at the table of contents of the Base Prospectusshows that the first 48 pages are devoted to, amongst other things, “References to Websites”, “Information on Websites”, a “Description of Selected Transaction Documents” and “Summary of the Programme”. Pages 48 to 119 purport to set out “Terms and Conditions of the Notes”, but the last sections of the Base Prospectus merely supply “Information Relating to [the defendant]” and “General and Statutory Information”. Clearly, therefore, not all of the Base Prospectus is given over to contractual terms; indeed, a significant portion of it is intended only to summarise, describe and explain the HN5 to investors. The Pricing Statement is no different. Its table of contents refers to information, descriptions, summaries and details on procedures and timetables. These are sections which are far more likely to be found in the preamble or explanatory note to a contract than amongst its actual terms and conditions.

[33] I also note that the Pricing Statement contains a section specifically entitled “Terms and Conditions of the Notes”. The reference in the Application Form to “Terms and Conditions set out in DBS High Notes 5 Pricing Statement” could, therefore, be easily construed referring to the section that purports to contain the terms and conditions and indeed, this is the argument that the defendant makes. It is, after all, no stretch of the English language to say that the section is one “set out” in the Pricing Statement. Thus, there is room to conclude that the phrase does not necessarily incorporate the entire Pricing Statement into the terms of the HN5 contract.

[34] Despite these considerations, I nevertheless conclude that while the words “set out” do not necessarily incorporate the whole of the Pricing Statement into the HN5 contract, when they are followed by the words “in particular”, they do have this result. The plaintiffs point out that the section entitled “Terms and Conditions of the Notes” in the Pricing Statement makes no reference at all to Appendix A. Yet, the Application Form appears to assume that it has been incorporated: the words “in particular” then single it out for the reader’s particular attention. There is no need to refer to something “in particular” unless it has been previously identified as part of a collective. In my view, this is a strong argument. The starting point for any attempt to interpret or construe a contract must be the wording of the contract itself. This does not mean that a court should only give effect to the literal meaning of a contract. Instead, it means that in order to persuade the court that the words of the HN5 contract bear another meaning from what they appear to, the defendant must show that the background context to the HN5 contract clearly and obviously indicates a different interpretation from that suggested by the plain language. As Lord Hoffmann put it in Investors Compensation (supra at [25]) when he set out the general principles by which contractual documents should be construed at 913:

This, therefore, is the threshold requirement spelt out by the Court of Appeal in Zurich Insurance (supra at [25]) at [129]. While ambiguity is no longer a pre-requisite for the court’s consideration of extrinsic material in construing a contractual document, the context behind a contract must be clear and obvious and render the language of the contract ambiguous or absurd before the court will be entitled to give the contractual term in question an interpretation that is different from that which the plain language demands: Zurich Insurance (supra at [25]), at [130].

[35] In my view, the defendant in this case has not overcome the threshold requirement. The plain language of the contract found in the Application Form supports the plaintiff’s reading of the contract: the use of the words “set out” followed by “in particular” do appear, by virtue of common syntax and linguistic use, to lead to the conclusion that the entire Pricing Statement (of which Appendix A is a part of) was incorporated into the terms and conditions of the HN5 contract. And it is not clearly and obviously indicated by the context of the contract that a different interpretation from that suggested by the plain language of the contract should be applied. Indeed, as my analysis in the preceding paragraphs above demonstrates, the context behind the HN5 contract is unclear: a glance at the documents overall may give the impression that not all of their contents constitute the terms and conditions of the HN5 contract; yet the paragraph from the section entitled “Risk Factors” found in the 2007 base prospectus (see [29] above) conveys – in express words – the impression that the entire Pricing Statement is incorporated as part of the HN5 contract terms. In fact, by describing the context as unclear, I am giving the defendant the benefit of the doubt. I am actually inclined to give great weight to the express words of the 2007 base prospectus. Those who invested in the HN5 (and this includes the plaintiffs) were not, after all, lawyers. It is unfair to expect them to know that those parts of the Base Prospectus and Pricing Statement which summarise and describe the HN5 to investors were not part of the terms and conditions of the contract – let alone that the very existence of such sections could indicate by implication that the contract did not consist of the documents in their entirety.

