Royal Courts of Justice, Rolls Building
Fetter Lane, London, EC4A 1NL
Before :
MR JUSTICE EDER
Between :
Taberna Europe CDO II plc | Claimant |
- and - | |
Selskabet AF1.September 2008 In Bankruptcy (formerly known as Roskilde Bank A/S) | Defendant |
Mr Alain Choo-Choy QC and Mr Michael d’Arcy (instructed by Duane Morris) for the Claimant
Mr Charles Béar QC and Mr Matthew Cook (instructed by Macfarlanes) for the Defendant
Hearing dates: 18-20, 24-27 November, 1, 3-4 December 2014
Judgment
Mr Justice Eder:
Introduction
In these proceedings, the claimant (“Taberna”) claims substantial damages in respect of one or more alleged misrepresentations made by or on behalf of the defendant (“Roskilde”) which, it is said, induced Taberna into entering into a secondary market purchase of certain subordinated notes originally issued by Roskilde.
At the outset, it is important to emphasise that although the claim was originally advanced on a number of grounds including deceit and negligence at common law, such grounds were abandoned in the course of the proceedings. It is now advanced solely under s2(1) of the Misrepresentation Act 1967 (the “1967 Act”). Further, in the course of the trial, Mr Choo-Choy QC on behalf of Taberna abandoned a number of the representations originally relied upon.
Taberna, an Irish public limited liability company, is an investment vehicle with beneficial ownership of a portfolio of intangible assets, including obligations or securities issued by banks and other financial institutions. Roskilde is a former Danish bank. As at the end of 2006, it was the fifth largest Danish bank in terms of market value with approximately 100,000 customers and a total business volume of approximately DKK 70 billion. For the preceding 10 years, it had been among the top five listed financial institutions in Denmark in terms of return on equity. However, in the course of 2008 it suffered substantial financial difficulties and went into bankruptcy in early 2009.
The subordinated notes in question were issued as part of Roskilde’s capital raising efforts in 2006 and 2007 which involved the issue of various categories of loan notes, including pursuant to a €2 billion Euro Medium Term Note (EMTN) programme. One such category of loan notes was the €80,000,000 Fixed/Floating Rate Dated Subordinated Callable Step Up Notes Due 2014 which were originally issued by Roskilde on 1 December 2006 to Merrill Lynch and subsequently marketed to prospective investors pursuant to the terms of a prospectus dated 4 April 2007 (the “Prospectus”). The subordinated notes which are the subject of these proceedings had a nominal face value of some €27 million and were purchased by Taberna in the secondary market from Deutsche Bank for the sum of €26,421,585 on 14 February 2008.
Within a very short time of such purchase, Roskilde encountered severe financial difficulties. In early July 2008, Roskilde sought liquidity assistance from the Danish Central Bank. On 24 August 2008, Roskilde announced inter alia that following the conduct of a structured sale process and an audit of Roskilde at the request of the Danish FSA, it was no longer able to meet the individually assessed solvency requirements of the relevant Danish legislation and a new bank had been established (“New Roskilde”) which had offered to buy all assets and assume all debts and liabilities of Roskilde, save for certain specified liabilities, including subordinated loan capital. Roskilde entered into a “Transfer Agreement” with New Roskilde pursuant to which, inter alia, New Roskilde would on 25 August 2008 take over all assets of Roskilde as well as all of its debts and liabilities, other than Roskilde’s liabilities in respect of its subordinated loan capital, hybrid core capital, encapsulated savings bank fund and equity capital. Shares in Roskilde Bank were suspended.
On 26 February 2009, the Board of Directors of Roskilde filed a petition in bankruptcy. On or about 3 March 2009, the Bankruptcy Court of Roskilde, on the recommendation of the Danish FSA, appointed Jørgen Holst trustee of the bankrupt estate of Roskilde. On 5 August 2009, there was published a “Summary Of Legal Inquiry Into Certain Matters At Roskilde Bank” (the “Summary of Legal Inquiry”), prepared by Danish lawyers, Mogens Skipper-Pedersen and Henrik Stenbjerre, at the request of the board of directors of Roskilde. On 24 February 2010, New Roskilde commenced legal proceedings against Roskilde’s former management (including Mr Christensen) and auditors in the District Court of Roskilde as recorded in the Danish Writ of Summons.
In these circumstances, one might perhaps ask: what is the purpose of these proceedings? In the ordinary course, one might expect Taberna simply to advance a relatively straightforward contractual claim against Roskilde, seek payment of the amounts due under the subordinated notes and advance its claim in the Danish bankruptcy proceedings. However, it appears that any such contractual claim will not produce any recovery for holders of such subordinated notes. Moreover, although Taberna acknowledges that the liabilities transferred to New Roskilde under the Transfer Agreement exclude the contractual liabilities of Roskilde to pay principal and interest under the subordinated notes because those notes constitute subordinated loan capital, nevertheless Taberna’s position is that there is no such exclusion in respect of other liabilities including any liability that may be imposed on Roskilde in these proceedings to pay damages for misrepresentation. Roskilde does not admit that the liabilities transferred by Roskilde to New Roskilde under the Transfer Agreement would cover any such liability of Roskilde in damages, such as for misrepresentation as claimed in these proceedings. The parties are agreed that the precise scope of the Transfer Agreement is a matter of Danish law which does not arise in these proceedings. If necessary, it will be a matter to be resolved under Danish law depending on the outcome of these proceedings. So whether any judgment in these proceedings will ultimately prove of any value to Taberna is a matter which will ultimately be determined in Denmark.
The alleged representations and main issues
As originally pleaded, Taberna advanced some eight or more alleged misrepresentations in support of its claim, although a number were abandoned in the course of the trial. Come the time of final submissions, the alleged misrepresentations relied upon by Taberna were in respect of the following:
non-performing loans (NPLs);
credit policy;
write-downs for 2007-2008; and
project financing;
Roskilde denies Taberna’s claim entirely. In particular, Roskilde denies:
making any of the alleged representations to Taberna;
making representations in order to induce Taberna to invest in or become a holder of the subordinated notes;
that, in making the statements or expressing the views that it did to Taberna, it impliedly represented that it had reasonable grounds for doing so;
that Taberna was entitled to rely upon any representations made or that Taberna in fact relied on any of the representations made;
that any representations which it made were false;
that its management knew that any representations which were made by Roskilde were false, or that there was an absence of honest belief in the truth of the representations or recklessness as to their truth or falsity; and
that there were no reasonable grounds for Roskilde’s belief in the truth of any representations made.
In addition, Roskilde contends:
that Taberna cannot, in any event, rely on s2(1) of the 1967 Act because Taberna entered into a contract with Deutsche Bank to purchase the subordinated notes rather than with Roskilde;
that it is entitled to rely on certain disclaimers included in two of the documents that contained certain of the alleged representations as negating the representations or disentitling Taberna from relying on the representations;
alternatively, insofar as Taberna succeeds on its claim under section 2(1), that Taberna’s damages should be reduced to such an extent as the Court thinks just and equitable having regard to Taberna’s share in the responsibility for the damage. In particular, Roskilde says that Taberna’s decision to purchase the subordinated notes was caused, or contributed to, by its own misunderstandings, errors and/or failures to follow reasonable procedures for an investment of this scale.
In short therefore, the following broad issues arise:
Did Roskilde make the representations alleged by Taberna?
If so, did Roskilde make them in order to induce Taberna to invest in or to become a holder of the subordinated notes?
Did Taberna rely on the representations and was it entitled to rely on them?
What is the effect of the disclaimers relied upon by Roskilde?
Were the representations false?
If the representations were false, did Roskilde’s then management make them without reasonable grounds for believing them to be true?
Is Taberna entitled to rely on section 2(1) in circumstances where it purchased the subordinated notes from Deutsche Bank on the secondary market rather than from Roskilde itself?
What damages are recoverable by Taberna if Roskilde is found liable as alleged by Taberna?
Was Taberna’s decision to purchase the subordinated notes caused by or contributed to by its own misunderstandings, errors and/or failures to follow reasonable procedures for an investment of this scale?
The evidence
In what has sadly become a matter of standard practice, there was a large number of bundles containing hard copies of documents – many of which were never referred to during the trial. In large-scale modern commercial litigation, I regard this as unacceptable. It is a waste of time, money and resources – both natural and human and, in my view, it is a practice which should be abandoned. I readily accept that it is often impossible to say with certainty that particular documents will or may not have to be referred to in the course of the trial. But, in my view, there is no reason in this modern electronic age why documents which might be described – at best – as “peripheral” and which are unlikely to be referred to need to be reproduced in hard copy on a “just-in-case” basis. In most cases, it seems that with good pre-trial case management by the parties’ representatives, it would normally be sufficient for such documents to be available, if necessary, in soft-copy format – although, if needed, hard copies could no doubt be produced relatively quickly. At this stage, I refrain from making any further comment other than to note that it was agreed that the contents of all these documents were admissible in evidence as to the truth of their contents. Even more sadly, there was no satisfactory Core Bundle. There were, in fact, two bundles that were said to constitute the Core Bundle as required by the Commercial Court Guide. First, there was a bundle (D1) containing Roskilde’s financial documents including various formal financial reports. This was arranged very helpfully. Second, there was another bundle (D2) containing various contemporary documents including emails and other correspondence and reports. Unfortunately, this was something of a jumble. It was (often) not in chronological order (as it should have been) and the pagination was unsatisfactory. This was not only confusing but caused real difficulty from time to time in the course of the trial and in understanding the proper sequence of events. Insofar as may be necessary, I will consider separately who bears responsibility for this state of affairs and what, if any, special order as to costs should be made.
The organisational structure of Taberna and the evidence served on its behalf
On behalf of Taberna, written statements were served from the following witnesses, all of whom gave oral evidence:
Christian Ramamonjiarisoa (“Mr Ramamon”): Mr Ramamon was the former Vice-President of Taberna Securities (UK) Limited (“Taberna Securities”), which acted on behalf of Taberna at the relevant time.
Adam Schneider: Mr Schneider was the Managing Director of Taberna Securities at the relevant time.
Kenneth Frappier: Mr Frappier was the Chief Credit Officer and then Executive Vice-President of Risk Management, of RAIT Financial Trust (the parent company of Taberna Securities) at the relevant time.
In summary, as appears from that evidence, Taberna is, as I have stated, an Irish public limited company. Taberna Capital Management LLC (“Taberna Capital”) was the collateral manager of Taberna. As I have stated, Taberna Securities was another company which acted on behalf of Taberna at the relevant time. It was also the agent of Taberna Capital and formed part of a financial group comprising RAIT Financial Trust and Taberna Capital.
Taberna Europe CDO II Plc was one of a number of collateralised debt obligation vehicles which were, in effect, run in London by a team at Taberna Securities although it is important to note that any investments required the approval of the RAIT Financial Trust and the Taberna Capital Management Investment Committee (the “Credit Committee”) which consisted of a number of individuals working out of Philadelphia, USA.
At the relevant time, the most senior member of the team in London was Mr Mitrikov. He was the CEO of Taberna Securities as well as an Executive Vice-President of Asset Management at RAIT Financial Trust. Under Mr Mitrikov were 3 Managing Directors including Mr Schneider. His responsibility was to find issuers of debt suitable for investment. His expertise was on the banking side so he would lead the review of bank credits. Mr Schneider reported to Mr Mitrikov. The other main individuals of the London team were (i) Mr Ramamon whose main responsibility was to originate or to develop the relationships with the bank/debt issuers and their advisors; and (ii) Francesco Bellopede who was the lead credit analyst for investment opportunities which the team in London assessed and then passed to the Credit Committee for approval.
The Credit Committee consisted of four named individuals including Mr Frappier who was based in the offices of RAIT Financial Trust and, by his own description, was the Credit Committee’s “gatekeeper” in the sense that he worked with the team proposing any particular investment in order to weed out from the process those proposals that the Committee would never accept. The other members of the Credit Committee were Betsy Cohen, Daniel Cohen and Jack Salmon who were, as I understand, also based at the same offices. Three out of the four named individuals could approve an investment. Here, it was Betsy Cohen, Jack Salmon and Mr Frappier who duly approved the investment by Taberna in Roskilde’s subordinated notes on or about 8 February 2008 i.e. a week or so before the investment was actually made on 14 February 2008.
It is noteworthy that Taberna did not call as witnesses Mr Mitrikov, Mr Bellopede, Betsy Cohen or Jack Salmon or otherwise serve evidence from these individuals, all of whom were potentially important witnesses. On behalf of Roskilde, Mr Béar QC submitted that the failure to call Mr Bellopede was particularly significant and that the inference must be that he would have given evidence unhelpful to Taberna’s case on the issues he could have spoken to, viz. the representations which he understood Roskilde to be making (or not), reliance and contributory negligence. In that context, Mr Béar relied, in particular, on the decision of the Court of Appeal in Wisniewski v Central Manchester HA [1998] PIQR P324. In summary, Mr Béar thus submitted that the Court should draw the inference that Mr Bellopede’s evidence would contradict Taberna’s case on the following issues:
What oral representations were made to Taberna.
What understanding Taberna had of the information it had received concerning Roskilde, i.e. what it thought the alleged representations meant.
What reliance Taberna placed on the alleged representations, the importance to Taberna of particular issues now highlighted in its claim, and whether Taberna would have made a different decision with different information.
Whether Taberna’s approach to the investment analysis was negligent and contributed to any loss suffered.
In his submissions on behalf of Taberna, Mr Choo-Choy QC sought to give certain explanations as to why Mr Bellopede had not been called. However, such explanations were unsupported by any evidence. Absent any such evidence and bearing in mind that it would at least seem that Mr Bellopede could have been compelled to give evidence, it seems to me that there is much force in Mr Béar’s submission having regard, in particular, to the fact that even if Mr Choo-Choy is right in his further submission that Mr Bellopede was not the critical witness, there is no doubt that (as appears below), he played an important role. However, in the particular circumstances of the present case, the right approach, in my view, is to look at the entirety of the evidence bearing fully in mind the points made by Mr Béar.
Further, Mr Béar drew attention to the fact that none of Taberna’s witness statements addressed any of the points raised in Roskilde’s contributory negligence defence; and that the inference must be that they did not have any helpful evidence to give. That is true. But the fact is that this defence was raised by amendment only very shortly before the trial. In my view, the right approach is, again, to look at the entirety of the evidence bearing fully in mind this point.
In addition, Mr Béar attacked the approach by Taberna and also its legal representatives with regard to its failures concerning disclosure which Mr Béar submitted were both numerous and serious. In particular, he highlighted the absence of any document concerning the “investigation” and “board packs” mentioned by Mr Frappier; and the non-production of Mr Bellopede’s notebook mentioned by Mr Ramamon. On this basis, Mr Béar submitted that the “probable” conclusion was that the reason for these failures was conscious or unconscious bias by Taberna’s solicitors arising from the conflict of duty created by Taberna’s funding arrangements for the litigation i.e. bias causing unwillingness to make unfavourable disclosure. More specifically, Mr Béar submitted that Taberna’s disclosure had not been undertaken properly and that the reason for this was that it and/or its solicitors believed that it had relevant material which was liable to harm its case; and that such material related at least to (i) the initial (and obviously non-privileged) investigations referred to by Mr Frappier; and (ii) the investment strategy being pursued by Taberna at the time.
As to these submissions, I accept that there is a possibility that certain documents that did originally exist have not been produced; and that there may be some “gaps” in the disclosure. However, I reject the submission that this was the result of any deliberate withholding of documents still less any bias by Taberna’s solicitors. Again, it seems to me that the right approach is to look at the entirety of the evidence bearing fully in mind the points raised by Mr Béar.
As to the witnesses that Taberna did call, Mr Béar submitted that they were all generally unsatisfactory. In particular, he submitted that there were clear signs that the witness statements which they signed had been drafted in a hurry by Taberna’s lawyers not just in the normal sense of acting as an amanuensis but more fundamentally; and that there was no attempt to engage in the potential difficulties in Taberna’s case nor, as I have already stated, with regard to Roskilde’s contributory negligence defence. Further, Mr Béar submitted that Mr Frappier was notably argumentative and evasive and also displayed an ability to advance evidence with an air of confidence which was based on nothing at all. On the most charitable view, Mr Béar submitted that Mr Frappier is someone fully prepared and able to engage in quick, but false, reconstruction and then to convince himself that it is truthful evidence; and on a less charitable view, that he is prepared to lie. Further, Mr Béar submitted that in addition to his spurious air of confidence, Mr Frappier came across as dogmatic, probably something of a bully, used to getting his own way with anyone not perceived as his superior, and not willing to listen to contrary views. Mr Béar submitted that this was important in considering reliance because, as well as the mystery of Taberna’s overall strategy, concealed from the court by its failure to give disclosure, it was clear, as Mr Frappier admitted, that he “liked the credit”; and that that admission carries all the more weight because his witness statement failed to describe his own actual attitude at the time. Finally, Mr Béar submitted that Mr Frappier gave a strong impression of bitterness towards Roskilde and willingness to paint its conduct in the blackest terms which was entirely self-serving but casts further doubt over his evidence of his own and his colleagues’ actual, and hypothetical, conduct. As to Mr Ramamon, Mr Béar submitted that his English was so poor that his statement should have been done in his native language – in accordance with the Commercial Court Guide; and that, for a variety of reasons, it would be wrong to accept his evidence as reliable. As to Mr Schneider, Mr Béar seemed to accept that he was an honest witness; but, as Mr Schneider said, he had only a “limited memory” of the events in question.
As to these submissions, there is no doubt that Mr Frappier was a most forceful individual prone to making dogmatic self-serving statements which, in certain respects at least, I do not accept. Further, I agree that the evidence of Mr Ramamon was, at least in some respects, not reliable; and that, as Mr Schneider himself acknowledged, he had only a limited memory of the events in question. I also readily accept that in matters of this kind, in particular matters of reliance, it is all too easy for an individual to assert long after the event that he or she relied on a certain piece of information. For all these reasons, it seems to me that I should approach the oral evidence of Mr Frappier (and indeed Taberna’s other witnesses) with great caution paying due regard to the contemporary documents and inherent probabilities – as well as bearing in mind the alleged gaps in disclosure.
The organisational structure of Roskilde and the evidence served on its behalf
On behalf of Roskilde, written statements were served from the following witnesses all of whom gave oral evidence:
Leif Gebel: At material times, Mr Gebel was Head of Roskilde’s “Team Økonomi” department, responsible for dealing with financial reporting, budget, profit and loss management, and other forms of financial reporting to the bank’s management.
Marc Dalgas: Mr Dalgas joined Roskilde in June 2007 and was at material times the Head of Capital Markets at Roskilde. In particular, Mr Dalgas was responsible for debt investor relations (including making sure that Roskilde was represented at presentations with the Danish banking industry and with responsibility for dealing with Moody’s, the well-known credit rating agency) and for long-term liquidity; from January 2008, he also became responsible for short term liquidity.
Flemming Nielsen: At material times, Mr Nielsen was Head of Administration at Roskilde Bank, and then (from Q4 2007) Chief Financial Officer.
Mr Højgaard Andersen: Mr Andersen was at material times the deputy Head of Credit at Roskilde.
In summary, as appears from that evidence, the CEO of Roskilde was at all material times until 30 April 2007 Niels V Hansen. After a short gap, he was replaced on 1 July 2007 by Søren Kaaren-Andersen. The deputy CEO was Mr Arne Wilhelmsen. Mr Flemming Nielsen originally joined Roskilde as Head of Administration in 2005. He was promoted to CFO in the fourth quarter of 2007 and continued in that role until February 2009. Although he was not a board director of Roskilde, he sat on the Management Council which reported to the CEO. In his capacity as CFO, Mr Flemming Nielsen’s role included the management and oversight of Roskilde’s financial position, preparing quarterly, semi-annual and annual financial reports and liquidity management. However, credit, legal matters and corporate and retail lending were managed elsewhere.
In all, around 17/18 people reported to Mr Flemming Nielsen. These included, in particular, Mr Jesper Flensted Nielsen, the Head of the Risk Management Department; Mr Leif Gebel; and Mr Marc Dalgas.
There was a separate Credit Department at Roskilde; and also a Credit Committee consisting of the CEO (i.e. Mr Niels V Hansen until April 2007 and Mr Søren Kaaren-Andersen from July 2007), the Head of Credit (i.e. Mr Knud Nielsen until 1 April 2008 and Mr Peter Hauskov from 1 April 2008), the Deputy Head of Credit (i.e. Mr Andersen) and, from May 2004, the overall Credit Executive Officer (i.e. Mr Allan Christensen). According to Mr Andresen, this Credit Committee met daily to discuss individual cases. A team known as “Team Erhverv” (the “Corporate Department”) was a branch of the Credit Department run by Mr Lars Jensen and serviced by Mr Andersen and three other credit employees.
