UBS AG (London Branch) & Anor v Kommunale Wasserwerke Leipzig GmbH

This is a joint majority judgment to which both Lord Briggs and Hamblen LJ have contributed, and with which we both fully agree.

The Appellants, UBS AG, a bank incorporated in Switzerland, and UBS Ltd, an English subsidiary of UBS AG (together "UBS"), are an investment bank. The main Respondent, Kommunale Wasserwerke Leipzig GmbH ("KWL"), is the Leipzig municipal water company, responsible for the supply of water and sewage services to the people of Leipzig.

In 2006 and 2007 KWL was persuaded to sell credit protection to UBS and to two intermediary banks, Landesbank Baden-Württemberg ("LBBW") and Depfa Bank plc ("Depfa"), the other Respondents. It did so by means of complex derivative products known as Single Tranche Collateralised Debt Obligations ("STCDOs").

The effect of these STCDOs was that if a certain number of the entities in the reference portfolios defaulted during an eight or ten year period, KWL would be liable to pay the banks hundreds of millions of dollars. Defaults duly occurred following the global financial crisis of 2008–9 and the banks sought payment of the sums due, making claims in excess of €350 million.

KWL was persuaded to enter into the STCDOs by its corrupt financial advisers, Value Partners Group AG ("Value Partners") (a Swiss company), assisted by a bribe of around US$3 million paid by Value Partners to Mr Klaus Heininger, one of KWL's two managing directors. This bribe was paid out of the premium of just over US$30 million paid to KWL for providing credit protection under the STCDOs, all bar €4.5 million of which was siphoned off by Value Partners. The premium payments were funded, and UBS's booked profit of over US$25 million was made, by UBS selling equivalent protection to the market through collateral hedging contracts.

Although UBS was not aware of the bribe, it had entered into an arrangement with Value Partners whereby Value Partners would advise their municipal clients to enter into STCDOs with UBS regardless of the clients' interests. KWL was the first client "delivered" under this arrangement. KWL was not aware of this arrangement.

The trial judge, Males J, held that KWL was entitled to rescind the STCDOs with UBS on the grounds of bribery and conflict of interest. The judge further held that Depfa and LBBW were entitled to rescind the STCDOs they had entered with UBS on the grounds of fraudulent misrepresentation.

In addition, the judge held that even if the STCDOs were valid and binding the losses on the portfolios were caused by their negligent management by the portfolio manager UBS Global Asset Management (UK) Ltd ("UBS GAM").

UBS and UBS GAM appeal the judgment on various grounds and there are 10 agreed issues to be determined on the appeal, including two grounds of cross appeal by KWL.

A detailed narrative of events is set out in the judgment of Males J. It is not necessary to repeat that full account in this judgment since there is only one main factual finding which is challenged relating to the knowledge of Dr Andreas Schirmer, KWL's other managing director. Instead a brief summary of the key events relevant for the purposes of the appeal will be set out, drawing on and giving cross-references to the judgment. More detailed reference to the judge's findings will be made when considering the various issues which arise on the appeal.

KWL is a German municipal water company. At all relevant times, its Executive Board consisted of two managing directors, namely Mr Heininger and Dr Schirmer. Mr Heininger took the lead on commercial matters while Dr Schirmer did so on technical matters. It also had a Supervisory Board to provide oversight of the Executive Board's activities.

In 2002, KWL engaged Global Capital Finance ("GCF") as a financial adviser. The key principals of GCF were Mr Berthold Senf, and Mr Jürgen Blatz. In around April 2004, Mr Senf and Mr Blatz left GCF and formed another financial advisory company called Value Partners. A corrupt relationship developed between KWL and Messrs Senf and Blatz of GCF (and later Value Partners). This began with the giving of generous gifts and expenses paid luxury trips to Mr Heininger and others at KWL from 2002 onwards.

Between 2000 and 2005, KWL entered into four cross-border lease transactions ("CBLs"), the provisions of which were materially identical for present purposes. Under each of these agreements, KWL leased certain of its infrastructure assets to a special purpose vehicle ("the Trust"). The Trust was funded by deposits from overseas investors and loans. It was domiciled outside Germany, allowing it to take advantage of depreciation provisions under foreign tax laws which KWL was not itself able to realise. This enabled the Trust to pay an upfront premium to KWL in return for the lease of the assets.