[36] To conclude this portion of my judgment, therefore: the phrase “the Terms and Conditions set out in DBS High Notes 5 Pricing Statement in particular [Appendix A]” should be given its natural and ordinary meaning, especially once it is read in light of the express incorporation of the “Issue Document” (or Pricing Statement) in the 2007 base prospectus. The background context in this case (limited as it is only to the three sets of documents handed to the HN5 investors with no other extrinsic facts) is equivocal at best. To my mind, however, the context really comes down in favour of the plaintiffs. It does not clearly or obviously support any other interpretation – let alone the interpretation propounded by the defendant. As such, I conclude that the HN5 contract consists of the terms and conditions set out on the reverse side of the Application Form and the entirety of the HN5 Pricing Statement, including but not limited to the section of the Pricing Statement entitled “Terms and Conditions of the Notes” and Appendix A.

[37] It follows from my conclusion on the constituent terms and conditions of the HN5 contract above that all four descriptions (or definitions) of CERA set out at [14] above fall within the terms of the contract. I turn to consider the plaintiffs’ contention that these four descriptions are inconsistent with one another.

[38] The plaintiffs’ summarised the different CERA descriptions in the following table:

[39] It should be noted at the outset that the Fourth CERA Description is grossly different from the first three descriptions. It is actually the inverse of the losses suffered by the Lehman Note and produces a curious result for an instrument whose value is supposed to depend on the value of the Lehman Note: namely, that the greater the loss suffered on the Lehman Note, the smaller the loss suffered on the HN5. Conversely, if the Lehman Note does not suffer a loss in price, the value of the HN5 plunges to zero. The defendant does not dispute that this is the result of the fourth CERA Description, but contends that this was an obvious clerical mistake that may and ought to be corrected either by construction or rectification. I deal with this latter submissions at [46] below.

[40] The plaintiff’s position in relation to the first three CERA Descriptions is as follows. It is evident from the above that all three formulae require the deduction of the CAAA and HC. Nevertheless, the three are inconsistent because:

(a) The First and Second CERA Descriptions can only be consistent if the prevailing market price of the Lehman Note is always equal to the nominal value of the Reference Notes less the loss suffered on the Lehman Note. However, this is not necessarily always the case as CERA would be zero under the First CERA Description if the market value of the Lehman Note were zero, but not necessarily zero in the Second CERA Description as the Lehman Note was only one of eight Reference Notes that the HN5 were linked to.

(b) The Third CERA Description materially differs on its face from the First and Second CERA Descriptions as the component factors it uses are completely different from the latter. It also requires the aggregate principal amount (“APA” or face value) of the Reference Note to be multiplied by the price of the Lehman Note expressed as a percentage. Yet it fails to say what the price of the Lehman Note should be expressed as a percentage of – effectively rendering the Third CERA Description unworkable.

[41] The defendant’s rebuttal of the plaintiffs’ submissions is made on several levels. Firstly, it argues that the First, Second and Third CERA Descriptions are entirely consistent as they clearly convey that the CERA is:

(a) Directly related to the market value of the defaulting reference entity’s Reference Obligation (in this case, the Lehman Note).

(b) Inversely related to the shortfall in market value of the CAAA and HC.

Secondly, the defendant argues that the three descriptions relied on by the plaintiffs were never intended to be comprehensive formulae. In the event of any uncertainty or inconsistency, the Pricing Statement provides a clear mechanism for its resolution: the Third CERA Description is expressly designated the prevailing one.

[42] I agree with the defendant’s position for the following reasons. The plaintiff’s reasoning flows from their having reduced the First and Second CERA Descriptions, both expressed in the form of words, into the precise mathematical formulae reproduced in the table at [38] above. However, a proper construction of the First CERA Description does not convey that level of precision. The First CERA Description is found on the second page of the Pricing Statement, and it is useful to see it in the context of the paragraph in which it appears (with the First CERA Description emphasised):

The words “based on” do not normally mean “equal to” and the First CERA Description, in the context of the paragraph in which it appears does not in fact convey that meaning. What it simply means is that the “prevailing market value of the defaulted Reference Entity’s Reference Obligation” will be an important factor in the computation of CERA: it does not specify the exact mathematical relationship between the two.

[43] The same reasoning applies in the case of the Second CERA Description. This is found on the fourth page of the Pricing Statement, in the context of the following paragraph (with the Second CERA Description emphasised):

Although the word “equal” is used in this case, this has to be seen in the context of the entire paragraph, particularly the repeated caveats in the two opening sentences that it is for “illustrative purposes” or “illustration” only and the words “in summary” used in the Second CERA Description itself. Furthermore, the expressions “loss suffered”, “depreciation of the market value of the collateral” and “costs and expenses” are not defined at all. It is therefore not correct to reduce it to the second formula in the table at [38] as the plaintiffs have done. As with the First CERA Description, it is no more than a general description of the manner in which CERA will be determined.