The expert evidence
In addition, both parties called an expert on matters of Danish accounting law and practice. Taberna’s expert was Jan Tønnesen (“Mr Tønnesen”) and Roskilde’s expert was H C Krogh (“Mr Krogh”). They both produced various written reports and gave oral evidence. In their primary expert reports, respectively dated 24 September 2014 and 16 October 2014, they addressed the following questions:
Under Danish law/accounting practice as at the end of 2007, on what basis were banks permitted and/or required to make write-downs?
Does the practice in relation to write-downs described at page 17 of Roskilde’s Annual Report 2007 accord with Danish law/accounting practice at the relevant time or, if not, how does it differ?
Under Danish law/accounting practice as at the end of 2007, on what basis were banks permitted and/or required to determine that loans were “non-performing”?
Whether any further write-downs after 31 December 2007 would have to be set against Roskilde’s Core Tier 1 capital (equity), whether such write-down would also reduce the amount of other permitted Tier 1 Capital, and whether Roskilde’s capital adequacy was under stress.
In their supplemental reports, respectively dated 31 October 2014 and 7 November 2014, the experts responded to each other’s views on the above four questions and addressed the following additional questions:
What is the relationship, if any, between the figure of approximately DKK 80 million referred to on pages 22 and 44 of Roskilde’s Annual Report 2007 and on page 10 of its Annual Results 2007 and the figure of approximately DKK 522 million referred to on page 17 of the “Virksomhedens Oplysningsforpligtelser” dated 8 February 2008?
What rules, including any transitional rules under the Basel II Treaty (to the extent that the Basel II Treaty was in effect), applied to Roskilde in respect of the financial reporting of, and/or disclosure of information about, its “loans in default” and/or “non-performing loans” as at the 2007 year end?
In addition, the experts produced a Joint Memorandum setting out areas of agreement and disagreement.
Outline of main events
The issue of the subordinated notes
Turning then to an outline of the factual background, the story starts with the issuance by Roskilde on 1 December 2006 of €80,000,000 Fixed/Floating Rate Dated Subordinated Callable Step Up Notes due 2014.
The Offering Circular
On 16 March 2007, Roskilde published an Offering Circular relating to a €2,000,000,000 Euro Medium Term Note programme (the “Offering Circular”). This is an important and detailed document but which it is unnecessary to set out in detail. For present purposes, it is sufficient to note the following.
First, on p2, it contained an express acceptance of responsibility in the following terms:
“The Issuer accepts responsibility for the information contained in this Offering Circular. To the best of the knowledge of the Issuer (having taken all reasonable care to ensure that such is the case) the information contained in this Offering Circular is in accordance with the facts and does not admit anything likely to affect the import of such information.”
Second, also on p2, it contained a “disclaimer” which provided in material part as follows:
“Neither this Offering Circular nor any other information supplied in connection with the Programme or any Notes (a) is intended to provide the basis of any credit or other evaluation or (b) should be considered as a recommendation by the Issuer or any of the Dealers that any recipient of this Offering Circular or any other information supplied in connection with the Programme or any Notes should purchase any Notes. Each investor contemplating purchasing any Notes should make its own independent investigation of the financial condition and affairs, and its appraisal of the creditworthiness, of the Issuer ...
Neither the delivery of this Offering Circular nor the offering, sale or delivery of any Notes shall in any circumstances imply that the information contained herein concerning the Issuer is correct at any time subsequent to the date hereof or that any other information supplied in connection with the Programme is correct as of any time subsequent to the date indicated in the document containing the same. The Dealers expressly do not undertake to review the financial condition or affairs of the issuer during the life of the Programme or to advise any investor in the Notes of any information coming to their attention.”
Third, on p51, it contained a description of Roskilde’s loan portfolio stating, in particular, that it was relatively active in the real estate area (34%) and describing a breakdown of its loan portfolio related to real estate in the following terms:
“6% to the financing of property projects. The projects are typically ring-fenced in SPVs, where all rights under the project is [sic] pledged or assigned to the Bank, and most importantly the projects are generally sold or leased before the project is activated. The Banks [sic] approves the buyer or lessee in each project. The projects have in general a maturity of 12-18 months.” (emphasis added)
Fourth, on p53, under the heading “Credit”, it contained a summary of Roskilde’s credit policy and operation of its credit procedures which was, in relevant part, as follows:
“The Bank has a conservative credit policy based on its activities in the market area which is reflected in the low losses. The Lending activities must create long-term and attractive business relationships. Any credit decision is based on the economical foundation of the borrower. Risks will always be sought to be covered to the largest possible extent by pledges and guarantees. Good faith and close co-operation with clients must ensure that any negative developments are discovered as quickly as possible.
There is a structured hierarchy of approval authority, whereby the local branch manager can approve loans up to a certain limit. Larger loans are presented to the Credit Department. Loans exceeding their limit go to the Credit Committee, which consists of a member of the Management and the head and the deputy head of Credit. Loan requests exceeding the authority of the Management are presented to the Board on a weekly basis. Thus, there are four levels in the approval hierarchy ...
The loan portfolio is monitored daily by the credit department, which refers directly to the Management.
The loan portfolio is subject to a provision process each quarter, where existing loan loss provision[s] are monitored and followed up on, and new risk of losses are provided for. This is approved by both the Management and the Board.
The Management has regular meetings with each of the branch managers and the credit department, where important figures of the branch are checked, such as all major loan facilities, loan loss provisions, overdrafts, risk profile in terms of retail lending, corporate lending and sector lending. In addition to that the credit department visits each branch at least once a year to control a number of randomly selected facilities.
At the monthly board meeting, the Management presents all facilities granted by the Board since the last board meeting. Furthermore, there is a presentation on a sector basis, whereby the Bank’s exposure to a given sector is discussed. This includes an overview of the 10-20 largest individual exposures to the sector, a risk/profit analysis and the credit department’s opinion on the expected development within the sector. It is ensured that all sectors where the Bank has exposure are presented at least once a year, or at the request of the Board.”
The Merrill Lynch Prospectus
On 4 April 2007, Merrill Lynch issued a prospectus (the “Prospectus”) to assist in the marketing of the subordinated notes which had been issued by Roskilde. The Prospectus incorporated by reference the terms of the Offering Circular.
October 2007
On 25 October 2007, Mr Ramamon received an email from a Mr David Sabbagh of Société Generale (“SocGen”) stating that “a client of ours” i.e. Roskilde was looking to issue subordinated paper that might interest Taberna. In the email, Mr Sabbagh gave certain further details and indicated a possible price range. He said that the Q3 results were out that day and stated: “Let me know what you think”.
This approach came to Mr Ramamon in his capacity as the main individual within Taberna’s team in the London office with responsibility for finding commercial investments and financial institutions to invest in. He was quite well known in the market and would generally be approached by email and invited to meetings. At that time, the two main investment vehicles operated by the London office (of which Taberna was one) held some US$1.5 billion. According to Mr Ramamon, the investment focus was to invest in the debt issues of financial institutions, real estate companies with property in Europe and structured credit Asset Backed Securities (“ABSs”). So, he immediately passed this email to Messrs Schneider and Bellopede. Later that day, Mr Sabbagh sent a further email to Mr Ramamon attaching a copy of Roskilde’s Q3 Report 2007 summarising certain of the information in the Q3 Report and referring to what he described as an “updated roadshow presentation” that Roskilde would publish later that day on the website.
In the course of the trial, there was some debate as to the role of Mr Sabbagh and SocGen. Relying in particular on the words in Mr Sabbagh’s first email that Roskilde was “a client of ours”, Mr Choo-Choy sought to assert that, in truth, they were to be regarded as agents acting on behalf of Roskilde and, in that context, Mr Choo-Choy sought to amend Taberna’s Particulars of Claim to plead such a case. In the event, I refused to allow such amendment for reasons which it is unnecessary to repeat.
Roskilde’s Q3 Report 2007
As referred to by Mr Sabbagh, Roskilde did indeed publish its results for the third quarter of 2007 (the “Q3 Report 2007” or more simply the “Q3 Report”) on 25 October 2007. The Report was published in both English and Danish, the English version containing a statement on p16: “In case of any discrepancies between the Danish and the English version of the financial report, the Danish version shall prevail.” Reading the English version, the Report was, and was on p26 expressly stated to be, approved by the Board of Directors of Roskilde. On the same page, it contained a statement that the Directors believed that the Report “… gives a true and fair view of the assets and liabilities, the financial position, and of the result of the activities of the bank for the Q1-Q3 2007”; and a declaration that the “… Report contains a true and fair and analysis of the bank's activities and financial standing as well as a description of the most important risks and uncertainties that may influence the operations of the bank.”
The Q3 Report is a detailed document which it is unnecessary to set out in full. For present purposes it is sufficient to note the following as appears in the English version.
First, on p4 under the heading “Activities”, it recorded that “The Management has launched a strategy process with the purpose of defining the most important focus areas for the next two to three years. The strategy process is expected to be completed at the end of November 2007.” Under the same heading, it also set out a “financial calendar" identifying the date of 8 February 2008 for publication of its Annual Report 2007.
Second, on p6, under the heading “Write-downs on loans etc.”, it stated that, on the basis of the forecast published by the Economic Council of the Labour Movement in Denmark that the price of owner-occupied flats might fall by a further 15%, “the bank does not expect to incur losses on building projects with unsold units”.
Third, on p8 under the heading “Capital structure”, it stated:
“The capital adequacy ratio is 13% – well above the statutory requirement of 8%.
The core capital ratio is 8%.
The large growth in loans and the exposure on the mortgage financing market mean that the bank’s minimum capital adequacy requirement was 10.3% at the end of Q3.”
Fourth, on page 13 under the heading “Additional Key Figures”, there was a detailed table of figures going back to 2003. In particular, under the sub-heading “Asset quality”, the table stated in material part as follows:
2007
Non-performing loans, NPL DKK 56,821
…
NPL/Gross loans and guarantees 0.14
It is noteworthy that that latter figure (i.e. 0.14) is to be compared with the equivalent figures for each of the previous years which were 0.14, 0.24, 0.41 and 0.69 respectively.
One of the major issues in this case concerns the proper meaning of the term “non-performing loan” or “NPL”. Although it appears that the term is often used in banking operations, it was common ground that there was no universal internationally accepted definition or meaning of the term. Various possible meanings have been canvassed in the course of the trial including (i) a loan where there has been a default (e.g. non-payment) of any kind; (ii) a loan where there has been a “significant” default e.g. a default which has extended for a period of time over (say) 30 days or 90 days and remains unrectified in whole or in part; (iii) a loan where there has been a default (whether or not “significant”) or there is at least a risk of default and the lender has agreed with the borrower to amend the terms of the loan e.g. to forego or to reduce interest payments either for a limited time or forever. I should emphasise that these are merely non-exhaustive examples of possible meanings that have been canvassed. I address this aspect further below. However, at this stage, I would merely note that (i) on p16 of the Q3 Report there is a “disclaimer” which states: “In case of any discrepancies between the Danish and the English version of the financial report, the Danish version shall prevail”; and (ii) the Danish version for what is translated in the English version as “Non-performing loans, NPL” uses the following Danish words: “Rentenulstillede udlån”. Translated literally, these words have a specific meaning in English: “Interest reset” or “Loans with reset interest”. As appears further below, this is to be contrasted with the Danish words: “Misligeholdte fordringer” which was translated as “Loans in default” (I was told that another possible translation is “delinquencies”) and would seem to have been understood to be equivalent to “non-performing loans”. I deal with this aspect further below.
The “Investor Presentation/Road Show” and “Roskilde’s Bank Presentation”
At about the same time as the publication of the Q3 Report, Roskilde also published what it described as an “Investor Presentation” for a “Road Show” as well as a separate “Bank Presentation”. These are in similar format consisting of what appear to be a series of coloured slides with some large bold print and various diagrams and charts. (I should mention that these followed on from an earlier presentation prepared in August 2007 – but nothing turns on that.) For present purposes, it is sufficient to focus on the first of these presentations (which I shall refer to as the “IP”) which was prepared by Mr Dalgas and which, as he explained, was not only put on Roskilde’s website but also used specifically in the course of various investor presentations including one which he conducted together with Mr Nielsen in Ireland in December 2007.
The IP contained in material part the following references:
On p4, under the heading “Earnings Drivers”, it stated that: “We will maintain our conservative risk profile”.
On p6, under the heading “Performance Highlights”, it stated:
“2007:Q3 2007:Q3
MDKK MEUR
Total impairment, write downs on loans etc. 38 5”
On p7, under the heading “Credit Risk”, it stated:
“Conservative risk policy Structured hierarchy of approval authority Long record of low credit losses.”
Underneath, there is further text and diagrams including a bar chart entitled “Coverage for non-Performing Loans”. Unfortunately, this is impossible to reproduce. However, in essence it provided information over the years 2003-2006 and for two quarters viz 2006Q3 and 2007Q3 in respect of what are referred to as (i) “non-performing loans”; (ii) “loans-loss provision”; and (iii) “Coverage ratio”. In particular, it shows for 2007Q3 non-performing loans of DKK 57 million; loans-loss provision of DKK 299 million; and a coverage ratio of approximately 5.
On p15, under the heading “Credit Risks”, there is a further bar chart entitled “Non-Performing Loans” giving relevant information with regard to what are described as “Non-Performing Loans” for each quarter between 2005Q2 and 2007Q3 both in absolute terms and as a percentage of gross loans. In relevant respect, the figures given for 2007Q3 are as stated in the Q3 Report i.e. DKK 57 million and 0.14% respectively. (I should mention that the line on the chart supposedly representing the relevant percentage figures had been drawn inaccurately – but nothing turns on this point.)
On p22, there is a “Disclaimer” which provides in material part as follows:
“This presentation has been produced by [Roskilde]… solely for use by investors met during the non-deal roadshow made in connection with the release of the bank’s Q3 2007 figures and may not be reproduced or redistributed to any other person without permission. This presentation is only directed at persons who have professional experience in matters relating to investments.
This presentation may contain certain forecasts made in statements relating to the business, financial performance and results of the bank and/or the industry in which it operates. Any such statements contained in this presentation, including assumptions, opinions and views of the Bank or cited from third party sources, are solely opinions and forecasts which are uncertain and subject to risks. A number of factors can cause actual events to differ significantly from any implied or anticipated development. Neither the Bank nor any officers or employees can guarantee that the assumptions underlying such statements are without errors nor does either accept any responsibility for the future accuracy of any opinions given in this presentation or the actual occurrence of any forecasted developments. No representation or warranty (expressed or implied) is made as to, and no reliance should be placed on, any information, including projections, estimates, targets and opinions, contained herein, and no liability whatsoever is accepted as to any errors, omissions or misstatements contained herein, and, accordingly, neither the Bank nor any officers or employees accepts any liability whatsoever arising directly or indirectly from the use of this presentation for any purpose …
The Bank is under no obligation to update or revise the information contained herein and will not publicly release any amendments it may make that may result from circumstances arising after the date of this presentation. The Bank accepts no responsibility for the accuracy of its sources.”
November/December 2007
In the course of November, Mr Ramamon had further contact with Mr Sabbagh as well as with a Mr Nils Colldahl of HSBC with regard to Roskilde; and work was being done internally in Taberna’s London office to evaluate investment opportunities. In particular, Mr Bellopede was analysing various European banks as possible subjects for investment including Roskilde which, in an internal email dated 7 November 2007, he described in the following terms: “… good rating for Moody’s NPL 0.1%, wholesale funding can be a problem but lets talk about it tomorrow …” The email identified a number of other banks including Marfin and Sparebank where he referred specifically to the relevant NPL figure. According to Mr Schneider, this was an important consideration because NPLs were, generally speaking, a good indicator of the overall quality of a bank’s loan portfolio: the higher the level of NPLs of a bank, the poorer the overall quality of its loan portfolio, which in turn entailed a greater risk of default by the bank’s borrowers and consequently a greater risk of loss to the bank.
Other internal emails within Taberna’s London team about this time also show discussions concerning pricing of the Roskilde subordinated notes and the possibility of investing in real estate companies.
On 12 November 2007, Mr Bellopede sent an internal email to Mr Schneider and Mr Ramamon with regard to Roskilde attaching certain information and stating:
“For me the credit profile looks ok, good profitably and efficient high quality of loan portfolio (high percentage of coverage), 45% deposits in their founding and some exposure to wholesale funding (but their unsecured lending is mitigated by a very low NPL) … I think that it's only a matter of spread …”
Attached to this email was a preliminary analysis carried out by Mr Bellopede with regard to Roskilde showing an NPL figure of 0.14% of the gross amount of Roskilde’s loans and guarantees for 2007 compared with loan loss provisions of about 0.71% of the gross amount of Roskilde’s loans and guarantees. It is impossible to say for certain where Mr Bellopede obtained these figures but, on a balance of probability, I accept that the very strong likelihood is that they were taken from the Q3 Report and the IP.
By the end of November 2007, Taberna’s London team had produced a list of some 41 possible investments (including not only banks but also other entities) which were set out in a spreadsheet entitled “Taberna Europe Pipeline” each with an indicated “probability” of likely investment. Roskilde was listed at number 19 with a 50% probability of likely investment of €25 million in the subordinated notes – a figure which Mr Schneider had himself assessed.
Following further internal discussions, Taberna’s London team decided that Roskilde merited further consideration and to press on with the due diligence process. The contemporaneous documents also show work commencing by the London team under the supervision of Mr Bellopede for the preparation of a report on Roskilde for submission to the Credit Committee in the USA.
January 2008
So far as Roskilde is concerned, the documents show that there were internal exchanges/discussions covering the financial statement and IP. For example, on 10 January 2008, Mr Flemming Nielsen sent an internal email stating:
“Dear all
Considering that there is 0% advance sales and 0% advance rentals in [particular name],I would like us to discuss the following:
1. Should we still state – and write – that: “usually, 70-80% has been rented out or sold in advance and as a minimum always 50%? (announced most recently by FBN+MAD in Ireland in December).
2. If not: What do we say/write instead?
3. How does it really look? (Do we have other in the range from 0% to 50% - or 70/80%).
4. Which explanation should we give for the write-downs? If any.
5. AOB
6. Another meeting necessary later in the month? (for financial statements + investor presentation)”
A first draft of an internal Credit Report on Roskilde was produced by Mr Bellopede dated 10 January 2008. This document extended to 13 pages and included detailed financial information and various tables, diagrams and charts concerning the history of the Bank, shareholders structure, overview of the Bank in the Denmark banking sector, banking activities including mortgage loans, securities portfolio, funding and capital, income statement, comparisons with other Danish financial institutions, net profit, balance sheet and details of the members of the Supervisory Board. In addition, it contains what is described as a “sensitivity analysis” showing “efficiency ratios” and “coverage ratios” between 2001-2006.
Again, it is impossible to say for certain where Mr Bellopede obtained the information contained in this Credit Report. Given the similarity of certain of the information and language, it seems relatively plain that much must have been taken by him from the Q3 Report and the IP and also the Offering Circular. But, as appears below, there are certain points where the position would seem to be otherwise and the source of the information is uncertain.
The information in this Credit Report is obviously important in the context of these proceedings but I do not consider that it is necessary to quote extensively from it. For present purposes, it is sufficient to note the following:
On p1, there is a summary table headed “Key financials and Data” viz:
Key financials and Data | |||||
Figures Presented in DKK million | 3Q 2007 | 2006 | 2005 | 2004 | 2003 |
Total assets | 41,580 | 30,502 | 18,697 | 12,903 | 10,021 |
Capital Adequacy Ratio (%) | 13.2 | 8.9 | 10.4 | 9.8 | 11.4 |
Return on average assets (%) | 1.74 | 2.12 | 2.75 | 1.75 | 2.41 |
Return on average equity (%) | 25 | 33 | 43 | 26 | 29 |
Efficiency ratio (%) | 38.58 | 40.64 | 40.28 | 50.12 | 48.68 |
Shareholders’ equity | 2,684 | 2,494 | 1,705 | 1,146 | 1,056 |
Loans | 30,058 | 23,765 | 14,962 | 9,750 | 7,028 |
Deposits | 17,821 | 12,360 | 9,506 | 6,387 | 5,119 |
NPLs/Loans (%) | 0.14 | 0.21 | 0.35 | 0.81 | 0.71 |
Loan loss provisions/Gross loans | 0.71 | 0.77 | 1.64 | 2.37 | |
Core capital ratio | 7.8 | 7.9 | 8.3 | 8.5 | 8.6 |
Net Interest Margin (%) | 2.8 | 2.96 | 3.20 | 3.54 | 3.83 |
Operating income/Interest + pref (x) | 1.51 | 1.67 | 1.33 | 2.23 | |
S&P | NA | NA | NA | NA | NA |
Moody’s | A2 | A3 | A3 | A3 | A3 |
Fitch | NA | NA | NA | NA | NA |
1 DKK=0.1340 EUR
As to the figures in this table, I would at this stage merely note that the source of at least some of the information is unclear. For example, the figure for nonperforming loans of 0.14% for 2007Q3 is identical to the figure which appears, for example, on p13 of the Q3 Report. However, the corresponding figures which are indicated in the above table for the earlier years i.e. from 2003 to 2006 are different to those which appear on p13 of the Q3 Report. There are also other differences.