KWL used part of this payment to obtain a payment undertaking or bond from a highly rated bank or insurance institution ("the defeasance provider"). The four defeasance providers were: i) Balaba (which undertook to pay £96.2m in 2014 to support the "Balaba CBL"); ii) Merrill Lynch (which undertook to pay US$65m in 2025 to support the "Wastewater CBL"); iii) General Electrical Capital Corporation, or "GECC" (which issued a floating rate note pursuant to which it agreed to pay €77m in 2014 to support part of the "Freshwater (F1) CBL"), and; iv) MBIA Global Funding LLC (which issued a zero coupon bond pursuant to which it agreed to pay US$50m in 2033 to support the balance of the "Freshwater (F2) CBL").

The Trust then sub-leased the infrastructure assets back to KWL. Under the terms of that sub-lease, KWL was required to pay rent to the Trust, which in turn enabled the Trust to pay interest on its loans and to pay a return to its investors. KWL funded these rent payments with the proceeds of the bonds provided by the defeasance providers. At the end of the sub-lease term, KWL had the option to re-purchase the infrastructure assets at a predetermined price, such purchase to be funded by the remainder of the bond provided by the defeasance provider.

The overall result of these CBLs was the generation of funds for KWL, which it needed to upgrade its facilities, reduce running costs and lower the water premiums paid by Leipzig citizens. At the same time, the leases exposed KWL to the risk that the defeasance providers might default. This would render KWL unable to buy back its infrastructure assets when the sub-lease term came to an end, thereby preventing it from carrying on its business.

Value Partners had advised KWL in relation to at least three of the leases and in 2005, KWL made Value Partners responsible for the regular reporting on and monitoring of all concluded and future CBLs. The CBLs provided the opportunity for the corrupt relationship between Value Partners and Mr Heininger to escalate to outright cash payments. In May 2005, Value Partners made a cash payment to Mr Heininger of €945,945 in connection with the Balaba CBL, and made a donation of €150,000 to Mr Heininger's favourite football team [54].

By 2006, KWL was cash-strapped and under pressure to raise further funds. At the same time, Mr Heininger, Mr Senf and Mr Blatz were keen to find new ways to profit personally at KWL's expense. KWL could not enter any new CBLs, as it had already leased all of its relevant infrastructure assets, and CBLs were in any case becoming less attractive as the tax loopholes of which they sought to take advantage were closed up. In early 2006, Mr Heininger asked Value Partners about the possibility of restructuring the existing CBLs. As part of their investigations into potential restructuring, Mr Senf and Mr Blatz contacted Mr Steven Bracy, a former colleague of theirs who now worked at UBS.

On 6 April 2006, a telephone conversation took place between Mr Senf, Mr Blatz and Mr Bracy, during which Value Partners made inquiries as to restructuring possibilities on behalf of a then-unnamed client (i.e. KWL). During this conversation Value Partners raised the possibility of restructuring the CBLs by means of a type of credit default swap ("CDS") known as a collateralised debt obligation ("CDO"), and asked for more information on the subject. Mr Bracy explained the way in which CDOs worked, and suggested a specific type of CDO known as a single tranche collateralised debt obligation ("STCDO") as being appropriate for Value Partners' purposes.

A traditional CDO involves a number of investors contributing capital which is used to acquire a portfolio of income-generating assets. The portfolio is divided into a notional hierarchy of 'tranches' which all carry different risk characteristics. Investors are allocated to a tranche, and accordingly stand to gain the income generated by that tranche, whilst also bearing the risk associated with that tranche. The parameters of each tranche are defined by an attachment point and a detachment point. The attachment point describes the point at which a default in the underlying portfolio will impact on the tranche, and is usually expressed as a percentage of the total value of the portfolio (or "CDO notional"). For example, an attachment point of 4% means that if the total losses from defaults on the portfolio are less than 4% of the CDO notional, that tranche will suffer no loss. However, once losses exceed 4%, the capital value of the tranche will begin to be written down.

The detachment point, also usually expressed as a percentage of the CDO notional, defines the level of default in the underlying portfolio at which the risk of losses associated with that tranche is exhausted. For example, a detachment point of 5.5% means that once losses from defaults on the portfolio exceed 5.5%, there is no income or capital remaining for distribution to the investors holding that tranche, and their investment is wiped out. The tranche width is the difference between the attachment and detachment points. The tranche in the above example would have a tranche width of 1.5%. Expressed in monetary terms, the tranche width represents the capital value of the tranche (i.e. the maximum that investors can lose). So, for example, for a portfolio of investments worth US$1bn, a tranche width of 1.5% would correspond to a CDO notional of US$15m.