[44] The Third CERA Description appears under the section “Summary of the Offering” and it is the only one of the three descriptions that sets out an exact formula based on terms that are fully defined. Further, in the section entitled “Glossary” at p 37 of the Pricing Statement, the following appears:

[45] The foregoing leads to the ineluctable conclusion that the Third CERA Description is the operative one under the contract.

[46] I turn now to the Fourth CERA Description. As noted at [39] above, this is actually the inverse of the Third CERA Description as well as the general scheme under the First and Second CERA Descriptions. The defendant’s case is that this is an obvious clerical mistake, which can and should be corrected by construction. It submits as follows:

(a) The Fourth CERA Description contains a formula that is identical to the Third CERA Description except for one thing: instead of stating that the APA should be multiplied by the “Final Price”, it stated that the APA should be multiplied by “(1 – Final Price)”. It is plain from the other CERA descriptions in the Pricing Statement and the Application Form that the CERA should directly correlate to the market value of the defaulted reference entity’s Reference Obligation (ie the Final Price). The additional “1 –” in the Fourth CERA Description would result in a negative correlation between the CERA and the Final Price. It is also patently apparent from the other parts of the Pricing Statement that the CERA should be calculated by multiplying the APA with the Final Price and not “(1 – Final Price)”:

(i) At p 9 of the Pricing Statement, under the section entitled “Risk Factors”, the Pricing Statement highlights that an HN5 investor would be subject to the credit risk of the eight reference entities in the first-to-default credit basket and goes on to elaborate that the value of the HN5 “may be affected by the activities undertaken by the Reference Entities and any financial or economic difficulties the Reference Entities may face”. It is clear that one of the risks faced by an HN5 investor is that “any financial or economic difficulties the Reference Entities may face” would have a negative effect on the value of the HN5 (and, therefore, the CERA, if a credit event occurs); and

(ii) The illustrations provided at p 8 of the Pricing Statement under “Scenario Analyses” put the issue beyond doubt. It is explained in Scenario 3 (Worst Case Scenario) that if a credit event occurs under the Reference Notes, and if the market value of the relevant Reference Obligation is zero, the HN5 will terminate immediately and the investor will receive a zero payout. This can only be true if the CERA is calculated by multiplying the APA with Final Price (ie, the market value of the relevant Reference Obligation which is zero) and not “(1 – Final Price)”.

(b) A literal reading of the formula contained in the Fourth CERA Description would give rise to an absurdity. The CERA under the Reference Notes (and therefore the CERA for the HN5) would have an inverse correlation to the Final Price. As a result, the lower the prevailing market value of the defaulted reference entity’s Reference Obligation, the more the defendant would receive as the Reference Notes CERA and the higher the amount each HN5 investor would receive upon the occurrence of a credit event. This could not by any stretch of the imagination have been the intention of the defendant or the HN5 investors when the HN5 contract was entered into. In fact, the plaintiffs do not make any such suggestion. In investing in HN5 and in return for the higher interest yield of the HN5, the investor agreed to take on the risk, inter alia, of any financial difficulties which the eight reference entities in the first-to-default credit basket may face. If the HN5 investor’s returns increased with the depreciation of the market value of the defaulted reference entity’s Reference Obligation, that would completely turn the risk which was undertaken and accepted by the HN5 investor in the HN5 contract on its head.

[47] I agree with the defendant’s submission regarding the Fourth CERA Description. There is no doubt in my mind that it is a clerical mistake. The conditions for correcting an obvious clerical mistake by the process of construction were laid down by Brightman LJ in East v Pantiles (Plant Hire) Ltd [1982] 2 EGLR 111, 112, as follows:

This was endorsed and applied by Belinda Ang J in Ng Swee Hua v Auston International Group Ltd and another [2008] SGHC 241, at [33]-[35], and more recently by the House of Lords in Chartbrook Ltd and another v Persimmon Homes Ltd and another [2009] 1 AC 1101.

[48] I find that there is ample basis for correcting the Fourth CERA Description by the removal of the “1 –” in the formula. To the extent that the Fourth CERA Description contradicts the Third CERA Description, this would eliminate that contradiction.