Also on p1, there are two lists of “Key Strengths” and “Key Risks”. The Key Strengths were listed as follows:
“– Roskilde Bank is a leading local bank in Copenhagen area (40-50% in the area around Roskilde and 5-10% in its other areas of operations)
Good asset quality with strong coverage of non-performing loans (0.14% of the gross loans)
Capital adequacy ratio (CAR) and Core Capital Ratio remarkably high (13.2% and 7.8% respectively)
Roskilde represents one of the best Nordic banks in terms of profitability and efficiency ratios (25% and 38.5% respectively)”
The Key Risks were listed as follows:
“- Roskilde Bank has a strong deposit base funding but this is steadily declining in recent years; this risk is mitigated by a very good liquidity buffer with the liquidity ratio at 126% as of Sept 2007 (minimum requirement 100%)
- Some credit concentration in the lending portfolio due to the regional nature of this bank
- Future revenue growth is at risk given intense competition in the national market and the limited franchise network of the bank”
On p4, there is an explanation relating to the basis upon which Roskilde made project financing loans in the real estate sector which stated in relevant part as follows:
“The Bank is relatively active in the real estate area (45%) and the loan portfolio related to the real estate can be broken down as follows:
- 22% to the financing of property projects. The projects are typically ring-fenced in SPVs, where all rights under the project is [sic] pledged or assigned to the bank, and most importantly the projects are generally sold or leased before the project is activated. The bank approves the buyer or lessee, the duration and the terms in the project. The projects have in general a maturity of 12–18 months and in the case of housing, a specific number of units must be sold before construction work is commenced, typically at least 50 – 70%.”
It is noteworthy that the percentage figures there stated of 45% and 22% are approximately those referred to on p7 of the IP; and that otherwise the language in this passage tracks virtually word for word what appears on p51 of the Offering Circular which I have already quoted above. In my judgment, these are probably the sources of the information which Mr Bellopede used to prepare this part of the Credit Report.
On p4, there is also a passage concerning Roskilde’s credit approval and monitoring policies which again tracks the language in the Offering Circular (on p53) quoted above and which again I find was probably the source used by Mr Bellopede to prepare this part of the Credit Report.
The following day i.e. 11 January 2008, Deutsche Bank sent the Prospectus to Taberna. In addition, about this time, the documents show that Mr Bellopede downloaded and printed off Moody’s Reports about Roskilde Bank dated March 2007 and about the Danish Banking System Outlook dated November 2007 and also the Offering Circular.
It appears from the contemporaneous documents that Mr Bellopede wanted further information to produce a final version of the Credit Report for submission to the Credit Committee in the USA and in anticipation of a conference call with Mr Frappier which was scheduled to take place in the near future. After conferring with Mr Schneider and obtaining Mr Schneider’s approval, Mr Bellopede sent an email dated 14 January 2008 to Mr Maerklin at Deutsche Bank saying that Taberna was hoping to close new deals in 2008; informing him that Taberna would like to know some additional details regarding Roskilde in order to complete their due diligence; and setting out a list of questions (in particular with regard to company structure, composition of Board of Directors, list of securities in Roskilde’s portfolio, the increase in share capital in 2007 and the reason why Roskilde’s share price had fallen which Roskilde was requested to answer.)
While still waiting for answers to this list of questions, Mr Bellopede produced the second draft of the Credit Report dated 14 January 2008. In broad terms, this is substantially similar to the first Credit Report. Certain of the information in the table on p1 was amended (including an increase to the stated efficiency ratio in 2007Q3 from 38.58% to 42.36%) but such changes are not material for present purposes. In particular, the figures otherwise stated for 2007Q3 were all unchanged. This was circulated internally by Mr Bellopede to Messrs Schneider, Ramamon and Mitrikov later on that day i.e. 14 January 2008 informing them that he was still waiting for additional information from Roskilde and asking Mr Schneider to let him know if he (Mr Schneider) wanted more comments for the financial performance of Roskilde.
Meanwhile, it appears that Deutsche Bank passed on Mr Bellopede’s list of questions to Roskilde. This request was dealt with by Mr Dalgas who responded by email to Deutsche Bank on 16 January 2008 who in turn immediately forwarded Mr Dalgas’ response to Taberna. This is an important chain of communication. For present purposes, it is sufficient to note that I find that, although this information passed through Deutsche Bank, it was, in effect, a communication from Roskilde which Mr Dalgas on behalf of Roskilde knew would be likely to be passed on to Taberna and intended would be relied upon by Taberna in the context of its purchase of the subordinated notes. For present purposes, it is sufficient to note that (i) included in the email was a statement with regard to Roskilde’s property book (in particular its project financing element of that book) that “[its] exposure was low” in that segment of the market; and (ii) Mr Dalgas attached to his email what he described as a “few files” which included the original Offering Circular and a single sheet public announcement by Roskilde signed by its CEO i.e. Mr Andersen dated 14 January 2008 to the Danish Stock Exchange (the “14 January Announcement”). In translation, it reads in relevant part as follows:
“Downgrade of the 2007 result
After the ordinary review of the banks accounts, [Roskilde] will increase the write-downs on a number of major accounts in the fourth quarter of 2007. Accordingly a total of DKK 267 million will be written down and set aside as provision against potential, future losses. The amount is equivalent to 0.6% of our total loans and guarantees …
2008 Forecast
Core earnings for 2008 are expected in the DKK 610-620 million range including write-downs of DKK 150 million on loans etc.
The 2007 Annual Report will be published Friday, 8 February 2008 as previously announced.”
It was the evidence of Mr Schneider (which I accept) that he probably read the 14 January Announcement following its receipt and that both his understanding and the understanding of Taberna’s team involved in the due diligence exercise into Roskilde was that Roskilde honestly and reasonably believed that the write-down of DKK 267 million for 2007 and the estimated write-down of DKK 150 million for 2008 were or would be appropriate write-downs to provide for in respect of Roskilde’s loan portfolio.
In an e-mail dated 29 January 2008 from Roskilde (Mr Nielsen) to Deutsche Bank (Mr Paxeus) and passed on (as, I find, Roskilde knew and intended) by Deutsche Bank to Taberna Securities (Messrs Bellopede and Ramamon), Roskilde stated as follows:
“We have never had a loss on any of the guarantees, and the risk of losses in the future is small – both in numbers of LGD and in actual figures (mio DKK). However, prices on houses have declined the past year, so we might see minor losses this and next year. Total loss will be insignifikant [sic].”
Thereafter, the contemporaneous documents show continuing internal discussions within Taberna’s London team and also between that team and both Deutsche Bank and Roskilde. In particular, it appears that there was at least one and possibly more than one conference call direct between Taberna’s London team and representatives of Roskilde during which further questions were put to Roskilde and subsequently answered by Roskilde in an email which was passed on by Deutsche Bank to Taberna on 22 January 2008. Again, this was, as I find, an email which Roskilde knew would be passed on to Taberna with the intention that it would be relied upon. For present purposes, it is the answer to question 5 which is potentially significant viz:
“5. [Q] Have you made provisions for future losses on these instruments? How much (if you are at liberty to say)? [A] Markets have deteriorated further since our announcement last week. So the remaining portion of the portfolio is at the moment showing a loss in line with expectations. Other investments made in the liquidity portfolio have to a certain extent recovered a part of this development.”
At about this time, it appears that Mr Bellopede produced a further version of the Credit Report on Roskilde. Although that third version also bears the same date as the second version i.e. 14 January 2008, there can be little doubt that it was in fact produced some time after that date. In any event, it appears that Mr Frappier received a copy of that Credit Report dated 14January 2008 from Mr Schneider on 23 January 2008. In his written statement, it was his evidence that he noted, in particular, the following points:
The second bullet point on p1 that there was good asset quality within the loan portfolio with 0.14% of loans being non-performing loans and strong coverage of non-performing loans (more than five times).
The information on p4 which I have already quoted above with regard to project finance and credit approval/monitoring.
The information on p8 based on the 14 January Announcement that Roskilde had decided to write down DKK 267 million for 2007 which was still only a modest 0.6% of its total loans and guarantees.
It was Mr Frappier’s evidence that when he saw this information about Roskilde (whether it related to NPLs, estimated write-downs or other estimates or forecasts), he assumed that Roskilde had an honest and reasonable basis for putting out such information.
Later that day, Mr Frappier emailed Betsy Cohen, Daniel Cohen and Jack Salmon a list of credits for discussion including the potential investment in Roskilde. Thereafter – probably on 24 January 2008 – there was a conference call between members of the London team and Mr Frappier to discuss the possible investment in the Roskilde subordinated notes. According to Mr Frappier, he was particularly interested in Roskilde’s off balance sheet commitments which could affect the recourse question. However, it appears that the Taberna London team were unable to provide satisfactory answers in that regard during the call and Mr Frappier did not think they had prepared themselves adequately. This prompted Mr Mitrikov to circulate an internal email to his team sating: “Guys, we kind of sucked on this call. Do you agree?” In any event, it was agreed internally that further work needed to be done with regard to the operation of the Danish market, the leverage on typical loans made by Roskilde compared with leverage generally in the market; and to concentrate on the “biggest question” i.e. recourse.
A few days later i.e. on 28 January 2008, Mr Ramamon emailed Mr Maerklin certain further questions which had been reformulated by Mr Bellopede in order to complete Taberna’s due diligence. In the event, written answers were duly provided by Mr Flemming Nielsen in an email dated 29 January 2008. For present purposes, it is sufficient to note that in answer to one of the questions posed by Mr Bellopede, Mr Flemming Nielsen responded as in paragraph 66 above. He also stated that Roskilde had no plans to issue covered bonds.
On 30 January 2008, there was a further conference call between members of the Taberna London team and Roskilde. Following that call, Mr Bellopede updated the Credit Report; and after further internal exchanges, Mr Bellopede produced a final version of the Credit Report dated 30 January 2008. This was sent by email to Mr Frappier and circulated internally to the members of the London team in the evening on that day. This contained various further amendments including on p5 at the end of a section headed “Banking activities”:
“According to the information provided by the management, the bank wrote down around DKK 200 million of its mortgage portfolio (land and property unfinished) during the fourth quarter 2007, due to the slowdown in the Denmark property market and in the meantime conservatively increased its loss provisions to 1% of the total loan portfolio. The Bank expects additional DKK 150 million write downs of its mortgage portfolio in 2008 but this risk is mitigated by the large cushions in the form of loan-loss reserves (increased to 1% of the total loan portfolio).”
February 2008
A few days later i.e. on 5 February 2008, Mr Schneider circulated further copies of the finalised Credit Report dated 30 January 2008 to the other members of the Credit Committee in the USA i.e. Betsy Cohen, Daniel Cohen and Jack Salmon. This was in advance of a conference call between members of the Credit Committee and the London team which had been scheduled for the following day i.e. 6 February 2008 to discuss the possible investment in the Roskilde notes and also another possible investment. Later that day, Betsy Cohen emailed Mr Frappier saying:
“… Have you looked at the Danish bank carefuly [sic]. Their loan portfolio has grown very fast.”
He responded almost immediately saying:
“… have concerns about the Danish Bank due to its significant residential exposure and heavy reliance on sub debt funding due to less robust deposit growth. I have asked Adam to focus the discussion on the mitigants to the resi loan concentration and why we should be comfortable with the debt leverage.”
A few minutes later, he sent a further email to Mr Schneider in London saying:
“… Roskilde Bank will be tougher [i.e. as compared with the other possible investment] due to its resi exposure and heavy reliance on sub debt with slowing deposit growth. You should pay particular attention to those issues in your presentation.”
On the following day i.e. 6 February 2008 and shortly before the scheduled call was due to take place, Mr Bellopede sent to Mr Mitrikov a Reuter’s article published a few days earlier i.e. on 3 February 2008. The article was headed “Nordic Banks eye edge in subprime crisis” and stated in its opening paragraph:
“Nordic banks remain largely insulated from the effects of the sub-prime crisis, but their shares have tracked sector peers lower since last August and may now be bargain priced.”
The article goes on to cite various analysts’ views. One quote, which is perhaps typical of the main theme of the article, was to the effect that although people in London said that this is the worst financial crisis in 20 years, in the Nordics “… banking is very much business as usual …” Mr Mitrikov immediately forwarded the article to Mr Frappier saying “Thanks. Ken – take a look …”; and Mr Frappier immediately forwarded it on to Betsy Cohen and Jack Salmon saying: “This may be helpful in our discussion of the Denmark bank today.”
The conference call between the Taberna London team and the three members of the Credit Committee in the USA did indeed take place on 6 February 2008. During the call, Mr Schneider made a presentation. However, there is no written record or even summary or notes of what he – or the other participants – said. Somewhat surprisingly, Mr Schneider said nothing in his written statement about the nature of his presentation or the discussion that took place; nor did Mr Ramamon say anything about this call – although it is not clear to me whether he was a party to the call; and, of course, neither Betsy Cohen nor Jack Salmon gave evidence. The only person who did give evidence on what seems to me this important topic was Mr Frappier who stated in his witness statement as follows:
“40. At the meeting Betsy Cohen asked for additional information on the concentration in real estate assets and the risk in the construction loan portfolio folio. We were concerned about the quality of that growth which made the confirmation of sales of 50 to 70% of houses on construction projects all the more critical. That deposit growth was decreasing but subordinated debt (as a source of Roskilde’s capital funding) was increasing was something to consider. That is not necessarily a disadvantage because deposits can simply be withdrawn. In our discussion we felt that Roskilde had good quality low growth opportunities that outstripped deposit growth. Whilst sub-debt had allowed Roskilde Bank to fund its expansion we were ultimately comfortable that they had these tight credit policies and procedures (as summarised in the Credit Report) to control the loan growth and to properly control the risk in the loan portfolio. Their very low NPLs and write-downs seemed to confirm this. Adam Schneider told the Committee that Taberna would be borrowing 27m DKK, explained how it was being priced and would have talked about the matter of covered bonds and Totalkredit. I remember particularly that the 50 – 70% general pre sale condition was a significant piece of the presentation on which we relied to approve the investment together with the other contents of the Credit Report.”
In any event, it is common ground that no final decision was taken by the Credit Committee for Taberna to go ahead with the purchase of the subordinated notes at that stage – although it is not clear exactly why this was so. I suppose one possibility is that everyone was waiting for the publication of Roskilde’s 2007 Annual Report which was due on 8 February 2008 but it is impossible to make any positive finding to that effect.
Roskilde’s VO document - Roskilde’s Disclosure Requirements
Meanwhile, Roskilde’s management was in the course of finalising its financial results for 2007 which were due for publication on 8 February 2008. Apart from the 2007 Annual Report, one of the documents which Roskilde was required to publish was entitled “Virksomhedens oplysningsforpligtelser” (the “VO document”). In translation, this means: “The company’s disclosure requirements”. The publication of the information contained in this document was, in effect, part of Roskilde’s statutory obligations under newly enacted (or at least modified) Danish law pursuant to Danish Executive Order 10113 dated 22 December 2006 on Capital Adequacy (“Order 10113”). Pursuant to s60 and Annex 20 para 6(g) of Order 10113, Roskilde was required in particular on 8 February 2008 (the date of its intended publication of its Annual Report 2007) to disclose the total value of its “defaulted loans” or “defaulted exposures”.
There are issues between the experts as to whether Annex 3, paras 20 and 21 of Order 10113 – which referred to loans or exposures in respect of which there were overdue payments or arrears of more than DKK 1,000 for more than 90 days – contained a definition of “defaulted loans” for the purposes of the disclosure obligation under Annex 20 para 6(g) or whether the legislative intention behind Annex 20 para 6(g) (when read in conjunction with para 6(a)) was to confer a discretion upon banks to choose their own definition of defaulted loans. Whatever may be the answer to these issues, what is clear is that, by 31 January 2008, Roskilde (through Mr Gebel and those assisting him) had calculated the total of its defaulted loans for disclosure purposes to be about DKK 3.5 billion in accordance with the definition of loans in arrears contained in Annex 3 paras 20 and 21. This appears from an email which Mr Gebel sent in the course of preparation of the VO document dated 31 January 2008 to Mr Flemming Nielsen (and others) under the heading “Corporate disclosure obligations” attaching what he said was a suggestion for “page 21” which met the Danish FSA's requirements and “… has been expanded by a column showing non-performing loans divided into industries …”. That particular “page 21” has not been disclosed and apparently can no longer be found – although another page headed “THE COMPANY’S DISCLOSURE REQUIREMENTS” and marked “2” which has been disclosed and which contains a table referring to various industry sectors showing total “loans in default” of DKK 3,539,683 (i.e. approximately DKK 3.5 billion) would seem likely to be another version of what Mr Gebel was probably referring to.
In any event, Mr Flemming Nielsen responded almost immediately to Mr Gebel’s email in terms which indicate that in the version of the table as it existed in the attachment sent by Mr Gebel, the figure stated for non-performing loans was some DKK 3.5 billion. On any view, that was a very large figure i.e. over 10% of Roskilde’s total loan portfolio; and the fact that Mr Flemming Nielsen’s response email is headed “RE: Corporate disclosure obligations – the pacemaker skipped a beat!” would indicate that he was truly shocked by the size of that figure. The body of Mr Flemming Nielsen’s email (which he copied to others including Mr Andersen) further confirms that he was indeed shocked by this figure viz:
“… I hope the excerpt is at least partially misleading (DKK 3.5 billion in the non-performing loans column) - is there any hope that some of the client numbers mentioned cover an entire group and that the number could thereby be reduced … Or is this really the number?
It won't look pretty to the outside world that 10% of the lending portfolio is in arrears but if that is indeed the case, the next challenge is tackling the communication in the wake of this …”
The email concluded by saying that he had booked a meeting for 5:30 pm to discuss the matter.
Shortly thereafter, Mr Wilhelmsen (the Deputy CEO) responded by email stating:
“Something must be completely wrong in the entered definitions … Terms such as non-performing loans could be misinterpreted.
Just because clients are in arrears, surely we cannot define it as non- performing loans.
I am anxiously awaiting your reply regarding this.”
Mr Flemming Nielsen then forwarded this email to Mr Gebel and others saying: “For inspiration for the meeting”.
A little later, Mr Gebel sent a further email stating (with original underlining):
“Just some input on defaulted definitions
According to the Danish Financial Supervision Authority, who use the standard method (as do we), institutes can use the same definition as appears from Appendix 3, items 20 and 21 in the Capital Adequacy Regulation (entries with our arrears):
Definition of arrear: when a counterparty has been in arrears or overdraft with an amount that is considered substantial for more than 90 days. It is a case of arrear when the counterparty fails to make payments when they fall due, fails to honour their debt on an agreed date, or when the granting of a maximum overdraft and similar is exceeded.
In order for arrears to be substantial, the total amount in arrears on the counterparty’s commitment, cf. Article 5 Section 1 and no. 16 in the FILE, must constitute more than DKK 1000 as regards the company, the company's parent and its subsidiaries.
Right off the bat, I believe that the excerpt we have made of DKK 3.5 billion meets the above mentioned definition – if that is not the case, I will ask Jesper and/or Allan to enter the scene.
If we do not apply the above-mentioned official definition at the end of 2007, we must indicate how we define non-performing loans.
However we need to bear in mind that the official definition must be used in connection with the solvency statement at the end of March 2008.
This will result in the need to weight the DKK 3.5 billion at 150%. Therefore, action must be taken to reduce the amount – otherwise all things being equal, our solvency percentage will decreased by 0.9% points.”
The relevance of the figure of about DKK 3.5 billion is that (i) if it represented (as by all accounts it appears to have represented) the loans on Roskilde’s books which were in arrears of more than DKK 1000 by more than 90 days and (ii) if the NPL term was properly understood by Taberna as including loans which were in arrears by more than 90 days then the NPL figure which Roskilde would be required to disclose should have been that figure i.e. DKK 3.5 billion.
The scheduled meeting duly took place later that afternoon i.e. on 31 January 2008. It was obviously an important meeting. Perhaps surprisingly, there are no notes, minutes, reports or other documents relating to that meeting and it is unclear whether any such documents ever existed (perhaps because the matter was dealt with on an urgent basis) or whether they did exist but are no longer available. In any event, the result was that an email was sent by Mr Knud Nielsen to all Roskilde’s branch managers that very same evening at 7.38pm which stated in material part as follows:
“Overdraft
In our New Year's letter, we wrote about our objective to reduce overdrafts older than 90 days. The objective was a 50% reduction up until 30/06/2008.
In the meantime, however, we recognise that reality has caught up to us faster than we what we were wishing for (sic). With reference to the Basel II regulations, we are already now able to establish that the total commitments with old overdrafts were too big.