The hierarchical structure of a CDO means that a tranche with a lower attachment point carries a greater risk, and is known as a junior or junior mezzanine tranche. A tranche with a higher attachment point carries a lesser risk, as it is protected by subordinate tranches which bear the first risk of loss. In theory, it will also take fewer defaults to extinguish a tranche with a narrow tranche width than one with a wider tranche width.

There are a number of modern variations on the traditional CDO model. These include 'synthetic' STCDOs, under which rather than putting in cash to acquire an investment portfolio of bonds upfront, the portfolio is composed of CDS contracts which reference such bonds. As an STCDO only has one investor, this hierarchy of tranches is wholly notional, in that the level of subordination of a tranche is not defined by reference to parameters of other actual tranches, but rather by reference to the credit buffer represented by the attachment point. Such a synthetic STCDO can be structured on an unfunded basis, in which case the investors make no up-front payment but instead take the risk of having to make a future payment in the event of sufficient defaults occurring in the portfolio to impact their tranche. Another potential variation is that the portfolio which underlies the STCDO can be either managed or static.

The STCDOs suggested for KWL by UBS were synthetic, unfunded, with a single, mezzanine tranche and related to a managed portfolio. They would operate as follows:

In the early 2006 discussions between Mr Heininger and Messrs Senf and Blatz about the possibility of restructuring KWL's CBLs in order to generate additional funds, it was apparent that any such restructuring would not only be for the legitimate purpose of raising funds for KWL, but would also involve the personal enrichment of Mr Heininger and the corrupt exploitation by Value Partners of its relationship with Mr Heininger and thus with KWL. The judge found that personal enrichment was throughout the primary motivation for all three of these dishonest individuals [591].

Both Value Partners and, when reported back to, Mr Heininger, appreciated the potential for STCDOs to generate large premiums for KWL, from which they hoped to benefit personally. Value Partners made it clear to Mr Bracy that KWL would want to maximise the upfront premium payable under any STCDO. A letter of advice written by Value Partners to KWL on 19 April 2006 made no mention of any premiums, but instead emphasised that the STCDOs would operate to diversify the risks borne by KWL under the CBLs. In reality, however, the transactions were intended to generate the maximum return for KWL, from which Mr Senf, Mr Blatz and Mr Heininger were planning to extract a large sum for themselves. On 26 April 2006, Mr Heininger signed a letter formally engaging Value Partners to act as KWL's advisor in relation to the CBL restructuring. Under the terms of this letter, Value Partners was to be paid a success fee of around US$7.875m [202]-[204][212].

Another corrupt arrangement developed between Mr Bracy and Mr Senf and Mr Blatz. As the judge found, the arrangement involved Value Partners' agreed role being to advise their clients to conclude STCDOs with UBS and to "deliver the client" to UBS, and to do so regardless of whether that was in the clients' best interests. The essential features of this arrangement were that they would work together to persuade Value Partners' existing municipal clients to enter into STCDOs with UBS whether or not it was in their interests to do so; get Value Partners engaged as advisers to other municipal entities with cross-border leases so that they could act likewise towards them; keep the relationship between UBS and Value Partners secret from the clients and generate significant profits for UBS and Value Partners [182]-[201].

One of the matters to be decided was who would manage the CDO portfolio once the STCDO was concluded. This would be for KWL to decide, advised as necessary by Value Partners. In view of the fees which could be charged for such management, this could be a valuable role. On 28 April 2006 Mr Blatz emailed Mr Bracy suggesting that Value Partners was "in a position to promote UBS as being the CDO manager over the term — of course, we would expect to receive a fair compensation for that!" Mr Blatz added that "Of course, that would also increase the pot we can spend for our closing party". The judge rejected Mr Bracy's evidence that he did not know what the "pot" was and did not understand that any payment would be kept secret from KWL. He found that the reference to spending the compensation on a closing party made it clear that this was money which KWL was not going to hear about or benefit from, and that it illustrated the way in which the interests of Mr Bracy and Value Partners were aligned in ensuring that the deal was done and did not correspond with KWL's interest.