Therefore, you will be receiving a list of your commitments that fall due for the 90 – day criterion sometime on 1 February.
In the context of Basel, commitments with old overdrafts are regarded as destitute. The bank must publish this number, which is much too large, on our website by the end of next week. I would therefore like to ask you to process the matter immediately upon receipt of the list and approve overdraft if the client is creditworthy. Unfortunately, I must ask you to please process all clients with a commitment in excess of DKK 5 million already by Friday, 1 February, so that loan applications can be in the hands of Team Credit by next Monday, so that applications can reach Team Credit by Tuesday. Please approve the cases you are able to personally approve by next Monday.
Please process the rest as soon as possible and by the end of February, after which we expect to have no overdrafts older than 90 days …”
In evidence, Mr Flemming Nielsen accepted that, although not stated precisely, the exercise required by this email was designed to effectively turn unauthorised overdrafts into authorised overdrafts with effect from 31 December 2007 i.e. retrospectively; that this would explain why the urgency was being pressed upon all branch mangers; and that the process of the conversion of unauthorised overdrafts into authorised overdrafts should also carry on into the future. Further, Mr Flemming Nielsen accepted that if a loan had been in arrears for more than 90 days and would therefore have to be disclosed pursuant to the disclosure rule, it was not, in fact, appropriate or fair to avoid the operation of the rule by simply authorising such overdraft from an earlier date.
For his part, Mr Gebel in his oral evidence sought to distance himself from this exercise. In particular, he stated in cross-examination: “I can’t imagine that overdrafts would be granted retrospectively”because “… that would be the wrong thing to do.” Further, he agreed that that would be consciously disregarding the requirement for disclosure. Nevertheless, there can be no doubt, in my view, that that is the exercise which the branch managers were, in effect, required to perform by Mr Knud Nielsen’s email.
By 4 February 2008, it appears from an email sent by Mr Gebel that morning that the “work initiated in relation to arrears” by Mr Knud Nielsen was still under way and that the defaulted loans figure had still at that stage not been reduced from the DKK 3.5 billion figure. It also appears from the last sentence of that email that Mr Gebel had begun to consider the advice received from Lopi (a trade association of regional banks), including that Roskilde could use “internal statements of loans in default” if it had not received statements of arrears from BEC, the Danish banks’ IT provider. Thereafter, further internal meetings took place on 5, 6 and 7 February 2008 during which there were further discussions between the individuals who had attended the meeting on 31 January 2008 as well as Mr Knud Nielsen. The gist of those discussions is not entirely clear, save that according to Mr Gebel the discussions related to the defaulted loans issue.
By 7 February 2008, it appears that a further or different way had been found to reduce the defaulted loans figure by some 85% from DKK 3.5 billion to DKK 521,783. According to Mr Gebel, the way in which this new defaulted loans figure of about DKK 511 million was arrived at was “by adding up the overdrafts on the daily list of overdrafts as per year-end 2007” – although the relevant overdraft list or lists have not been produced by Roskilde. Nevertheless, it was Mr Gebel’s recollection that this is how the figure of about DKK 511 million was reached. As submitted by Mr Choo-Choy, this would seem to be consistent with Mr Gebel’s evidence that, in reliance on Lopi’s advice that the disclosure requirement under Order 10113 might be met “by using the institution’s list of overdrafts”, Roskilde decided to use its lists of overdrafts to calculate the defaulted loans figure and what can be seen from the monthly notifications to the Roskilde Board of Directors of overdrafts of more than DKK 5 million. If this is the correct explanation for the figure of DKK 511 million, then as submitted by Mr Choo-Choy, it is clear that that figure is not of the gross amount of loans in default or in arrears for more than 90 days, but only an aggregate of the amounts overdrawn in respect of such loans, and that the gross amount of the loans would have been a materially higher figure, as Mr Gebel fairly accepted in evidence.
Mr Gebel’s evidence was that he “felt comfortable that the figure was correct” although, as he himself there stated, he did not do the exercise himself - he only received the final figure. In his witness statement, Mr Gebel also stated that, once Roskilde appreciated in accordance with guidance provided by ‘Lopi’ that it could use its internal overdraft lists in order to calculate the NPLs (pursuant to “transitional arrangements, rather than immediately adopting the Basel II approach) “… it was not a problem to reach the 522 million. This figure was not treated or produced in any artificial way. It was just the total of the overdrafts.”
In any event, the result of the exercise carried out was to produce a table in the final VO document which was published on 8 February 2008 which showed the figure of DKK 521,783 under the heading “Misligeholdte fordringer” instead of the original figure of about DKK 3.5 billion – although it is important to emphasise that Mr Béar’s submission was that these two figures were not comparable in particular because of the manner in which the former figure was calculated as described above.
The VO document is another lengthy and detailed document extending to 22 pages covering numerous topics including targets and risk policies with regard in particular to credit risks, market risks, liquidity risks, operational risks, business risks and risks relating to Roskilde’s capital base. Most of the information in the VO document is not directly relevant. For present purposes, it is sufficient to focus on p17, Section 5 headed “Credit risk and dilution risk”. This is similar in form to the earlier draft table already referred to above – but there are important differences. In material part, the new table at p17 stated as follows:
“Definitions and methods
The definitions for accounting purposes of loans in default and impaired loans as well as a description of the methods applied for determining value adjustments and write–downs are set out in the Danish Executive Order on Financial Reports for Credit Institutions and Investment Firms, etc … Roskilde Bank complies with the Executive Order, and we therefor refer to sections 51-54 of the Executive Order. As the bank uses the transitional arrangement for the implementation of Basel II, the calculation of loans in default is based on the banks lists of overdrafts.” (emphasis added)
Underneath this text, there is a table under the heading “Individual loans in default and impaired loans, and write-downs by sector at year-end 2007” containing 4 main columns, the first of which is headed “Misligholdte fordringer” or (in the English translation) “Loans in default” although (at the risk of repetition) I was told that another possible translation for the original Danish is: “Delinquencies”. The total as shown at the bottom of this column is the figure referred to above i.e. DKK 521,783. In addition, there is a column for “Loans and guarantee debtors for which write-downs/provisions have been made” with a total of DKK 1,586,255 and “Write-downs/provisions at year-end” with a total of DKK 446,482.
These figures – in particular the original figure for what Mr Gebel described as non-performing loans of about DKK 3.5 billion and the reduced figure of about DKK 511 million – and the exercise described above that led to this reduced figure and revised table being included in the final published VO document were the focus of much debate in the course of the trial. The nature and propriety of the exercise carried out in arriving at this substantially reduced figure for (using the Danish term) “Misligholdte fordringer” was a major part of Mr Choo-Choy’s attack in the trial although it was Mr Béar’s submission that the exercise carried out complied with Roskilde’s disclosure obligations under Section 60 and Annex 20 para 6(g) of Order 10113 and was, therefore, entirely legitimate and proper. So far as necessary, I consider this below. However, at this stage and at the risk of repetition, it is sufficient to note a number of points highlighted by Mr Choo-Choy in his final submissions with regard to this exercise viz:
None of the overdraft lists allegedly used in this re-calculation – which resulted in a seven-fold reduction in the NPL figure – have been disclosed by Roskilde or any of its witnesses;
There is no explanation of the distinction between the internal records used in calculating the original DKK 3.539 billion figure and the overdraft lists used in arriving at the much reduced DKK 522 million figure;
Nor is there any explanation of how and why those internal records differed from the contents of the overdraft lists – the enormity of the difference is all the more puzzling since the internal records originally used must have allowed Roskilde to determine that the DKK 3.539 billion of loans in question satisfied the definition of NPLs in Appendix 3 items 20 and 21 of the capital adequacy rules (i.e. overdue for payment for more than 90 days and for more than DKK 1,000);
Indeed, as Mr Gebel himself put it in his 3:14 pm email on 31 January 2008, after referring to the NPL definition in the capital adequacy rules: “[r]ight off the bat, I believe that the excerpt we have made of DKK 3.5 billion meets the above-mentioned definition.”
The 2007 Annual Report
On 8 February 2008, Roskilde also published a document entitled “Annual Report Announcement” (the “2007 Announcement”) as well the main 2007 Annual Report (the “2007 Annual Report”). Both these documents were published in Danish as well as English. A copy of the 2007 Announcement in the English version but not the former was immediately forwarded by Deutsche Bank to Taberna early in the morning i.e. at 9.18 am London time. It is important to note that Taberna never themselves looked at the Danish version of the 2007 Announcement nor the 2007 Annual Report (either in Danish or in English).
Roskilde’s 2007 Annual Report Announcement
So far as the English version of the 2007 Announcement is concerned (which, to repeat, Taberna did see), the potentially significant information may be summarised as follows:
On p4, under the heading “Strategy 2010”, it stated that Roskilde had established a “new strategy for the period till 2010” including, with regard to “risk management”, “a whole new focus area enabling us to live up to present and future legislation”.
On p4, it showed, in effect, an intention to reduce the proportion of its loan portfolio in respect of real estate from 42% in 2007 to a “goal” of 25% during the period 2008-2013.
On pp8-9, under the heading “Write-downs on loans etc.”, it stated:
“In 2007, the bank increased write-downs on loans by DKK 266 million equal to 0.6% of the bank’s loans and guarantees. The accumulated write-downs amount to 1.2% ...
Due to the continued uncertain economy growth prospects, the requirement for write-downs is expected to be DKK 150 million in 2008.”
On p10, under the heading “Capital and solvency”, it stated:
“The capital adequacy ratio was 12% at the end of 2007, and the core capital ratio was 7%. The individual requirement of the bank was set at 10.5%.”
On p12, there was a disclaimer which stated:
“In case of any discrepancies between the Danish and the English version of the financial report, the Danish version shall prevail.”
On p13, under the heading in large type “Main figures”, there was a detailed table of financial information covering the five year period 2003-2007. In particular:
Under the sub-heading “Profit and loss account”, the figure in 2007 for “Loan impairment write-downs etc.” was stated to be DKK 266 million.
Under the sub-heading “Credit risk”, the figure for “Percentage of loan assets at reduced interest rate” was stated to be 0.2%. This was broadly in line with the stated figures for the previous years viz 0.1%, 0.2%, 0.6% and 0.5%.
On p21, under the heading in large type “Equity and solvency”, there was a table with further financial information including under the sub-heading “Solvency” figures for 2007 in respect of “Capital adequacy ratio” – 11.9; “Core capital ratio” – 7.0; and “Statutory capital adequacy requirement” – 8.0.
In broad terms, it was common ground that the English version of the 2007 Announcement was a relatively good translation of the Danish version. The only suggested potentially relevant difference was in respect of certain wording on p13 of the English version i.e. “Percentage of loan assets at reduced interest rate”. Strictly translated, the Danish version reads “Percentage of loan assets with reduced interest rate.” In my view, this difference is of no practical significance.
Roskilde’s 2007 Annual Report
The 2007 Annual Report was a much fuller and detailed document extending to 52 pages. To repeat, it was never seen by Taberna before the internal approval of the deal by the Credit Committee on 8 February 2008 nor even before the actual purchase went through on 14 February 2008 – although it was common ground that it was a document which was available and which Taberna could have obtained shortly after its publication on 8 February 2008 and certainly before the actual purchase went through. For present purposes, it is sufficient to say that, as one would expect, it contained general statements with regard to historic performance and future “goals” and much detailed financial information; that it was signed off and approved by the Board of Directors; and that it was also signed off by both Roskilde’s internal auditors and external auditors (Ernst & Young) as giving a true and fair view of Roskilde’s financial position at 31 December 2007. For present purposes, it is sufficient to note the following:
On p6, there was a table headed “Main figures” including, under the subheading “Credit risk”, figures for “Percentage of loan assets at reduced interest rate” which for 2007 was stated to be 0.2.
On pp15-19, there was an important and lengthy section under the heading “Risk Management” dealing with a number of matters. In particular, this included a passage under the sub-heading “Credit Risks on Corporate Accounts” which stated:
“Project financing typically has a maturity term of 1-18 months. The bank requires security in the project and approves the contractor. Also, the project will be supervised by the bank’s own building experts. In the case of residential projects, a certain number of units must [typically] be sold before construction is commenced. The advance sale to buyers providing a cash deposit or a bank guarantee as security is to ensure that the bank’s credit facility can be redeemed upon finalisation. The advance sale requirement is typically between 50 and 70%. In the case of commercial projects the bank requires that a sales agreement on the property, or, as the case may be, a lease where the tenant and the term of the lease are approved by the bank, is obtained before the construction has commenced.” [The word equivalent to “typically” included in square brackets above appears in the Danish version but this is omitted in the English version.]
Also under the “Risk Management”, there is a further passage under the sub-heading “Policy for Write-downs on Loans etc, and Provisions” which stated in material part:
“We currently review our loans etc for any objective criteria for loan impairment and the consequent need to write any individual accounts down. Furthermore, we consider whether any general impairment of our consumer as well as corporate loans is indicated.
The assessment of impairment by subgroups of homogeneous loans is based on models involving our rating schemes. The models have been completed in 2007 in line with the specific accounting instructions in this respect.
If a loss is considered inevitable, we make a provisional write-down on the account in question and monitor the account on a current basis. Major impairments trigger a concrete assessment at least twice a year. The final write-down or write-off is made when the account is finally closed (bankruptcy, rescheduled debt, etc.).
The bank follows up on the written-off claims on a regular basis in order to recover our claims. If the bank considers interest computation to lead to further losses, the account will be transferred to the category of non-performing and non-accrual loans. Any claim against customers in order to recover the uncomputated interest will be upheld, however.
For information on write-downs and non-performing loans, cf note 25.”
On p22, there was a table headed “Additional key figures”, including under the sub-heading “Asset quality”, figures for “Non-performing loans, NPL”, the stated figure for 2007 being DKK t. 80,216.
On p31, there was the Balance Sheet which referred, at the bottom, to supplementary notes at the end of the Report. With regard to these notes:
On p38, included in note 12 are the figures for “Total loans” which were stated to be (approximately) DKK t. 32,551 (2007) and DKK t. 23,765 (2006).
On pp43/44, included in note 25 under the heading “Credit risk”:
The 2007 figure for “Accumulated write-downs end of year” is stated to be DKK t. 443,267. The text states: “The Bank has partially written some of its loans and outstandings on the basis of an objective indication of impairment.”
The 2007 figure for “Total write-downs on loans etc” is stated to be DKK t. 502,541. Including write-downs on guarantees etc, the 2007 figure is slightly higher i.e. DKK t. 534,293
Under the sub-heading “Write-down on loans and outstandings”, the following figures are stated for 2007 for “Claims at reduced interest rate end of year” i.e.
Amount DKK t. 80,216
Percent of loans, guarantees and provisions 0.2%
Roskilde’s 2007 Annual Results Presentation
Also on 8 February 2008, Mr Andersen, the CEO, produced a presentation on Roskilde’s 2007 annual results, the slides for which were published on Roskilde’s website entitled: “Annual Results 2007”. Included in this presentation was, at p10, a bar chart headed “Coverage for Non-performing Loans” which, in effect, showed “Non-Performing Loans” at the end of 2007 of DKK 80 million; and at p11, under the heading “Real Estate Portfolio” and sub-heading “Project Financing”, the statement: “LTV 80-85% as a general rule – Presale condition of 50-70% …”.
Taberna’s purchase of the subordinated notes
Meanwhile, Mr Maerklin had also emailed Mr Ramamon that morning i.e. 8 February 2008 saying that Deutsche Bank’s trader “… has received various bids for this position. I would need to know if Taberna is going to invest or not by today. I will not be able to protect this amount without any firm commitment from your side on the EUR 27 Mio position. Can you please let me know on your timing, processes.” That email was forwarded by Mr Ramamon to other members of the London team including Mr Mitrikov who immediately replied: “We will try to come back to him, but today is probably not realistic for a final decision. I understand we might lose the deal but I think Deutsche is bluffing (still we’ll try to get it approved fast.)” Somewhat surprisingly perhaps, Mr Schneider then sent an email a few minutes later at 9.54 am to Mr Kane of Citibank (the administrators for the subordinated notes) stating: “We are going to purchase the attached bonds in the following amounts: Roskilde - €27m …” However, later in that same email he said that he (Mr Schneider) would get back to Mr Kane“… as soon as I firm trades for the cash bonds …” The fact is that at this stage Taberna’s Credit Committee in the USA had not yet given its final approval to the purchase of the subordinated notes. According to Mr Schneider, the reason he sent the email in the terms he did was because he wanted to make sure that Citibank would be ready to proceed with the steps required to effect the purchase of the subordinated notes if and when the decision to purchase was made by the Credit Committee.
In the course of that same morning, it would appear that Mr Bellopede reviewed the English version of the 2007 Announcement (but not the other documents referred to above published on 8 February 2008) and prepared a briefing setting out his “considerations” which he sent in an email to other members of the London team at 12.38pm on 8 February 2008; and Mr Mitrikov forwarded it on shortly thereafter to Mr Frappier together with a copy of the 2007 Announcement. Surprisingly, it does not appear that the 2007 Announcement or Mr Bellopede’s briefing were ever forwarded to the other members of the Credit Committee. Exactly what happened during the rest of that day remains somewhat unclear. Although not referred to in either Mr Frappier’s or Mr Schneider’s witness statements, the contemporaneous documents would seem to indicate that there was a call between Mr Frappier and the London team when further questions were raised; and also a further call between Taberna and Roskilde’s management when Roskilde provided Taberna with further information concerning “loan growth” and its strategy to diversify its loan portfolio. This resulted in Mr Bellopede producing a further email summarising this new information which he sent to Mr Frappier and the rest of the London team by email later that afternoon. It would appear that this was forwarded by Mr Frappier to Betsy Cohen and Jack Salmon shortly thereafter with his own summary explanation as follows:
“The loan growth was primarily related to an acquisition which included deposits as well as a loan portfolio primarily consisting of retail and small business loans. They are focused on reducing their resi exposure but have not had any losses in this portfolio. The off–balance sheet exposure related to the "covered bond" program is already accounted for by the company in their risk based capital adequacy analysis.”
Mr Frappier ended his email to Betsy Cohen and Jack Salmon by saying: “I recommend we approve the investment. Let me know if you would like to schedule another call with the group to discuss further.” Virtually instantaneously, Betsy Cohen responded by email saying: “Thanks for the explanation. I approve.” Jack Salmon responded by email the following morning saying: “I approve based on the clarification re credit issues”. Mr Frappier confirmed such approval to the London team on Monday 11 February 2008; and a formal resolution was adopted and signed by Mr Frappier, Betsy Cohen and Jack Salmon approving the purchase of the subordinated notes (as well as other investments). As already noted, the actual deal when Taberna purchased the subordinated notes and transferred the purchase money went ahead a few days later i.e. on 14 February 2008.
Reviews of Roskilde’s operations
I have so far attempted to follow a broadly chronological summary of the main events culminating in Taberna’s purchase of the subordinated notes. However, it is important to note that throughout this period, Roskilde’s own management carried out various reviews of its operations; and Roskilde was also the subject of review by third parties including the Danish FSA. So far as relevant, I address these matters in the context of the specific issues which I consider later in this Judgment.
The Law
As I have already noted, Taberna abandoned any claim in deceit or negligence at common law and restricted itself to a claim under s2(1) of the 1967 Act which provides as follows:
“1) Where a person has entered into a contract after a misrepresentation has been made to him by another party thereto and as a result thereof he has suffered loss, then, if the person making the misrepresentation would be liable to damages in respect thereof had the misrepresentation been made fraudulently, that person shall be so liable notwithstanding that the misrepresentation was not made fraudulently, unless he proves that he had reasonable ground to believe and did believe up to the time the contract was made that the facts represented were true.”
Thus, it was common ground that the burden lay on Taberna to show (i) that Taberna entered a contract; (ii) that the contract was entered into after a misrepresentation had been made to Taberna “… by another party thereto …”; (iii) that, as a result thereof, Taberna suffered loss; and (iv) that Roskilde would be liable in damages in respect of such loss had the misrepresentation been made fraudulently. Subject to proof of those four matters and one important point of principle, it was also common ground that Roskilde would be liable in damages unless it proved that it had reasonable ground to believe and did believe up to the time the contract was made that the facts represented were true. Notwithstanding this broad agreement as to the applicable principles, it is necessary to clarify certain points and address a number of general issues before turning to consider the particular alleged misrepresentations relied upon by Taberna.