Mr Bracy passed the suggestion for "compensation" on to Mr Sanz-Paris who discussed it with Mr Czekalowski (both UBS employees involved in the transactions). Eventually on 2 May 2006 Mr Bracy responded to Value Partners that payment for the promotion of UBS GAM involved an "appearance of impropriety" and therefore could not be considered. The judge found that this episode ought to have alerted the UBS deal team to the fact that Value Partners was prepared to act contrary to the interests of the client which it purported to represent, and was happy to be paid by both parties to the proposed transaction, but that the deal team were quite content to continue to deal with Value Partners on this highly lucrative transaction and did not even think of ensuring that KWL was aware that this request had been made [214]-[217].

On 9 May 2006, a 'kick-off meeting' took place. Those present included Mr Heininger, Dr Schirmer, Mr Senf, Mr Blatz, and Mr Bracy. Value Partners told UBS to avoid mentioning figures during this meeting and to emphasise the advantages of an STCDO for managing risk. The judge found that this was because it did not want Dr Schirmer, who was making a first appearance in the negotiations at the proposed meeting, to hear figures which would indicate what premium KWL could expect to receive. During the meeting UBS described the STCDOs as achieving the twin objectives of risk diversification and yield enhancement. Whilst the meeting resulted in Mr Heininger understanding the nature of an STCDO, on the judge's findings Dr Schirmer had plainly understood very little [225]-[240].

At the same meeting, or immediately afterwards, it was agreed that UBS GAM would manage KWL's STCDO portfolios. The judge found that it was clear from the documents that there was no serious consideration by Value Partners of any other candidate, that the choice of UBS GAM was promoted by the UBS deal team, and that the choice was made before negotiations took place to determine what level of fees UBS GAM would charge [218].

Because the STCDOs were large, complex and unusual transactions, they required approval by UBS's internal credit risk control ("CRC"). Mr Bracy and others in the deal team regarded this necessity as something of a nuisance. They understood that for this purpose it would be important to emphasise the risk diversification aspects of the STCDOs since, as the judge found, the view of UBS's CRC would have been that an essentially speculative transaction designed primarily to achieve an upfront cash premium for KWL, which exposed it to even a modest risk of liabilities running into hundreds of millions of dollars, would have been a thoroughly unsuitable transaction for a municipal water company. The judge found that it was the need to obtain UBS's internal approval for the deal which largely explains the increasing emphasis on risk reduction which appeared in the documents from this time [242]-[243].

On 16 May 2006 Mr Cox, another senior figure at Value Partners, queried UBS's proposed collateral hedge costs which he commented "seem wildly excessive". In an incident described as "beating up Mr Cox", Mr Senf, Mr Blatz, Mr Bracy and Mr Sanz-Paris saw off this unwelcome interference by Mr Cox. Mr Senf sent an e-mail reassuring Mr Bracy that Mr Blatz had spoken to Mr Cox and that "everything ok!" Mr Bracy passed this on to Mr Sanz-Paris, joking: "Anyone else you want me to beat up?". The judge found that Mr Cox was sat upon by Mr Senf and Mr Blatz to make sure that he did not cause further trouble and that this confirmed the picture of Messrs Senf and Blatz working together with Mr Bracy to ensure the smooth conclusion of the deal regardless of KWL's interests. It was also at this stage that Value Partners told UBS that they and KWL were not talking to anyone else about the deal. The judge found that this was extremely valuable information as it left UBS free to maximise its profit from the deal free of any competition from another bank [251]-[258].

The information needed for the CRC approval process included information about KWL's credit rating. KWL had received a shadow rating of A- from Standard & Poor's in February 2004 and a further lower rating of BBB+ in October 2005. On 16 May 2006 Mr Senf emailed both of these to Mr Bracy commenting "use the papers however you want". Mr Bracy acted on this suggestion by causing the two ratings to be split into two separate documents and passing on only the out of date higher rating to Mr Sanz-Paris. This rating then featured reasonably prominently in the deal team's application for approval of the deal. The judge found this to be blatant dishonesty on the part of Mr Bracy, intended to deceive CRC [259]-[260].

On 24 May 2006 UBS and UBS GAM were formally engaged. The relevant documentation included an Engagement Letter, a Risk Disclosure Letter and a letter purportedly confirming KWL's authority to enter into the transaction [267]-[276].