Reliance
First, by way of clarification, I should say that Mr Choo-Choy accepted that it was trite law that to found a claim under s2(1), the representee must show that he relied on and was induced by the relevant representation. However, Mr Choo-Choy submitted that the representee need not show that he relied solely and exclusively on the representation; and that it is sufficient that the representation plays a real and substantial part, albeit not a decisive part, in inducing the representee to act. In support of that submission, he relied upon the following authorities:
Dadourian Group International Inc v. Simms (Damages) [2009] EWCA Civ 169, where the defendants were found liable for fraudulent misrepresentation and Arden LJ at [99] summarised the approach which the judge (Warren J) had adopted at first instance (which she approved at [101]):
“(1) it is a question of fact whether a representee has been induced to enter into a transaction by a material misrepresentation intended by the representor to be relied upon by the representee; (2) if the misrepresentation is of such a nature that it would be likely to play a part in the decision of a reasonable person to enter into a transaction it will be presumed that it did so unless the representor satisfies the court to the contrary (see Morritt LJ in Barton v County NatWest Limited [1999] Lloyd's Rep Banking 408 at 421, paragraph 58); (3) the misrepresentation does not have to be the sole inducement for the representee to be able to rely on it: it is enough if the misrepresentation plays a real and substantial part, albeit not a decisive part, in inducing the representee to act; (4) the presumption of inducement is rebutted by the representor showing that the misrepresentation did not play a real and substantial part in the representee's decision to enter into the transaction; the representor does not have to go so far as to show that the misrepresentation played no part at all; and (5) the issue is to be decided by the court on a balance of probabilities on the whole of the evidence before it.” [Emphasis added]
Assicurazioni Generali v. Arab Insurance Group [2002] EWCA Civ 1642, where Clarke LJ stated at [59]:
“It seems to me that the true position is that the misrepresentation must be an effective cause of the particular insurer or reinsurer entering into the contract but need not of course be the sole cause. If the insurer would have entered into the contract on the same terms in any event, the representation or non-disclosure will not, however material, be an effective cause of the making of the contract and the insurer or reinsurer will not be entitled to avoid the contract. Thus I agree with Sir Christopher Staughton, whose judgment I have seen in draft, that, in this context at least, causation cannot exist when even the ‘but for’ test is not satisfied; cf the recent decision of the House of Lords in a very different context in Fairchild v Glenhaven Funeral Services Limited [2002] UKHL 22.”
This should be contrasted with what Ward LJ said at [218]:
“I am happy to express my agreement with the analysis of the law conducted by Clarke L.J. subject to this reservation. I am not entirely sure that it is necessary to require the misrepresentation to be an effective cause of a party’s entering into the contract on the terms on which he did. If by that qualification my Lord means no more than that it did actually play upon his mind and influence his decision then I have no argument. In other words I readily accept it must have some causative effect. I would be concerned if the insistence on an effective cause were to lead to an evaluation of the weight placed by the representee upon the various matters which in combination lead to the agreement. We must be careful not to be led back into the error that the cause has to be a decisive cause.”
The but for test suggested by Clarke LJ and Sir Christopher Staughton was followed by Christopher Clarke J (as he then was) in Raiffeisen Zentralbank v Royal Bank of Scotland 2010 EWHC 1392 (Comm) and also by Hamblen J in Cassa di Risparmio della Republica di San Marino S.p.A. v Barclays Bank Ltd [2011] EWHC 484 (Comm), at [467].
In the light of those authorities, I accept Mr Choo-Choy’s submissions as stated above.
In this context, it is also convenient to mention a general theme in Mr Béar’s submissions which is relevant generally in the context of Taberna’s alleged reliance. In particular, Mr Béar submitted that it was apparent that there were many other reasons impelling Taberna towards the acquisition of the subordinated notes which did not involve the alleged reliance on the fine detail of the credit report which, as he submitted, underpins the whole of this claim. In summary, Mr Béar submitted:
As Mr Frappier readily admitted, he “liked the deal”. Indeed, he liked it so much that Mr Schneider told Citibank (the administrators of Taberna’s investment fund) early on 8 February that the deal would be proceeding. This not only suggests that the result of the committee decision of 8 February was a foregone conclusion (because Mr Frappier and his team would find ways to make it attractive to the other members) but also indicates that they were engaged in such a process from before that point in time.
The very large inducement offered by the “fee” of €810,000, increased for no explicable reason from €405,000 which had been expected only 2 weeks earlier. The suggestion that chasing fees was not a major incentive, or any incentive, is simply naïve given the obvious pressure on Mr Hogentogler of RAIT to report something in the fee pipeline to Daniel Cohen at the top of that business. Mr Schneider could not explain why his report to Mr Hogentogler focused only on the fee-earning deals but the answer is surely obvious.
The impelling factor of Taberna’s then-rosy view of the Nordic property market i.e. Taberna had a positive view of a high-performing market.
The obvious attractiveness of something which had no taint of sub-prime.
The murky background of Taberna’s overall strategy which has not been the subject of proper disclosure (and hence justifies an adverse inference). The background of negative publicity relating to other Cohen family businesses (which had previously held Taberna) only adds to this concern.
The formulaic and exaggerated quality of much of Taberna’s evidence, coupled with the failure to get to grips with obvious features of the facts. For example, Mr Frappier asserted with great vehemence in his written evidence that they would not have invested in a business with hundreds of millions of kroner of NPLs yet shortly before, Taberna had invested in AB Ukio Bankas of Lithuania, with 1.79% of NPLs (equivalent to 9 times the figure they claim they believed for Roskilde, or about DKK 720m if it had been applied to Roskilde).
In light of the above, it was Mr Béar’s broad submission that the Court is simply not in a position to be satisfied that Taberna has demonstrated that it would not have reached the same decision even if it had received different information about the transaction in question. I do not accept that submission. For example, the fact that Mr Frappier admittedly “liked the deal” (as he did) and that it was attractive because it was not (or at least did not appear to be) tainted by sub-prime (as it was) does not necessarily mean that Taberna would have entered into the deal regardless of the specific information that was received relating to the deal; and, for the avoidance of doubt, it is my clear view that this was not the case. However, I fully recognise that the points summarised above provide an important backcloth to the central question of Taberna’s reliance which I should and do take into account; and highlight the importance of considering critically the particular alleged misrepresentations which Taberna says were relied upon in its decision to purchase the subordinated notes.
Relevant contract ?
Mr Béar submitted that s2(1) has no application in the circumstances of the present case because (i) the remedy under s. 2(1) of the 1967 Act is only available against the other party to the contract under which the loss is suffered; and (ii) the “contract” under which Taberna sustained the reliance loss which it now claims (the price it paid for the purchase of the subordinated notes) was the contract with Deutsche Bank to which Roskilde was not a party. Thus, Mr Béar submitted that the claim under s2(1) of the 1967 Act fails at the outset.
In support of that submission, Mr Béar relied on a number of points which he summarised in his closing submissions as follows:
S2(1) was designed to regulate pre-contract negotiations. A contracting party can regulate all the terms of the bargain as well as the content of his pre-contractual statements. The former can take account of the latter, the most obvious examples being by choosing not to enter into the bargain at all, or by requiring exclusion clauses which (subject to the statutory reasonableness controls imposed in the same enactment) the representor can also agree with his counterparty, the representee, as well as by any other adjustment of other terms. If entry into the contract with the representor causes the representee loss, the representor can be made responsible for that loss in the same way as a fraudster, i.e. the tort measure.
This coherent scheme does not hold for the present case brought against Roskilde where the pre-contract negotiation was between the parties to the sale contract, Taberna and Deutsche Bank. In particular:
While correct that the result of that contract was to bring Roskilde and Taberna into a contractual relationship, that is fortuitous. The asset in question might as easily have been rights against borrowers rather than against Roskilde, for example, if the asset had been a package of loans originally granted by Roskilde.
The loss sustained by Taberna, and claimed in this action, was the price paid to Deutsche Bank under the contract with Deutsche Bank. No loss was sustained as a result of entering into the contract with Roskilde. This can be easily tested by considering what claims for damages Taberna could make if it could not plead its contract with Deutsche Bank – the answer is none, since it is the price under that contract which is the claimed loss.
Point (b) above is sufficient to take the case outside s 2(1). Further, the operative contract was that between Taberna and Deutsche Bank. That contract alone was induced (if anything was induced) by representations made by Roskilde; the contract between Roskilde and Taberna was simply a consequence of Taberna having acquired the notes from Deutsche Bank and was not itself induced by any representations.
As to these submissions, I readily accept that the facts of the present case are somewhat unusual. In particular, this is not a simple case of only two parties (A and B) where a representation is made by A to B and, in reliance on such representation, B enters a bilateral contract with A. Here, the position is more complicated. Thus, it is plain that at least certain pre-sale negotiations took place between Deutsche Bank and Taberna; that Taberna entered into a contract with Deutsche Bank; and that it was pursuant to that contract (to which Roskilde was not a party) that Taberna acquired the subordinated notes from Deutsche Bank. However, there is equally no doubt, and Mr Béar accepted, that the effect of such acquisition was to bring Taberna and Roskilde into a contractual relationship - although the precise mechanism whereby such contract came into existence is not entirely clear to me. It is perhaps also noteworthy that, contrary to a “normal” contract, the consideration for the subordinated notes i.e. the purchase price was paid by Taberna to Deutsche Bank not Roskilde. However, I am unpersuaded that these somewhat unusual features take the present case outside the scope of s2(1) although, of course, the burden remains on Taberna to prove the matters referred to above. For the avoidance of doubt, I would emphasise that if the position were that the relevant representations were made by Deutsche Bank rather than by Roskilde, then, of course, Taberna could have no claim against Roskilde under s2(1): cf: Chitty on Contracts, (31st Edition) Vol 1 para 6-024. However, that is not the case advanced by Taberna.
Contributory Negligence
Mr Béar submitted that even if it be held that Roskilde is liable under s2(1) of the 1967 Act, Roskilde would be entitled to rely upon s1(1) of the Law Reform (Contributory Negligence) Act 1945 (the “1945 Act”) which provides as follows:
“Where any person suffers damage as the result partly of his own fault and partly of the fault of any other person or persons, a claim in respect of that damage shall not be defeated by reason of the fault of the person suffering the damage, but the damages recoverable in respect thereof shall be reduced to such extent as the court thinks just and equitable having regard to the claimant’s share in the responsibility for the damage …”
Fault is defined in s4 as follows:
“… negligence, breach of statutory duty or other act or omission which gives rise to a liability in tort or would, apart from this Act, give rise to the defence of contributory negligence …”
The law on this topic is summarised in Chitty on Contracts (31st Ed) vol 1 para 6-079:
“It was held in GranGelato Ltd v Richcliff (Group) Ltd that damages for negligent misrepresentation under s.2(1) of the Misrepresentation Act 1967 may be reduced under s.1 of the Law Reform (Contributory Negligence) Act 1945 if the loss was partly the fault of the representee. Liability under s.2(1) applies unless the representor “had reasonable grounds to believe and did believe … that the facts represented were true” and thus is “essentially founded on negligence”. However, it would not be just and equitable to reduce the damages when the representor had intended, or should be taken as having intended, that the representee should act in reliance on the answers which had been given to his questions. The decision was based on the fact that there was concurrent liability under s.2(1) and in tort for negligent misrepresentation under the principle of Hedley Byrne & Co Ltd v Heller & Partners Ltd.It may happen that a defendant is liable under s.2(1) without being concurrently liable in tort for negligent misrepresentation, for instance because the court considers that there was on the facts no undertaking of responsibility towards the claimant. In such a case it seems that the claimant's damages could not be reduced on account of any contributory negligence. This is because s.2(1) makes the misrepresentor who cannot prove reasonable grounds liable as if the statement had been fraudulent. It has been held that at common law contributory negligence is not a defence to fraud and that therefore the Law Reform (Contributory Negligence) Act does not apply to fraud. Because the misrepresentor is to be liable under the Misrepresentation Act 1967 s.2(1), “as if the representation had been fraudulent”, the Law Reform (Contributory Negligence) Act seems not to apply to claims under s.2(1) where there is no concurrent liability in tort for negligent misrepresentation.”
I did not understand Mr Choo-Choy to challenge what was stated in Gran Gelato; nor what is stated in this passage from Chitty. However, Mr Choo-Choy submitted that Roskilde’s plea of contributory negligence under the 1945 Act cannot avail Roskilde in circumstances in which Roskilde did not owe any concurrent common law duty of care. In the circumstances of the present case that is perhaps a somewhat surprising submission because Taberna’s pleaded case included an allegation of negligent misrepresentation based on the breach of a duty of care – although that allegation was abandoned at a late stage i.e. in Taberna’s written opening.
In any event, it does not seem to me that the question as to whether a defendant can rely upon the 1945 Act should depend on the happenstance as to whether the claimant advances a claim based on a breach of a concurrent duty of care. Rather, the question should, in my judgment, depend on whether or not there is or would be concurrent liability. That seems a more principled approach which also derives at least some support from what Lord Hoffmann stated in Standard Chartered Bank v Pakistan Shipping Corporation [2003] 1 AC 959 at p967F where he referred with apparent approval to what Sir Donald Nicholls V-C said in Gran Gelato viz that, in principle, a defence of contributory negligence should be available in a claim for damages under s2(1) of the 1967 Act.
Here, Mr Béar submitted that insofar as any representations are found to have been made to Taberna (made with the necessary intent), the circumstances giving rise to that conclusion would also be likely to generate a common law duty of care. I agree. It follows that if Roskilde is guilty of any misrepresentation giving rise to a liability under s2(1) of the 1967 Act, it is my conclusion that it would in principle be entitled to rely on a defence of contributory negligence under the 1945 Act.
However, it is important to bear in mind that it does not necessarily follow automatically that Roskilde would be entitled to a reduction in the damages recoverable. That is so for two main reasons viz:
Because of the so-called rule in Redgrave v Hurd i.e. as paraphrased by Mr Choo-Choy, it lies ill in Roskilde’s mouth to complain of contributory negligence when it intended Taberna to rely on the representation(s); and
Under the 1945 Act, the damages recoverable are reduced only “to such extent as the court thinks just and equitable having regard to the claimant’s share in the responsibility for the damage …”
Here, Mr Choo-Choy submitted that no convincing case had been made out that Taberna failed to take reasonable care of its own interests in making the purchase; alternatively if Taberna was guilty of such failure, the lion’s share of the responsibility for Taberna’s loss should nevertheless lie with Roskilde. So far as necessary, I address these points later in this judgment.
Against that background, I turn to consider the four specific sets of representations which are relied upon by Taberna in support of its claim.
NON-PERFORMING LOANS (“NPLs”)
What representation was made?
Taberna relies on the following representations:
[A] Roskilde’s statement in the IP as referred to in paragraph 50(iv) above i.e. that its “Non-Performing Loans” or NPLs as at the end of the third quarter of 2007 amounted to DKK 57 million, being approximately 0.14% of its total loans and guarantees and the (virtually) identical statement in the Q3 Report as referred to in paragraph 47 above;
[B] Roskilde’s statement at p13 of the 2007 Announcement that the “Percentage of loan assets at reduced interest rate” was 0.2% of its gross loans and guarantees as referred to in paragraph 91(vi)(b) above;
[C] Roskilde’s statement on page 10 of the Annual Results 2007 that “Non-Performing Loans” as at the end of 2007 amounted to DKK 80 million as referred to in paragraph 94 above.
For convenience, I shall refer to these representations as [A], [B] and [C].
As to these representations, Mr Frappier confirmed that he did not personally read the Annual Results 2007; and, on the evidence, I am not persuaded that these were read by Mr Bellopede or anyone else at Taberna. I deal below with the separate question of reliance but, for present purposes, this means that I am concerned only with representations [A] and [B].
Mr Béar submitted that the case advanced by Taberna on the basis of those representations must necessarily fail because (i) as a matter of law, a representation must be made with intent that it be acted on; and (ii) Taberna cannot satisfy this criterion in relation to the IP, the Q3 Report or the year-end accounts (by which I understood him to include the 2007 Announcement) because these documents were publications to shareholders and possibly others but were not aimed at the secondary market i.e. they were not “aimed at” Taberna. As to the former submission, Mr Béar relied upon a passage in Chitty on Contracts (31st Ed) Vol 1 para 6-031; this was uncontroversial. As to the latter submission, he relied in particular, on the celebrated decision of the House of Lords in Peek v Gurney (1873) CR 6 HL 377, 399-400, 410 and, in more modern times, Gross v Hillman [1970] Ch 445, 463E-H, 461F-H.
In this context and with particular reference to representation [A], Mr Béar advanced a number of specific submissions to the effect that the representations contained in the IP were not “made” to Taberna with intent that they be relied upon. In particular, with regard to the IP, he submitted that this was clearly a set of (PowerPoint) slides which Roskilde used at presentations given after the Q3 Report was released in October 2007; that there is no evidence as to how the IP reached Taberna; that it might have been downloaded from the Roskilde website or sent by a third party – there is no basis for saying which; that if the latter, then there is no basis for imputing the communication to Roskilde in the absence of a plea of agency; and that, if downloaded by Taberna (as to which there is no evidence anyway), the document could not sensibly constitute a representation to the whole world. Further, Mr Béar relied on the terms of the “Disclaimer” contained in the IP. I have already quoted the full terms above but it is convenient to identify the particular wording relied upon by Mr Béar (with added lettering for ease of reference) viz:
“This presentation has been produced by [the Bank] solely for use by investors met during the non-deal roadshow made in …”
“No representation or warranty (express or implied) is made as to, and no reliance should be placed on, any information including projections, estimates, targets and opinions contained herein”;
“no liability whatsoever is accepted as to any errors, omissions or misstatements contained herein”;
“neither the Bank nor any officers or employees accepts any liability whatsoever arising directly or indirectly from the use of this presentation for any purpose”;
“Neither this presentation nor any part of it shall form the basis of, or be relied upon in connection with any offer, or act as an inducement to enter into any contract or commitment whatsoever”;
“the Bank is under no obligation to update or revise the information contained herein …”.
As to these “points”, Mr Béar submitted: (i) point (a) has the effect of ensuring that the document cannot be treated as a representation other than to the identified category of recipient – investors who actually met Roskilde during the non-deal roadshow; (ii) points (b) and (e) are clear duty-negativing clauses; (iii) points (c) and (d) are exclusion clauses; (iv) point (f) both defines the scope of the representation being made (i.e. making clear that the representation applies to the date of the document only – October 2007) and excludes any duty which might otherwise arise to update or correct the information in the light of subsequent events or after-acquired knowledge. Further, Mr Béar submitted that it was well established that a binding exclusion clause can be included in a non-contractual notice; that the same was true of exclusion clauses in relation to misrepresentation, relying upon a passage in Cartwright on Misrepresentation at para 9-01; and that, so far as relevant, the disclaimer satisfied the requirement of reasonableness, relying upon Photo Production Ltd v Securicor Transport Ltd [1980] AC 827 per Lord Wilberforce at p843 and IFE v Goldman [2007] Ll R 264 per Toulson J at [54].
Specifically with regard to point (a), Mr Béar submitted that there were good reasons to limit the scope of a representation contained in the slides for a presentation to those individuals who attended the presentation, because a summary contained in the slides will inevitably be explained and expanded upon in the oral presentation, making the slides inherently an incomplete part of any representation actually being made. In the abstract, I agree that there is certainly some force in that submission. However, in my judgment, it ignores the particular facts of the present case. In particular, although the IP (and also the Q3 Report) were “public” documents and to that extent were subject generally to the “rule” in Peek v Gurney, it is clear that they were intended by Roskilde generally to be available for use by potential investors of the subordinated notes in the secondary market, including Taberna. As submitted by Mr Choo-Choy, both Mr Dalgas and Mr Nielsen accepted as much in cross-examination; and as a member of the class of potential investors, Taberna was part of the “target audience” for Roskilde’s publications. Further, as also submitted by Mr Choo-Choy, there were more specific inter-actions between Roskilde and Taberna that support that conclusion. In particular, as referred to in paragraph 40 above, SocGen sent a copy of the Q3 Report and directed Taberna’s attention to the IP on Roskilde’s website. Although I was not prepared to allow Taberna to plead a new case that SocGen were agents acting on behalf of Roskilde (see paragraph 41 above), Mr Dalgas accepted in evidence that he was content that SocGen should approach a “very tight” group of potential investors that might be interested in purchasing Roskilde’s subordinated note paper and that, in that connection, SocGen should feel free to provide such potential investors with copies of the IP and the Q3 Report or to point such potential investor(s) to those documents on Roskilde’s website. This conclusion is also supported by the terms of Mr Dalgas’ email dated 16 January 2008 (see paragraph 64 above). Against that background, I do not consider that Roskilde can seek to rely upon point (a) in the disclaimer in the IP.
For the avoidance of doubt and contrary to Mr Béar’s submissions, I do not consider that the effect of the foregoing is to reintroduce a general duty of care for published statements of the very kind rejected by the House of Lords in Caparo. As stated by Bridge LJ in Howard Marine v Ogden [1978] QB 574, 596E-F, liability of the representor under s2(1) of the 1967 Act does not depend on the existence of any duty of care. Rather, my conclusions as summarised above are simply to the effect that on the particular facts of the present case I am satisfied that relevant representations were made by Roskilde to Taberna through Roskilde’s published reports as referred to in above.
Nor do I consider that Roskilde is entitled to rely upon the other “points” in the disclaimer as referred to above in the particular circumstances of the present case. As to points (b) and (e), I accept that as submitted by Mr Choo-Choy the disclaimer in the IP was never part of the subordinated notes contract between Taberna and Roskilde; that accordingly it cannot create any contractual estoppel of the type considered in JP Morgan v Springwell [2010] 2 CLC 705 against Taberna; and that, in any event, a mere declaration of non-liability by the representor cannot have the effect of preventing a representor from incurring liability for misrepresentation: see IFE Fund v Goldman Sachs [2007] 1 Lloyd’s Rep 264 in particular at [65]. As to points (c) and (d), I am prepared to assume that these are to be regarded as, in effect, exclusion clauses which might be relied upon by Roskilde and (so far as relevant) that they satisfy the requirement of reasonableness. However, in my view, they are to be construed contra proferentem and, as such, the words used are insufficiently clear to exclude liability for damages for misrepresentation under s2(1) of the 1967 Act. Having regard to the particular facts of the present case, it seems to me that point (f) has little, if any, relevance; but, insofar as may be relevant, it cannot, in my view, avail Roskilde for one or more of the foregoing reasons.
In any event, at least under this head, it is noteworthy that the representation relied upon in the IP was in virtually identical form to the representation relied upon in the Q3 Report which did not include any similar disclaimer.
I turn then to consider the two main representations i.e. [A] and [B].
As to representation [A], I have already referred (in paragraph 48 above) to the fact that one of the major issues in the trial concerned the proper meaning to be attributed to the term “non-performing loans” or NPLs. This is important because it is relevant to two main points viz (i) whether the representations were true or false; and (ii) whether Roskilde had reasonable grounds to believe and did believe that the representations were true. In summary, Mr Choo-Choy’s primary submission was that the definition of an NPL is set out in the Capital Adequacy Regulation implemented into Danish law as Executive Order 10113 of 22 December 2006 Appendix 3 item 20 and 21; and that, in essence, a loan becomes non-performing if it has been in default or in an unauthorised overdraft state for an amount of more than DKK 1000 for more than 90 days. In support of that submission, Mr Choo-Choy relied, in particular, on two main points:
On its natural and ordinary meaning, the NPL term must cover loans where the borrower is failing to perform in accordance with the terms of the loan. Such a loan would, as a matter of simple language, be “non-performing” and would also give rise to a risk of default which would be expected to be material to an investor although he accepted that the NPL term cannot have been intended to cover any failure to perform or default, no matter how minor or short-lived the failure or default.
Mr Tønnesen’s evidence is that the NPL term is well known in the accounting/banking field, in Denmark as well as internationally, pointing to the description of NPL in para 4.84 of the IMF’s Compilation Guide on Financial Soundness Indicators 2004 (the “IMF Guide 2004”) suggesting a definition of NPL which covers “loans … when payments of principal and interest are past due by three months (90 days) or more …” and, in that context, observing: “The 90-day criterion is the time period that is most widely used by countries to determine whether a loan is nonperforming.”
Although it was common ground that this suggested definition of the term NPL was never formally adopted by the IMF (or any other institution), Mr Choo-Choy also relied on the evidence of Mr Frappier which was, in summary, as follows:
He understood the terms “non-performing loans” and “NPLs” to mean “[g]enerally a loan that is not meeting the terms of its credit agreements and is in default … [t]ypically … [for] 90 days” as well as “loans where you don't expect to collect the full amounts of interest and principal on the underlying obligation”;
He reiterated the same point later in his evidence by describing NPLs as “loans 90 days or more past due, or loans on which you do not expect to collect all principal and interest”;
He described the latter category as “non-accrual loans”. As he put it, “if you don't expect to get all principal and interest, the loan is impaired; and if a loan is impaired, you should not accrue interest”;
In his view “[a] non-accrual loan would be a sub-set of non-performing loans”.
In essence, the case advanced by Roskilde was that the term NPL had a much narrower meaning. In particular, the evidence of Mr Nielsen was that, by the terms “non-performing loans” and “NPLs”, he only meant to refer to loans attracting a reduced interest rate, being loans where Roskilde had made an assessment that the relevant customer would not be able to service full interest payments and the bank therefore decided to reduce (or cease) the accrual of interest (at least until the customer was again able to make payments), such assessment being made either when the customer was in default of payment or even without there being a current default if the bank nevertheless assessed that the customer would ultimately be unable to make the required payments.
In support of this narrower meaning, Mr Béar also relied upon what he described as the “totality principle” and an argument that Taberna’s case to the contrary descends into what he described as “strange syllogisms” as set out in paragraphs 121 to 123 of his written closing submissions which it is unnecessary to set out at length. I would summarise the main points advanced by Mr Béar as follows:
That the meaning of the phrase “NPL” was limited to reduced rate loans is, in effect, demonstrated by the information contained in the 2007 Announcement which identifies a figure of 0.2% against the rubric “loan assets at reduced rate” – see paragraph 91(vi)(b) above. In fact the previous page of that document contains the “disclaimer” referring the reader to the Danish version in case of any discrepancy (see paragraph 91(v) above); the same text appears in the Q3 Report. So although the Q3 Report uses the phrase “NPL” it does so, in the first such instance, when giving a figure of DKK 56,821,000 which in the Danish version translates as “rentenulstillede udlån” (literally: reset-interest loans). An English reader even if he ignored the disclaimer and translation risk would have access to the 2006 annual accounts which in the table at p33 showed a figure, again obviously likely to be the same line item, of DKK 51,350,000, recorded as “Claims at reduced interest rate end of year”.
The authoritative Danish version of Roskilde’s audited accounts used no terminology equivalent to “NPL”.
In Denmark, there is no accepted use of the terminology “NPL”/non-performing loan. There is no equivalent language. A remarkable feature is that no-one has suggested any equivalent language. Even on Mr Tønnesen’s arguments, NPLs must include defaulted loans (misligholdte fordringer) as well as reduced-interest loans (rentenulstillede lån or similar language) – and no combination or “portmanteau” term to cover even these two categories, let alone to refine them further by reference to time and amount, has ever been put forward.
In Europe at large, there is no accepted definition. This is entirely clear from the Reuters’ report of 2013 on proposed EU standardisation. That report states that there are “a host” of definitions.
Moody’s own report, on which Taberna relied heavily (at least to the extent of following their positive view at the time), did not give the figure in question as being one for “non-performing loans” but rather for “problem loans” which were defined as the “sum of doubtful and non-performing loans”, neither of which terms were then defined.
In my view, simply as a matter of language, the phrase “non-performing loan” or NPL does not have the narrow meaning suggested by Mr Nielsen or submitted by Mr Béar. In any event, as submitted by Mr Choo-Choy and on the assumption that the phrase is capable of different meanings, the correct approach in the context of a claim under s2(1) of the 1967 Act (as opposed to a claim in deceit) is as described by Christopher Clarke J (as he then was) in Kingspan Environmental Limited v. Borealis A/S, Borealis UK Limited [2012] EWHC 1147 (Comm) at [423]-[424]:
“423. The representee must show that the representor intended him to act on the statement: Banque Keyser Ullmann SA v Skandia (UK) Insurance [1990] 1 QB 665, 790. That was always the rule in deceit; and the latter case indicates that the same applies in a claim under section 2 of the Misrepresentation Act.
424. If a statement has more than one meaning, the question is whether or not it was understood by the representee in the meaning which the court ascribes to it — which is the meaning which would be attributed to it by a reasonable person in the position of the representee — and that having that understanding he relied on it. Arkwright v Newbold (1881) 17 Ch D 301; Smith v Chadwick (1884) 9 App.Cas 187. But for a claim in deceit it would be necessary to establish that the representor intended the representee to understand the representation in the sense in which he did or was willing that he should do so: see Goose v Wilson, Sandford & Co [2001] 1 Lloyd's Rep P.N. 189 paras 41,42.” [emphasis added]
On this basis and as submitted by Mr Choo-Choy, since the Court is now only concerned with a claim under s2(1) rather than in deceit, it is not necessary for Taberna to prove that Roskilde specifically intended it to understand the terms “non-performing loans” and “NPLs” in the sense in which Mr Frappier says that he understood them; it is sufficient that the meaning that Mr Frappier attributed to those expressions was one that he was reasonably entitled to adopt as an addressee of the representation. It is his understanding that is critical in this connection because he represents the Credit Committee that ultimately approved the investment into the Subordinated Notes. Here, I am satisfied that Taberna (through Mr Frappier) understood the phrase “non-performing loans” and NPLs not in the suggested narrow meaning stated above, but as stated by Mr Frappier in his evidence as summarised in paragraph 124 above; and, subject to consideration of Mr Béar’s “totality principle”, that was a meaning which would be attributed to such phrase by a reasonable person in the position of Taberna.
In reaching this conclusion and for the avoidance of doubt, I should make plain that I have considered the expert evidence of Mr Krogh. However, he did not take issue with Mr Tønnesen’s evidence on this issue beyond stating (as is common ground) that the NPL term is not a statutorily defined term under Danish legislation. Moreover, as emphasised by Mr Choo-Choy, he acknowledged that “non-accrual loans and reduced-accrual loans – where the reduction is due to the customer’s inability to pay – would, in my opinion, be a subset of “non-performing loans” [and] [m]oreover, overdrawn accounts and loans in arrears will typically be a subset of “non-performing loans” and that “it must be assumed that there is a certain overlap between “non-performing loans” and defaulted loans”. In cross-examination, Mr Krogh confirmed that his reading of the NPL term as comprising non-accrual and reduced-accrual loans and loans that were in arrears for a certain period (he did not define for how long) was not based on any international banking experience (he claimed no such experience), but on a logical understanding of the expression “non-performing loans”. His experience was primarily in the Danish banking market where he had not used the NPL term, but he had heard of it and assumed that its meaning was as described.
As to Mr Béar’s “totality principle” (which I have summarised in paragraph 126 above), I fully accept that such principle is important and potentially relevant both generally and in the circumstances of the present case. However, in my judgment, the points raised by Mr Béar as summarised above are flawed for a number of reasons.
First, as to representation [A] as contained in the English version of the Q3 Report (and putting aside the representation relied upon in the IP), it is true that (i) the Q3 Report contained the “disclaimer” which I have already quoted in paragraph 42 above; and (ii) the Danish version of the Q3 Report used the term “Rentenulstillede udlån” which, translated literally, means in English “interest reset” or “loans with reset interest” rather than “non-performing loans” or NPLs. However, the “disclaimer” does not, in my view, assist Roskilde for one or more of the reasons set out in paragraph 14.9 of Mr Choo-Choy’s written closing submissions. In summary:
I accept that the purpose of the disclaimer is to ensure that in the case of inadvertent discrepancies in translation, the Danish version should prevail. In the case of use of the NPL term, however, there is no question of any inadvertent use in the English versions of the relevant reports (including the English version of the Q3 Report) of the NPL term. As Mr Nielsen testified there was a conscious and deliberate use by Roskilde of the NPL term in the English language in consultation with Moody’s. In that context, it seems to me that the language disclaimer is irrelevant.
In any event, the disclaimer does not have the effect of negating the statements or representations made in or by the English versions of the relevant documents (which are distinct documents in their own right).
Nor does the disclaimer have any contractual effect as between Roskilde and Taberna. The only effect of it having been ignored and of the Danish version(s) of the relevant report(s) not having been checked by Taberna because Taberna did not have or engage any Danish speaking advisers on its due diligence team is that such omission may support an allegation of contributory negligence on Taberna’s part. But the disclaimer cannot contractually or legally preclude Taberna from relying on the statements contained in the English versions of the relevant reports and presentations: cf Toulson J’s observations in IFE Fund v. Goldman Sachs [2006] EWHC 2887 (QB) (Comm) at [65]: “A mere declaration by a misrepresentor that he does not accept liability for the consequences which the law attaches to a misrepresentation would be ineffectual”.
Second, I am not persuaded that Roskilde is able to derive any assistance from Mr Béar’s argument as summarised above based upon a comparison of the language in the Q3 Report with that contained in the previous 2006 annual accounts or the 2007 Announcement. As to the previous 2006 annual accounts, it is right that different language was used in those accounts. However, it would, in my view, have required extraordinary mental gymnastics for anyone reading the 2007 Q3 Report to conclude from a reading of the 2006 annual accounts that the reference in the former to “non-performing loans” or NPLs should, in effect, be read to mean the same as what Mr Béar said was the “same line item” in the latter.
As to the language in the 2007 Announcement, Mr Frappier’s evidence was in summary that he interpreted the reference to “Percentage of loan assets at reduced interest rate” on p13 of the 2007 Announcement to be a reference to NPLs; that he assumed that that expression was simply the way in which Roskilde had translated the reference to “non-performing loans” or NPLs in its 2007 Announcement i.e. he assumed that it was “just a translation issue”; and that he therefore regarded the stated figure of 0.2% as, in effect, being the figure for NPLs (as he understood that term) at the end of 2007. The main thrust of Mr Béar’s submission was, in effect, that Mr Frappier could not have understood the figure of DKK 80 million or 0.2% to be a reference to NPLs if that figure covered both loans that had been in default for at least 90 days and what Mr Frappier described as “impaired loans”. However, having heard Mr Frappier give evidence and bearing fully in mind the reservations referred to earlier, I accept his evidence as to his subjective understanding. Moreover, as submitted by Mr Choo-Choy, most of Mr Frappier’s cross-examination on this aspect ignored what he had explained as to his understanding of the NPL term. In particular, he explained that, in addition to loans that had been delinquent for at least 90 days, NPLs included “loans on which you do not expect to collect all principal and interest”, what he described as “non-accrual loans”; and he continued “if you don't expect to get all principal and interest, the loan is impaired; and if a loan is impaired, you should not accrue interest”. Thus, in that sense, Mr Frappier’s notion of “impaired loans” was essentially the same as Roskilde’s version of “non-accrual loans” or “reduced accrual loans” (as explained by Mr Nielsen). Yet, it is not suggested by Roskilde that there was internal inconsistency between its own non-accrual / reduced-accrual loan figure of DKK 80 million and the accumulated write-down figures published as part of its full 2007 year end results. In any event, as again submitted by Mr Choo-Choy, the premise of Mr Béar’s argument in this connection is that Mr Frappier should have been studying every item in the 2007 Announcement as if he were a forensic accountant aiming to find inconsistencies and ambiguities, rather than taking Roskilde’s representations as to NPLs at face value; and the highest that Roskilde might be able to put it is that what Mr Frappier might have been able to discover if he had acted as a forensic accountant might be relevant as a matter of contributory negligence, but it does not go to the objective meaning of the NPL term.
Third, I am also not persuaded that Roskilde is able to derive any assistance in this context from anything contained in the Annual Results 2007. That is simply because, as I have found, this document was never read by Mr Bellopede or anyone else at Taberna. In any event, for the detailed reasons set out in paragraph 14.6 of Mr Choo-Choy’s written closing submissions (which it is unnecessary to set out), I do not consider that anything stated in the Annual Results 2007 would have indicated to a reasonable reader that the NPL term was being used by Roskilde in the narrow sense of “loans at reduced interest rate”.
So far, I have focussed on what I have referred to as representation [A]. In so doing, I have also considered the effect of the wording which constitutes representation [B] in the context of Mr Béar’s so-called “totality principle”. However, it is important to bear in mind that representation [B] is relied upon by Mr Choo-Choy as a discrete positive representation by Roskilde and it is therefore necessary to consider the alleged meaning of that representation. In summary, Mr Choo-Choy submitted that representation [B] should be regarded as having the meaning which Mr Frappier said in evidence he attributed to it as summarised above. As I have already stated, I accept that that was Mr Frappier’s subjective understanding. However, in deciding the proper meaning of representation [B] it seems to me that it is at least capable of having more than one meaning and, that being the case, the question is (as stated by Christopher Clarke J in paragraph 424 of Kingspans) what is the meaning which would be attributed to it by a reasonable person in the position of the representee. In my judgment, the answer to that question is that having regard to the language used and viewed in isolation, a reasonable person in the position of Taberna would not have understood representation [B] in the way in which (as I have found) Mr Frappier subjectively understood it. Thus, it is my conclusion that Taberna’s case under s2(1) of the 1967 Act based on representation [B] must fail. However, for the reasons already stated, I accept Mr Choo-Choy’s submission as to the meaning of representation [A] i.e. as referring to the totality of Roskilde’s NPLs including its loans in default for 90 days or more; and that it was made by Roskilde to Taberna with the intention that it be relied upon by Taberna.
Reliance
As to the law, I have already set out the applicable principles. As to the facts, there is overwhelming evidence of Taberna’s reliance on Roskilde’s statements about its NPLs. In particular, as summarised by Mr Choo-Choy in his written submissions:
From the outset of the due diligence exercise by Taberna, the low level of Roskilde’s NPLs (and consequently the high coverage of loan loss reserves to NPLs ratio) as at the end of the 3rd quarter of 2007, as represented in its Q3 Report 2007 and IP, was noted by Mr Bellopede – see for example:
his reference to “NPL/gross loans and guar 0.14%” in his email dated 25 October 2007 to Messrs Schneider and Ramamon – which was written in the context of the SocGen approach;
his reference to Roskilde having a “good rating [from] Moody’s” and “NPL 0.1%” in his email dated 7 November 2007 to Mr Schneider; and
his further reference to Roskilde’s “high quality of loan portfolio (high percentage of coverage) … (… their unsecured lending is mitigated by a very low NPL)” in his email dated 12 November 2007 to Mr Schneider.
The low level of Roskilde’s stated NPLs was thereafter seen throughout as a “key strength” of an investment: see e.g. the recording of this key strength by Mr Bellopede in the very first draft of the Credit Report dated 10 January 2008 and in the latest version of the Credit Report circulated to the Credit Committee on 5 February 2008 shortly before the Credit Committee’s decision to approve the investment on 8 and 9 February 2008.
Mr Frappier’s direct evidence at para 18 of his witness statement (which I accept) was that the information provided by Roskilde as to its NPLs was “very important” (first sentence), that “[t]he NPL figures were obviously crucial to the assessment by both the London team and the Credit Committee” (penultimate sentence), and that “[if] we had … been told that, instead of being of the order of a few tens of millions of Danish Kroner, the NPLs amounted to many hundreds of millions or even billions of Danish Kroner, we would have had a much more negative view of the Roskilde investment and would have been most unlikely to approve it” (last sentence). Yet, no case was actually put to Mr Frappier during cross-examination that he or other members of the Credit Committee did not in fact rely on the statements that Roskilde had made and which had been incorporated into the Credit Report with regard to NPLs. In addition, Mr Frappier’s oral evidence (to which I have already referred) was that he had noted and taken account of the increase in what he understood was the NPLs from 0.14% to 0.20%.
In the light of this evidence (which I accept), I am satisfied that Taberna did rely on representation [A] as contained in the Q3 Report and the IP in entering into the purchase of the subordinate notes. For the avoidance of doubt, it is my conclusion that Taberna also relied on representation [B] but for the reasons already stated, I do not consider that representation [B] is relevant as a matter of law.
Falsity
Under this head and in light of my conclusions as stated above, two main questions arise:
Was representation [A] false?
If so, did Roskilde have reasonable ground to believe and did believe up to 14 February 2008 that representation [A] was true?
On the basis of the meaning of representation [A] as I have held, it is my conclusion that representation [A] was false. Indeed, Mr Béar did not (at least initially) contend otherwise. I was initially troubled by the fact that Mr Choo-Choy was unable to identify from Roskilde’s disclosure the precise total size of NPLs (in the sense I have held) when representation [A] was originally made i.e. at the end of October 2007. The figure at the end of 2007 was approximately DKK 3.5bn. However, on a balance of probabilities, it seems to me that even if the figure was less than DKK 3.5bn at the end of October 2007, the overwhelming likelihood is that the total size of NPLs (in the sense I have held) must have been vastly in excess of the figure of DKK 57m as stated in representation [A] and that representation [A] was therefore false when it was made and at all material times thereafter. Again, I did not understand Mr Béar to suggest otherwise. Thus it is my conclusion that representation [A] was false.
Equally it is my conclusion that if representation [A] had the meaning as I have held, then Roskilde has failed to prove that it had reasonable ground for believing and did believe that such representation was true. Indeed, it is my conclusion that Roskilde did not have reasonable ground for believing and did not believe that such representation (as understood by Taberna) was true. (For the avoidance of doubt, my conclusion would be the same in respect of representation [B].) Indeed, I did not understand Mr Béar to suggest otherwise. In support of such conclusions, Mr Choo-Choy further relied on the events in late January/early February 2008 which I have already summarised above in relation to the preparation by Roskilde of the VO document. In the event, it is unnecessary to determine whether such exercise complied with the requirements of Order 10113. For present purposes it is sufficient to say that I accept that Roskilde’s calculations during that period are inconsistent with any reasonable ground for believing or any actual belief that representation [A] (or [B]) was true in the sense (as I have held) such representations were understood by Taberna.