As part of the approval process, on 30 May 2006 CRC held a due diligence meeting with Mr Heininger, to ensure that KWL understood the nature and risks of the transaction, that the transaction was suitable for KWL, and that KWL had the necessary approvals to enter into it. Mr Bracy provided Mr Heininger with a script for the meeting which he hoped would satisfy CRC. Mr Bracy knew that it was likely to be important to CRC that the purpose of the transaction was to reduce KWL's risk, that all necessary approvals were in place, and that KWL had been advised by appropriate professionals. Mr Bracy therefore set out the approach which Mr Heininger should adopt at the meeting, whether or not these things were actually true. The judge found that the fact that Mr Bracy was telling Value Partners in this way how best to brief Mr Heininger illustrates that their joint objective was to persuade CRC to approve the transaction, if necessary by providing a deceitful account of the purpose of the transaction and that there was no question of Value Partners providing any independent advice to KWL [285]-[297].

Despite these efforts, on 1 June 2006 CRC decided not to authorise the transaction, essentially on the basis that the credit risk was disproportionate to KWL's capacity, and that there was a real reputational risk to UBS should the transaction go wrong and UBS be required to take enforcement action. Mr Bracy subsequently escalated the approval decision to UBS's senior management, and a protracted struggle for approval ensued. Value Partners made it clear to Mr Bracy that they did not want to give Mr Heininger a full explanation of what had happened [298]-[315].

Eventually a number of very senior UBS figures decided to approve the transactions. This approval was made conditional on a number of factors, including a proviso that all CBL restructuring other than that which related to the Balaba CBL be done via intermediaries.

Another of the conditions was that the KWL Supervisory Board be informed of the transaction. Pursuant to this condition, on 8 June 2006 Mr Sanz-Paris sent a presentation, headed "Collateralised Debt Obligations — Presentation to KWL", to Mr Heininger for him to present at KWL's next Supervisory Board meeting. Although the presentation referred to KWL's investment in the STCDO as being "to diversify the risk concentration and achieve an enhanced yield", and although it had from the outset been made clear to UBS that at least one important objective of KWL was to achieve a substantial upfront cash premium, there was no mention in the presentation of the fact that the enhanced yield was to take the form of upfront cash, let alone of the amount of the premium. The judge found that consistently with Value Partners' instruction to Mr Bracy, even the KWL Supervisory Board was not to be told about one of the most important features of the transaction [320]-[322].

A third condition required UBS to keep on file Value Partners' marketing material to KWL, so as to provide evidence on UBS's files that Value Partners had the necessary expertise and had marketed the transaction properly to KWL. This posed a problem as there was no such material. Having discussed it with Value Partners it was proposed by Mr Czekalowski of UBS that in order to fulfil this condition a letter should be obtained from Value Partners stating that it had "conceived and marketed the structure to KWL", that it "had a mandate from KWL prior to approaching UBS for execution" and that it had "explained to named senior management of KWL the risks of and the structure of the Transaction and that KWL understood and has accepted the structure and any and all associated risks". The judge found that the first two of these statements were untrue, as Mr Czekalowski knew. Value Partners had not conceived or marketed the transaction and did not simply approach UBS to execute an already developed transaction for which it already had a mandate from KWL. The transaction had been conceived in its general outline by Mr Bracy in his initial discussions with Value Partners. This outline had then been proposed to Mr Heininger who had responded enthusiastically to the idea of a transaction which could generate US $30 million in cash. It had then been marketed to KWL by UBS, for example at the 9 May 2006 meeting, and in the course of this marketing, the transaction had been developed to meet KWL's requirements [323]-[326].

A closing meeting was arranged in London for 7 June 2006 and Mr Heininger and Dr Schirmer came over for that purpose. Value Partners again instructed Mr Bracy not to talk about numbers to Dr Schirmer. The judge found that, as must have been obvious to Mr Bracy, Value Partners was keen to ensure so far as possible that one of the two managing directors of KWL who were to sign the contracts did not have a full understanding of the transaction, as in fact he did not, and that UBS assisted Value Partners to achieve this objective by passing on and complying with this instruction [315].

The contracts were signed on 7 June 2006 and the Balaba transaction traded on the 8 June 2006. On that day UBS and KWL entered into an STCDO and a CDS in respect of the Balaba CBL. KWL purchased credit protection from UBS on Balaba for a premium of US$1.6m by means of a CDS. KWL also sold credit protection to UBS for an 8-year term on an A3-rated £153.7m single tranche (with a tranche width of 1.5% covering losses between 3.6% and 5.1% of the portfolio) STCDO in return for a premium of US$22.7m. The premium was paid into KWL's account at Wilmington Trust in the United States from which (unknown to UBS) it was subsequently extracted and distributed among Mr Heininger, Mr Senf and Mr Blatz [369]-[370].