For these reasons, it is my conclusion that by virtue of misrepresentation [A] as contained in the Q3 Report, Roskilde is, in principle, liable for damages for misrepresentation under s2(1) of the 1967 Act subject to consideration of the question of contributory negligence which I propose to consider at the end of this Judgment after considering the other alleged misrepresentations relied upon by Taberna to which I now turn.
CREDIT POLICY
Under this head, the alleged misrepresentations relied upon by Taberna were as follows:
The statement on p7 of the IP that Roskilde had a “[c]onservative risk policy” and a “[s]tructured hierarchy of approval authority” and at p4 that it would “maintain [its] conservative risk profile”. In this context, Mr Choo-Choy also sought to rely upon what Mr Dalgas accepted in evidence viz. that the references to conservative risk policy and to a structured hierarchy of approval authority were direct references to the more detailed statement of Roskilde’s credit policy in the Offering Circular relating to the €2 billion EMTN programme as described below.
The statements on p53 of the Offering Circular (as forwarded by Mr Dalgas to Deutsche Bank for passing on to Taberna in answer to Taberna’s questions on 16 January 2008) that Roskilde had:
“a conservative credit policy”;
“a structured hierarchy of approval authority, whereby the local branch manager can approve loans up to a certain limit. Larger loans are presented to the Credit Department. Loans exceeding their limit to the Credit Committee, which consists of a member of the Management and the head and the deputy head of Credit. Loan requests exceeding the authority of the Management are presented to the Board on a weekly basis. Thus, there are four levels in the approval hierarchy”;
a rigorous system of monitoring whereby “[t]he loan portfolio is monitored daily by the credit department, which refers directly to the Management. The loan portfolio is subject to a provision process each quarter, where existing loan loss provision[s] are monitored and followed up on, and new risk of losses are provided for. This is approved by both Management and the Board. The Management has regular meetings with each of the branch managers and the credit department, where important figures of the branch are checked, such as all major loan facilities, loan loss provisions, overdrafts, risk profile in terms of retail lending, corporate lending and sector lending. In addition to that the credit department visits each branch at least once a year to control a number of randomly selected accounts. At the monthly board meeting, the Management presents all facilities granted by the Board since the last board meeting. Furthermore, there is a presentation on a sector basis, whereby the Bank’s exposure to a given sector is discussed. This includes an overview of the 10-20 largest individual exposures to the sector, a risk/profit analysis and the credit department’s opinion on the expected development within the sector. It is ensured that all sectors where the Bank has exposure are presented at least once a year, or at the request of the Board.”
Thus, in summary, it was Taberna’s case that (i) by the above statements, Roskilde represented that it had until then implemented and operated conservative risk approval and monitoring policies and procedures as described in the above statements; and (ii) these statements and associated representations were made by Roskilde to Taberna as follows:
Taberna’s attention was specifically directed to the IP by SocGen in the context of SocGen’s approach to potential investors in Roskilde’s subordinated note paper following discussions with Roskilde;
Roskilde placed the IP on its website specifically so that it would be accessible to potential investors such as Taberna – it did not seek to restrict access to the presentation in any way; and
A copy of the Offering Circular was specifically provided by Mr Dalgas to Deutsche Bank for passing on to Taberna on 16 January 2008 in answer to various questions raised by Taberna, including questions relating to Roskilde’s EMTN programme.
As to the representations contained in the IP, it is my conclusion (for the reasons already stated above and which I do not propose to repeat) that, contrary to Mr Béar’s submissions, they were made and intended by Roskilde to be relied upon by potential investors in the subordinated notes, including Taberna. Similarly, for the reasons already stated above, I do not consider that Roskilde is entitled to rely on the disclaimer in the IP.
As to the representations contained in the Offering Circular, Mr Béar submitted that that document was obtained by Taberna, and the information in it extracted, before any contact with Roskilde, presumably from Deutsche Bank. That would seem correct. However, as already noted above, Mr Dalgas sent the Offering Circular to Deutsche Bank for passing on to Taberna on 16 January 2008. Notwithstanding, Mr Béar submitted that this does not convert all the contents of the Offering Circular into a representation. In particular, Mr Béar submitted that the critical point is that Mr Dalgas, having been sent Mr Bellopede’s questions (which appear higher up on the page), explicitly sent the Offering Circular and other documents in answer to those questions; that the questions on their face do not concern credit policy (or any of the other matters complained of by Taberna); that this was admitted by Taberna’s witnesses; and that it therefore follows that the representation Mr Dalgas made was only that the contents of the Offering Circular provided answers to (some of) the questions actually asked. Thus Mr Béar submitted that although this no doubt involved a representation that the parts of the Circular relevant to those questions were correct, it does not involve any representation concerning unrelated parts of the Offering Circular.
I do not accept that submission in particular because, as submitted by Mr Choo-Choy, Mr Dalgas accepted in evidence that he provided the entirety of the Offering Circular fully appreciating that a potential investor in Taberna’s position would naturally want to take account of the information contained in the section entitled “DESCRIPTION OF THE ISSUER” including key information about the historical performance, business operations, credit policy and management of Roskilde (including a description of its credit policy). For these reasons, I am satisfied that the relevant representations were made by Roskilde to Taberna with the relevant “intent”.
In this context, it is also convenient to consider Mr Béar’s further submissions with regard to the “notice” in the Offering Circular which stated in material part (with added lettering for ease of reference):
“Neither this Offering Circular nor any other information supplied [in connection with the Notes] shall form the basis of any credit or other evaluation…”;
“Neither the delivery of this Offering Circular nor the delivery, sale or offering of any Notes shall in any circumstances imply that the information contained herein concerning the Issuer is correct at any time subsequent to the date hereof …”.
As to this wording, Mr Béar submitted that point (a) was a clear duty-negativing clause; and that point (b) both defined the scope of the representation being made (i.e. making clear that the representation applies to the date of the document only – 16 March 2007) and excluded any duty which might otherwise arise to update or correct the information in the light of subsequent events or after-acquired knowledge. Given the other terms of the Offering Circular which I have already quoted above (including the broad statement to the effect that Roskilde accepted responsibility for the information contained in the Offering Circular), I do not accept Mr Béar’s submission with regard to point (a). However, I do accept Mr Béar’s submission with regard to point (b) – although given my conclusions with regard to the IP and my further conclusions below, it does not seem to me that this point is ultimately of much, if any, significance in the particular circumstances of the present case.
Mr Béar submitted that the first task was to analyse the true effect and meaning of what Roskilde said; and that in that context, two particular issues arose viz (i) does the representation have a sufficiently definite content; and (ii) what was the effect of the totality of Roskilde’s published communications concerning this topic?
As to point (i), Mr Béar submitted in summary that a representation must have some definite content i.e. a representation is not actionable beyond the parameters of its indisputable meaning; that a subjective term such as “conservative” is open to many shades of interpretation; that no expert evidence has been led (or indeed any other evidence) to suggest that “conservative” has a known meaning; that no evidence has been led as to the typical practices of Danish banks up to 2007 so as to provide a benchmark; that the statement that the policy was “conservative” was no more than Roskilde’s comment on or summary of the other features which were referred to – namely (i) the “long record of low credit losses”; and (ii) the “structured hierarchy of approval authority” and monitoring processes; that even if that were wrong, the burden was squarely on Taberna to show a meaning which any reasonable bank in Roskilde’s position would have to agree was within the scope of “a conservative credit policy”; and that this Taberna have not only failed, but not even attempted, to do.
As to point (ii), Mr Béar submitted in summary that it was not permissible to look solely at the IP and Offering Circular; that the correct principle in assessing what representation was made is to have regard to the totality of publications; that, as Taberna knew, the Bank was about to publish its annual report, something which by its nature was inherently liable to supersede earlier information; that in circumstances where Roskilde had announced substantial write-downs in the 14 January Announcement and had also announced that it had a new “Strategy 2010” involving “a whole new focus” on “risk management” and loan portfolio diversification which involved reducing real estate lending from 42% to 25%, Roskilde could not reasonably be understood as continuing to believe (and consequently to represent that it believed) that its previous credit policy remained conservative in the light of the substantial changes in market conditions; that the lending policy which had generated low levels of losses during positive market conditions and was, therefore, properly and reasonably to be viewed as “conservative” during 2007, could no longer be seen in that light in the new market conditions and indeed required modification; and that is exactly what Roskilde announced.
In my view, these submissions roll together a number of different points. First, I agree that the various elements of the representations identified under this head are somewhat vague. For example, the statement that Roskilde had a “conservative risk policy” is, I accept, open to many shades of interpretation. On this basis, I also accept that there is, at least, a strong argument that this set of representations are, in whole or in part, no more than a “puff” or the kind of vague statement which is not properly regarded as a representation under the 1967 Act. However, for present purposes, I am prepared to assume in Mr Choo-Choy’s favour that such argument is wrong and that the statements identified under this head are properly to be regarded as representations. Notwithstanding, as submitted by Mr Béar, it is necessary and important to consider the representations in their proper context. This is relevant not only in considering whether a representation has been made but also whether Taberna can establish “causation” i.e. reliance. In that context, Mr Choo-Choy relied upon a number of matters, in particular (i) the fact that, as already noted, the relevant representations were, in effect, reproduced in Taberna’s internal credit reports for use by its Credit Committee; (ii) Mr Frappier’s evidence that such information gave Taberna the “comfort” that everything Roskilde was doing was much better than in the past and that that was what Taberna needed to hear even though real estate was not doing well; and (iii) as Mr Dalgas accepted during his cross-examination, Roskilde appreciated that the nature of its credit policy would be important to potential investors because it was part and parcel of the overall picture as to the bank’s loan portfolio quality.
In considering these matters, I bear well in mind the relevant legal principles which I have already set out above. In particular, at the risk of repetition, it is important to emphasise that I fully accept that, as a matter of law, a misrepresentation need not be the sole cause which induces the misrepresentee to make the contract; that it is sufficient that the misrepresentation can be shown to be one of the inducing causes; and that the misrepresentee would not have entered the contract but for the misrepresentation: see Chitty on Contracts (31st Ed), Vol 1 para 6-037, 6-038. However, I am unpersuaded that this set of representations was one of the inducing causes of Taberna’s decision to purchase the subordinated notes; or, in other words, that Taberna would not have entered the contract but for one or more of these representations. As to the points relied upon by Mr Choo-Choy, it seems to me that neither the inclusion of these representations in the credit report nor the evidence of Mr Dalgas is necessarily determinative. As to the evidence of Mr Frappier on this point as summarised above, I accept that it is potentially relevant. However, it is, at best, somewhat tenuous; and I have to say that it was, in my view, one of the least satisfactory parts of Mr Frappier’s evidence which I am unable to accept. In my judgment, it is difficult, if not impossible, to believe that any hard-nosed sophisticated investor like Mr Frappier might purchase subordinated notes of this type at this price relying even in part on the vague representations identified under this head – particularly having regard to Mr Béar’s important “totality” point which I fully accept and recognise. For these reasons, I am unpersuaded that Taberna has established, on a balance of probability, that it relied upon this set of representations. Given this conclusion, it is unnecessary to consider the question of falsity and related aspects. But, so far as may be relevant, my observations and conclusions are as follows.
Falsity
I should mention that in the course of his oral opening, Mr Béar appeared to concede (at least initially) that he did not seek to suggest that it would have been correct to say in January 2008 that Roskilde had had a conservative credit policy in the past; and that Roskilde had no such belief. Be that as it may, I did not understand Mr Béar to make any formal concession; and in closing, I understood him to maintain that none of the representations identified under this head was false. It is therefore necessary to consider the case advanced by Mr Choo-Choy with regard to the alleged falsity of this set of representations which was, in essence, that the stated representations painted a rosy picture of the bank’s credit policy which had no basis in reality; and that they did not provide a fair and balanced picture of Roskilde’s credit policy in the relevant period. In support of that submission, he relied upon three main points.
First, he submitted that there was a long history of the ex post facto approval of large exposures (including changes to the terms of large exposures). Such grants or changes required the approval of the Board of Directors of Roskilde but were frequently approved as “urgent” or “pressing” cases without the knowledge of the Board of Directors and only subsequently submitted (at the next board meeting) for ratification by the Board of Directors, after the loan had been advanced or its terms changed. The details of what Mr Choo Choy submitted were a “practice” were set out at some length in paragraph 51 of Mr Choo-Choy’s closing submissions. For present purposes, it is sufficient to say that I accept that the Danish FSA made various criticisms in this regard from time to time, in particular in a number of letters dated 21 April 2005, 3 January 2007, 7 February 2008; and that the letter dated 22 January 2008 from Roskilde’s Credit Committee to branch managers implicitly acknowledged the undesirable nature of such practice and made clear that “after-authorisation situations will be ended … ‘after-authorisations’ will be a thing of the past.” However, it is impossible to say on the evidence before me how widespread this alleged “practice” was; or whether the so-called after-authorisations failed to comply with other objective standards of lending.
Second, Mr Choo-Choy drew attention to several reports which referred to what he submitted were “significant deficiencies in Roskilde’s credit policy over a long period” as summarised in paragraph 52 of Mr Choo-Choy’s closing submissions including the letters from the Danish FSA dated 21 April 2005, 3 January 2007 and 7 February 2008; the Internal Audit Reports dated 11 May 2007, 9 July 2007 and 8 February 2008; and the Risk Report dated 28 February 2008. I accept that these reports (which it is unnecessary to quote at length) provide some evidence of “violations” of credit policy and procedure and inadequate credit monitoring from time to time both prior to the date of the Offering Circular (so far as representations in that document are concerned) and after. However, again, I find it impossible to say on the evidence before me how significant such “violations” were – particularly since, as Mr Choo-Choy fairly accepted, there were occasionally positive reports as well within the Credit Department.
Third, Mr Choo-Choy relied on evidence to the effect that the proportion of Roskilde’s exposures which were marked as “poor” or worse within Roskilde’s internal rating system was relatively high. In particular, he relied on the Risk Report dated 28 January 2008 which showed that 6% of private customers and 23% of business customers (i.e. a total of 29% of customers) had a “poor” rating or worse; and that 17% of the grants were for weak commitments in December 2007 and 21% were grants of weak commitments in January 2008. In addition, he relied on a letter from the Danish FSA dated 7 February 2008, to the effect that Roskilde had “a bias towards business with debtors with a low rating i.e. with a high risk profile”. However, in my judgment, this alleged weak risk profile of debtors as it existed in late 2007/early 2008 is not, of itself, necessarily proof of an earlier failure or weakness in Roskilde’s credit policy at the time of lending.
For these reasons, I am not persuaded on the evidence before me that this set of representations was in whole or in part materially incorrect. It also follows that Taberna’s claim under this head must fail.
WRITE-DOWNS
Under this head, the representations relied upon by Taberna were as follows:
The statement(s) in the 14 January Announcement that (i) following an ordinary review of its accounts during the fourth quarter of 2007, Roskilde had decided to increase its loan write-downs for 2007 to a total of DKK 267 million as provision against potential, future losses, being 0.6% of its total loans and guarantees; and (ii) its forecast of loan write-downs for 2008 was DKK 150 million.
Similar statements were contained in the 2007 Announcement viz:
“In 2007, the bank increased write-downs on loans by DKK 266 million equal to 0.6% of the bank’s loans and guarantees” (p8); and
“Due to the continued uncertain economy growth prospects, the requirement for write-downs is expected to be DKK 150 million in 2008” (p9).
By these statements, it was Taberna’s case that Roskilde impliedly represented to Taberna that (i) it had reasonable grounds for believing that a write-down of no more than DKK 267 million was appropriate for 2007; and (ii) it had reasonable grounds for its forecast write-down of DKK 150 million in 2008. Further, Mr Choo-Choy submitted that the fact that Roskilde proceeded to give notice of those write-down figures by way of a special announcement to the Danish Stock Exchange and following “the ordinary review of the bank’s accounts” would have tended to reinforce the notion that Roskilde had a reasonable basis for the write-down figures that it was announcing.
In the light of my earlier conclusions with regard to the 14 January Announcement and the 2007 Announcement, it is my conclusion that these representations were similarly made by Roskilde to Taberna with the requisite “intent”. Further, I am also satisfied that these representations were relied upon by Taberna. Quite apart from the evidence of Mr Frappier (which, in this regard, I accept), that conclusion is also supported by the contemporaneous documents – in particular, the inclusion of these representations in the credit reports for use by the Credit Committee and also Mr Bellopede’s email dated 8 February 2008 which was forwarded by Mr Mitrikov to Mr Frappier later that same day.
Falsity
In considering the question of falsity and at the risk of repetition, it is important to emphasise the nature of the case advanced on behalf of Taberna in relation to these representations. In particular, Mr Choo-Choy did not dispute that the figures stated were indeed the write-down provision in fact made for 2007 and forecast for 2008 respectively. Rather, his case was that such stated figures necessarily involved the implied representations as set out above. This was disputed by Mr Béar.
In principle, I accept that certain statements may carry an implied representation: see, for example, Chitty on Contracts, Vol 1, paras 6-013 to 6-015 in particular at para 6-014. As stated by Toulson J in IFE Fund SA v Goldman Sachs International [2007] 1 Lloyd’s Rep 264 at [50], the court has to consider “… what a reasonable person would have inferred was being implicitly represented by the representor’s words and conduct in their context”. I agree generally with that “test” although I would myself emphasise that a degree of caution is necessary in considering whether or not a particular statement carries an implied representation. Even so, applying that test to the circumstances of the present case, it seems to me that a reasonable person in the position of Taberna would indeed have inferred that Roskilde had reasonable grounds for believing that a write-down of no more than DKK 267 million was appropriate for 2007; and that it had reasonable grounds for its forecast write-down of DKK 150 million. Thus, I reject Mr Béar’s submission to the contrary.
On this basis, Mr Choo-Choy submitted that there are substantial grounds for doubting the reasonableness of Roskilde’s write-downs for 2007, as well as its write-down estimate for 2008. In particular,
As to the write-down figure of DKK 267 million for 2007, Mr Choo-Choy relied upon what he said were two “key points”. First, he relied upon what he described as “inadequate procedures”. In particular, he submitted that it was clear from the experts’ evidence (para 1.3 of the Joint Memorandum and Section 52 of the Executive Order 1466 of 13 December 2006) that the exercise of determining whether write-downs were necessary required (i) consideration of whether there was Objective Evidence of Impairment (OEI), and (ii) if so whether such OEI was expected to impact on the future expected payments under the loan; that as Mr Krogh accepted in evidence, it logically follows that, in order for a bank to be able to make a reasonable assessment of the need for write-downs, it should have adequate procedures in place for the identification of OEI as well as adequate procedures for the calculation or estimation of expected future payments in respect of loans suffering from OEI; and that unless a bank has adequate procedures in both areas, it will not have a reasonable basis for properly calculating the relevant write-down. However, he submitted that by or before 8 February 2008, it was known to Roskilde that (i) the Danish FSA (in its letter dated 7 February 2008) continued to be concerned about Roskilde’s write-down assessment procedures and had expressed the view that Roskilde was “deficient in failing to introduce control procedures to ensure that there is proper identification of situations where there is objective evidence of a need for a write-down (OIV)” and in failing to adopt strategies to enable it to calculate the present value of expected future payments from loans suffering OEI; and (ii) as appears in the Internal Auditors’ Report dated 8 February 2008, its internal auditors had found “several instances” where there were no established cash flows for exposures which were OEI-marked, and were also of the view that “closer monitoring of significant exposures” was necessary. Second, Mr Choo-Choy submitted that Roskilde must have known of (i) its poor credit culture as summarised in the Risk Report of 29 January 2008; (ii) the fact that it had not implemented a conservative credit policy as referred to above; (iii) the fact that it had a substantial proportion of weak and very weak accounts (amounting to nearly 30% of its total loan portfolio); and (iv) the real likelihood of a worsening of the real estate market in 2008.
As to the write-down estimate of DKK 150 million for 2008, Mr Choo-Choy relied on what he said was Mr Andersen’s own evidence to the effect that this estimate was calculated at an earlier stage in 2007 prior to the fall in the housing market gathering momentum during the second half of 2007 but that this estimate was never updated in the light of those market developments. Further, Mr Choo-Choy submitted that Mr Andersen’s evidence in this respect is borne out by the course of discussions between the Board of Directors and Mr Christensen concerning write-downs on 11 January 2008; that the DKK 150 million write-down figure for 2008 was adopted at that meeting, but without any clear basis or rationale for it; and that (as appears from the Danish FSA’ letter dated 7 February 2008 and as again confirmed by Mr Andersen) the Danish FSA did not consider the estimated write-down of DKK 150 million for 2008.
As to these submissions, Mr Béar submitted that it is incumbent on a party alleging negligence against a defendant acting in a professional capacity (which would include the management of a bank) to lead expert evidence to establish precisely what the departure from acceptable standards was: see Sansom v Metcalfe [1998] PNLR 542, CA, approved by the Privy Council in Caribbean Steel v PwC [2013] 4 All ER 338; that only in exceptional cases is it open to a judge to find professional negligence proved “as a matter of common sense”, without the benefit of expert evidence; that, in practical terms, a plea that there was a lack of reasonable basis for an estimate is a plea of negligence and, therefore, this principle applies; that here Taberna led no such evidence save in one narrow respect; and that therefore there is no scope for a finding that Roskilde did not have a reasonable basis for the stated figures. Without deciding the point, I am prepared to assume that the stated principle applies here; but, I do not accept that the mere fact that the Taberna has led no significant expert evidence on this topic is necessarily fatal. Rather, it seems to me a factor which I can and should properly take into account.
Here, I agree that there are at least some doubts as to whether Roskilde had reasonable grounds for believing that these were appropriate figures. However, the burden lies on Taberna to prove that this was indeed the case; and, whatever doubts I may have, I remain unpersuaded on a balance of probability that this was indeed the case for the following main reasons.
First, it is important to bear in mind the nature of a “write-down” as opposed to a “write-off”. As to this, there was some dispute – and also expert evidence with regard to Danish law/practice – in particular with the terminology in the English and Danish versions of the 2007 Annual Report. I do not consider that it is necessary to engage in that dispute. For present purposes, it is sufficient to say that I accept Mr Krogh’s evidence (which was consistent with the unchallenged evidence of Mr Nielsen) that the difference between the two is that a “write-down” involves an offsetting provision in the accounts against a likely loss, whereas a “write-off” involves the loan actually being removed from the accounts because non-payment is considered inevitable.
Second, it was common ground between the experts that as a matter of Danish law/practice at the relevant time, OEI had to be identified in order for a bank to be allowed and required to recognise any write-down. As Roskilde’s witnesses explained, the rules on write-downs in Denmark had changed a few years earlier so that banks could no longer make general provisions to protect against changes in circumstances or to reflect general concerns about how market forces might impact on their customers’ ability to pay. In summary, write-downs could only be made once OEI existed and only then by reference to the expected loss on that loan.
Third, although Mr Andersen was involved in determining the level of write-downs, Mr Choo-Choy did not put to him in cross-examination (or to any other witness) any specific challenge to the write-downs made. In particular, Taberna has not identified any customer who should have been identified as having OEI who was not so identified by Roskilde still less any particular customer in respect of whom it might be said that a write-down should have been made but was not made or in respect of which a bigger write-down should have been made.
Fourth, I bear in mind the evidence before me (which I accept) that the write-downs made by Roskilde were subject to four levels of checks:
First, as Mr Andersen explained, there was a detailed internal procedure for write-downs undertaken by the individual branches, the Credit Department and the Board of Directors. This was unchallenged evidence.
Second, the approach adopted by the Bank was audited by the Bank’s Internal Audit Department. Although Taberna relied heavily on the comments and recommendations for improvements made by the Bank’s Internal Audit department in its Report of 8 February 2008, the fundamental conclusions reached by the Bank’s internal audit department can be seen from the Internal Audit Report in the Annual Report 2007 which states that having conducted its audit, the Bank’s Internal Audit department concluded that: “the procedures and controls established, including the risk management organised by Management relevant to the entity’s reporting processes and significant business risks, are working satisfactorily”. And “In our opinion, the Annual Report gives a true and fair view of Roskilde Bank A/S’ financial position at 31 December 2007.”
Third, the approach adopted by the Bank and the Bank’s Internal Audit Department was then audited by the Bank’s external auditors Ernst & Young (“E&Y”).
Fourth, as explained by the Danish FSA in its letter of 8 February 2008the Danish FSA conducted a review of Roskilde’s loan book including 33 of the largest exposures amounting to DKK 16.6 billion at the end of September 2007 (out of a total loan book of DKK 32.5 billion). In particular, as part of this review, the Danish FSA considered whether the correct level of write-downs was being made and made various recommendations about the level of write-downs to be made until the Board of Directors proposed a level which the FSA were content with. This resulted in the DKK 267 million figure announced.
In response, Mr Choo-Choy submitted that there was no discussion in E&Y’s report of the nature of the work done by Roskilde in conjunction with the Danish FSA; and that it is unclear therefore to what extent E&Y took account of what the Danish FSA required of Roskilde or of the Danish FSA’s continuing concerns. I see some force in these submissions and I bear them well in mind – as I do the comments in various internal documents indicating certain shortfalls in Roskilde’s internal procedures and recommending various improvements; but, in my judgment, they provide a very slender basis for the positive finding sought by Taberna in this context.
Fifth, I am much concerned that Mr Choo-Choy’s argument is coloured by hindsight. For example, it is right that Mr Andersen agreed that the market had “started to fall” in the second half of 2007; but, Mr Andersen did not (I think) say that it was already “gathering momentum” at that stage. In any event, the suggestion that the write-downs at the end of 2007 and the forecast for 2008 should have been higher is founded, at least in part, by the notion that Roskilde should have predicted the dramatic fall in the property market which occurred in 2008. However, I had no expert evidence to suggest that Roskilde was somehow at fault in failing to make such prediction and to take into account the probability or even possibility of such fall at the relevant time. It is right that much larger write downs were taken in July 2008. However, as submitted by Mr Béar, pointing to larger write-downs in a subsequent accounting period does not establish that those write-downs ought to have been taken in a previous accounting period; and it is particularly not probative where the market appears to have taken a steep downturn in the course of 2008 rather than earlier. In this context, Mr Andersen explained in his witness statement the “dramatic change in the real estate market [during the first half of 2008] and the financial crisis” and how circumstances changed during the first half of 2008 in relation to the large customer credits, with unforeseen problems appearing during the first half of 2008 for large numbers of customers. This evidence was unchallenged. Further, it is noteworthy that as appears from the 2008 Q1 Report write downs were still relatively modest during that first quarter of 2008 i.e. DKK 53 million.
For these main reasons, it is my conclusion that Taberna has failed to establish on a balance of probability that the representations in relation to write-downs were false. It follows that Taberna’s claim under this head must fail.
PROJECT FINANCING
Under this head, Taberna relied on the following representations:
The statement at p51 of the Offering Circular (as forwarded by Roskilde to Taberna, through Deutsche Bank, by email on 16 January 2008) with regard to “the financing of property projects” that “[t]he projects are typically ring-fenced in SPVs, where all rights under the project [are] pledged or assigned to the Bank, and most importantly the projects are generally sold or leased before the project is activated” (emphasis added).
The statement at p11 of the Annual Results 2007: “Project Financing … LTV 80-85% as a general rule – Presale condition of 50-70%”.
As already stated, I was not persuaded that the Annual Results 2007 were seen or read by anyone at Taberna. Thus, in my view, the second of these representations can be ignored. There is a separate issue as to whether the representation as to project financing was also made orally in the course of one or more of the telephone conversations between Taberna’s representatives and Roskilde’s representatives. As to this, the evidence (which it is unnecessary to recite) was uncertain and conflicting. Although, Mr Dalgas accepted that he “possibly” told Taberna about Roskilde’s policy in this respect, the evidence in this regard is, in my view, too weak to justify on a balance of probability a positive finding as sought by Mr Choo-Choy. Thus, it is only the statement on p51 of the Offering Circular as referred to above which requires to be considered under this head. As to the representations in this document, I have already dealt with certain general submissions made by Mr Béar; and it is unnecessary to repeat my conclusions in that regard. For present purposes, I proceed on the basis that the particular representation under this head was made to Taberna with the requisite “intent”. Further, there is no doubt that Taberna’s credit reports included various references to project financing as already quoted above. However, absent evidence from the main author of those reports (i.e. Mr Bellopede), the source of such information is uncertain, particularly since such references are different from what is stated in relevant respects in the Offering Circular. For this reason, I am unable to make a positive finding that Taberna relied upon this particular representation in the Offering Circular. On this basis, it is my conclusion that this claim under this head cannot succeed.
In any event, I am unpersuaded that Taberna has satisfied the burden of proof of showing that the representation on p51 of the Offering Circular was false. In that context, it is important to focus on the nature of the representation as expressed in the words used i.e. that the projects were “generally” sold or leased before a particular project was activated. Thus, it was common ground that the representation did not suggest any “universal” rule; but there was much debate as to what was meant by “generally”. In that context, Mr Choo-Choy submitted that Taberna’s natural understanding of this representation, as explained by its witnesses, was that the activation of such projects without 50-70% pre-sales having been achieved would be an exception to the general rule; that Messrs Dalgas and Nielsen both accepted that the meaning conveyed and intended to be conveyed was that it would only be as an exception to the general rule that pre-sales of less than 50% would suffice; and that therefore there does not appear to have been any real difference of understanding between Roskilde’s witnesses and Taberna’s witnesses. I am prepared to proceed on this basis – although, even so, it seems to me that the nature of the representation is extremely vague.
Here, Mr Choo-Choy fairly accepted that there was a paucity of evidence with regard to whether the general rule was observed throughout by Roskilde but he nevertheless submitted that there were what he described as “serious indications” in the evidence that it was not. In particular, he relied on the following evidence:
A number of instances (over 10 and possibly up to 15 or 16) of apparent non-observance in the list of project finance loans in excess of DKK 50 million;
Mr Nielsen’s calendar invitation in which he refers in terms to a particular project finance loan with “0% advance sales and 0% advance rentals” – in circumstances where he clearly appears to consider there should have been pre-sales within the specified range.
An entry at item 7 in para 2.8 of the Risk Report of 28 January 2008 where it is stated:
“There is a case concerning project sales that turns out not to have the pre-sale that it should have before we make financing available according to the credit policy. When asked directly, the Credit Director says that there are probably a few more cases. The scope, however, is unknown.”
Thus, Mr Choo-Choy submitted that what is significant in this passage is that the Credit Director stated that there were probably a few more cases which should have had pre-sales but did not, i.e., cases where the general rule was violated rather than exceptions to the general rule; and that, more worryingly, the Credit Director stated that the precise scope of such violations was unknown.
Despite the concern about whether the general rule was being properly applied and whether Roskilde could properly continue to make the representation as it continued to do e.g. in its Annual Results 2007, there is no evidence of any discussion or study confirming that the rule was being applied and that the representation could therefore continue to be made. Further, Mr Andersen confirmed in evidence that he was “not aware of any study being conducted by the credit department in either late 2007 or in early 2008, before the annual report was published, to consider whether or not the general rule for pre-sales was observed in relation to project finance loans”. He said he had no knowledge of such a review and that, if such a review had taken place, he would have heard of it.
In my view, this evidence (even taking it at its highest) falls short of showing that this particular representation was false. Moreover, as submitted by Mr Béar, a more detailed analysis of the list of project finance loans shows that the instances of a material failure to follow the suggested 50% pre-sale requirement was significantly less than Mr Choo-Choy submitted i.e. 2% by value and only about 6% of the 81 examples on the list. For these reasons, I am not persuaded that the representation on p51 of the Offering Circular was false. It follows that Taberna’s claim under this head must fail.
Preliminary conclusion
For these reasons and subject to the question of contributory negligence as referred to below, it is my preliminary conclusion that Roskilde is liable to Taberna in damages under s2(1) of the 1967 Act for misrepresentations in respect of NPLs as set out above but not otherwise. As to the amount of such damages, it is common ground that (i) Taberna paid the sum of €26,421,585 as the purchase price for the subordinated notes; and (ii) there is no realistic prospect of Roskilde having the funds to redeem, purchase or repay the subordinated notes. It follows that the amount recoverable by Taberna by way of damages is that sum plus interest (which I hope can be agreed) subject only to the question of contributory negligence.
Contributory negligence
I have already considered the applicable legal principles earlier in this Judgment and concluded that Roskilde is in principle entitled to rely on a defence of contributory negligence under the 1945 Act subject to the two points referred to in paragraph 111 above.
Before turning to the facts, it is convenient to say something more about the first point viz. the so-called rule in Redgrave v Hurd (1881) 20 Ch D 1 which was, of course, decided by the Court of Appeal almost 150 years ago, being preceded some 30 years earlier by the celebrated dictum of Lord Cranworth in Reynell v Spyre (1852) 1 De G.M. & G 660, 710 (“No man can complain that another had too implicitly relied on the truth of what he has himself stated”);and, in effect, subsequently adopted and applied in numerous later cases including Nocton v Ashburton [1914] AC 932, 962 (where Lord Dunedin famously stated “No one is entitled to make a statement which on the face of it conveys a false impression and then excuses himself on the ground that the person to whom he made it had available the means of correction”); The Arta [1985] 1 Lloyd’s Rep 534; and Strover v Harrington [1988] Ch 390, 410 (where Sir Nicolas Browne-Wilkinson V-C stated “… if it is once shown that a misrepresentation has been made, it is no answer for the representor to say that the representee has been negligent and could have found out the true facts if he had acted otherwise …”). In truth, it has always seemed to me difficult to reconcile the conclusion that the defence of contributory negligence is available to a claim for misrepresentation under s2(1) of the 1967 Act with these strong statements which are expressed in terms which almost suggest a general principle amounting to a rule of law. Notwithstanding, the approach adopted by Sir Donald Nicholls V-C in Gran Gelato was that although the 1945 Act applies, the court must still consider what, if any, apportionment to make; that, in that context, the question is whether any reduction in damages is “just and equitable”; and that, although “in principle” carelessness in not making other enquiries provides no answer to a claim when the plaintiff has done that which the representor intended he should do, it would need to be a “very special case” before carelessness would make it just and equitable to reduce the damages otherwise payable. In my view, the main difficulty is to identify even in general terms the kind of case which might properly be regarded as a “very special case”. So far as I am aware, there is no reported case where damages otherwise payable for misrepresentation under s2(1) of the 1967 Act have been reduced by reason of contributory negligence.
These observations are, in my view, important because they provide the necessary backcloth to the submissions made by Mr Béar that there were numerous and repeated failures on the part of Taberna which are relevant in this context.
In particular, Mr Béar relied upon a number of general submissions viz. (i) this is not a case where Roskilde intended Taberna to act on its representations: Roskilde was essentially indifferent and gained nothing from the potential transaction; (ii) the matters of contributory negligence raised are wide-ranging and not connected with anything said by Roskilde of which complaint is made; and (iii) Taberna’s failures are far more serious than anything alleged against Roskilde. Taberna, the manager of many billions of dollars of funds (i.e. including both Europe funds as well as numerous US funds: the Europe CDO funds alone were €1.363 billion as at September 2008, approximately DKK 10 billion) was in truth a far more sophisticated operator than a regional Danish bank.
In addition, Mr Béar relied upon a number of specific matters. Given my earlier conclusions, some of these are no longer relevant. In summary, those that remain potentially relevant were Mr Béar’s allegations that (i) Taberna failed to take any specialist advice; (ii) Taberna relied on out of date advice/information; and (iii) Taberna failed properly to assess Roskilde’s capital adequacy.
Failure to take any specialist or country advice
It is common ground that Taberna did not take any specialist or specific country advice despite not having familiarity with the Danish market for either banking or real estate or even having anyone who could speak/read Danish. In that context, Mr Béar sought to rely, in particular, on the evidence of Mr Schneider who acknowledged that it would have been reasonable to engage the services of someone who understood Danish and Danish accounts to help them evaluate the investment. In summary, the result, submitted Mr Béar, was, for example that (i) Taberna had no independent person to review all the reports and look for any discrepancies between the Danish version and the English version; (ii) Taberna was unaware of the VO document published on 8 February 2008 which included a large amount of information including the fact that Roskilde’s loans in default amounted to DKK 522 million which, on Taberna’s case, would have been critical to its decision to invest; and (iii) Taberna did not carry out any general research into the market which would have revealed, for example, the more up to date views of other institutions, such as Carnegie Bank, a Nordic financial house specialising in this area.
Reliance on out of date advice/information
In summary, Mr Béar submitted that it was clear from the credit report (which gives the Moody Credit rating as one of the first pieces of information) and from the discussions leading up to the investment decision that Taberna based its decision to invest on a credit rating and report which Moody’s had issued in May 2007 based on Roskilde’s Q1 2007 Report; and that this was obviously “dangerous” in particular because it was produced before the 14 January Announcement which identified substantial write-downs. Second, Mr Béar submitted that the decision to invest was also taken (under obvious time pressure) almost exclusively on the basis of the data from the Q3 Report, since the final version of the credit report was produced by Mr Bellopede on 30 January 2008 prior to the publication of the Annual Report 2007 and VO document on 8 February 2008 which (as I have found) were never read by Mr Frappier (either in the Danish or English versions) either prior to the decision to purchase or even completion.
Capital adequacy
As appears from the credit report, Taberna’s analysis of the capital strength of Roskilde was that it was “well capitalised with a capital adequacy ratio of 13% and a core capital ratio of 8%”; and that the “stress test” carried out to test the strength of Roskilde’s capital adequacy by reference to the “statutory capital adequacy requirement” of 8% indicated a capital surplus of 5%. As a result, Taberna’s analysis concluded that Roskilde could easily incur losses at the level tested in the stress test and consequently had a “very comfortable” capital adequacy and solvency ratio. While Mr Bellopede did review the 2007 Announcement, he concluded that this showed that “the bank remains well capitalized”. However, Mr Béar submitted that this fundamentally misunderstood Roskilde’s capital adequacy because (i) the Annual Report 2007 showed that Roskilde’s capital adequacy ratio was only 11.9% and (ii) the Bank had an individual solvency need of 10.5%, giving it a capital surplus of only 1.4% rather than 5%; and that the unchallenged evidence of both Mr Nielsen and Mr Krogh was that this meant that it would only take losses of DKK 430 million for Roskilde no longer to meet its capital adequacy requirement, which would require Roskilde to stop trading unless it could raise further capital. The result, submitted Mr Béar was that had the “stress test” been carried out based on the correct figures, it would have shown the Bank ceasing to be viable in two of the five scenarios being considered by Taberna which considered losses of up to DKK 594 million; that the conclusion in the credit report that Roskilde had a “very comfortable” capital adequacy was incorrect; and that this conclusion is supported by (i) the Danish FSA who concluded in its later report dated 17 June 2009 that “Roskilde Bank A/S’s buffer to resist future losses were in reality limited, since the solvency margin at the end of 2007 amounted to 1.4% (11.9 to 10.5) of the risk weighted items corresponding to approximately 430 million DKK … The Bank had thus chosen low capital cover” (emphasis added) and (ii) the evidence of Roskilde’s expert, Mr Krogh, that Roskilde’s capital adequacy was under stress for reasons set out in paragraphs 4.1 to 4.36 of his report.
In summary, Mr Béar submitted that these matters both individually and together show that the dominant cause of Taberna’s decision to invest was its own failure to carry out due diligence competently and with proper care; that these failures were so substantial that they collectively break the chain of causation from any cause of action which would otherwise exist; alternatively that these failures (and each of them) are grounds for a very substantial reduction in damages to reflect the sustained and repeated failure by Taberna to carry out any proper analysis of the investment opportunity.
I do not accept those submissions. In my view, the test as to what constitutes an intervening act sufficient to break the chain of causation is a high one: see e.g. per Gross LJ in Borealis v Geogas Trading [2011] 1 Lloyd’s Rep 482 at [44] to the effect that in order to break the chain of causation, there must be an event of such impact that it “obliterates” the wrongdoing of the defendant. Even accepting Mr Béar’s submissions as summarised above at face value, the matters relied upon – whether individually or collectively – do not, in my view, meet this test. Nor do I consider that they constitute a “very special case” which would make it just and equitable to reduce the damages otherwise payable. I should mention that Mr Choo-Choy advanced a number of detailed submissions to the effect that the matters summarised above and relied upon by Mr Béar were factually incorrect or overstated. However, given my conclusions as just stated, it is unnecessary to consider these submissions further.
In any event, I should mention that even if I had concluded that any of the matters relied upon by Mr Béar as summarised above (whether considered individually or together) would or might amount to contributory negligence on the part of Taberna so as to constitute a “very special case” justifying a reduction in damages, it would be necessary to consider all relevant circumstances in order to decide whether it would be just and equitable to do so – in particular the exercise performed by Roskilde as referred to in paragraphs 76-86 above. However, in the light of my conclusions, it is unnecessary to do so.
Conclusion
For these reasons, it is my conclusion that Roskilde is liable to Taberna in damages in the sum of €26,421,585 under s2(1) of the 1967 Act for misrepresentations in respect of NPLs as set out above but not otherwise. Counsel are accordingly requested to seek to agree a draft order to reflect the terms of this Judgment (including interest and costs). Failing agreement, I will deal with any outstanding issues.