Following the conclusion of the Balaba transaction the atmosphere at UBS among the deal team and senior management was euphoric. Many plaudits were handed out, to Mr Bracy in particular, encouraging him to foster his relationship with Value Partners. On 9 June 2006 Mr Bracy emailed Mr Ryan, part of UBS's senior management, suggesting that consideration be given "on how properly to incent" Value Partners to bring UBS opportunities. The judge found that the effect of this email was that UBS's senior management, in the person of Mr Ryan, was aware of Mr Bracy's inappropriate approach to the relationship with Value Partners which he was being asked to foster and aware too of Mr Bracy's inability or unwillingness to recognise obvious conflicts of interest [371]-[390].

KWL and Value Partners entered into a further engagement letter, dated 31 May 2006 but signed by Mr Heininger and Dr Schirmer on 14 June 2006. This recorded that "KWL's main objective in doing this transaction is to diversify the risk of its underlying crossborder lease portefeuille". It purported substantially to increase the fee to which Value Partners was entitled to include all proceeds of the transactions in excess of €4.5 million.

Dr Schirmer denied signing this letter but the judge found that he had done so. He further found that he did so because he was prepared to do what Mr Heininger asked him to do, without asking questions even though it was in a language which he did not understand, and in all probability without attempting to understand what he was signing.

The judge found that although Mr Bracy and Mr Sanz-Paris both denied having any knowledge of what Value Partners was being paid for the KWL transaction, they both had at least a good idea. Whilst they did not know any precise figures, he found that they both knew that Value Partners was being paid a very substantial fee which bore no real relationship to the services or advice that it had actually provided. They knew too that Mr Senf and Mr Blatz had been anxious to keep Dr Schirmer in the dark about the premium being earned from the deal, and must therefore have realised that it was because of their relationship with Mr Heininger that they were able to extract such a high fee [405].

Once the Balaba transaction had been completed it was necessary to find a bank which could act as intermediary between KWL and UBS for the remaining transactions. Discussions with LBBW began in July 2006. The LBBW transaction eventually closed on 7 September 2006.

On 8 September 2006 KWL entered into a CDS with UBS in respect of the GECC CBL, under which it purchased credit protection from UBS for a premium of €1.2m. KWL also sold credit protection to LBBW, as intermediary bank for a 10-year term on an Aa3-rated €76.7m single tranche STCDO (with a tranche width of 1.5% covering losses between 3.5% and 5% of the portfolio) in return for a premium of €7.6m. LBBW in turn sold the same protection to UBS on back-to-back terms. The KWL-LBBW swap has been referred to as the "LBBW Front Swap" and the LBBW-UBS swap as the "LBBW Back Swap". UBS paid LBBW's intermediation fee of €2m.

At this meeting Mr Heininger made a presentation to the Supervisory Board entitled "Further risk minimisation and optimisation of the existing CBL agreements". The judge found this presentation to be thoroughly misleading. In particular in relation to the STCDOs there was no mention of risk or of the premium which KWL was meant to receive or of the role or remuneration of Value Partners. Instead it claimed that KWL would receive about €4.5 million in the form of interest payments over time. The judge found that although Dr Schirmer, who was present, had not grasped that an upfront premium had been paid, and therefore did not realise that the reference to interest payments over time was misleading, he was content to allow Mr Heininger to mislead the committee and board members in other respects and did not intervene to clarify anything which Mr Heininger said. The judge further found that "amazingly" no member of the Supervisory Board asked a single question about why Mr Heininger was making the presentation.

Later in September 2006 articles began appearing in the Leipzig press questioning the conduct of Mr Heininger and Dr Schirmer regarding various business trips and expenses in 2002-2003 paid for by Mr Senf's and Mr Blatz's former company, Global Capital Finance. On 28 September 2006 the Mayor's office in Leipzig requested KWL's Supervisory Board to investigate and PricewaterhouseCoopers ("PwC") were instructed. On 6 October 2006 there was a meeting of the Supervisory Board at which Mr Heininger lied about the purpose of his trip to Dubai. Once PwC had produced their reports, a further Supervisory Board meeting was held on 8 November 2006. The conclusion of the Supervisory Board was set out in a press release issued on 8 November 2006. This stated that it had decided that travelling first class and by Concorde was not appropriate for KWL management, that no gifts had been accepted, and that the statements about the trip to Dubai made by Mr Heininger should be accepted. These statements included that the invitation had come from the "arranger" of the cross-border leases (i.e. Global Capital Finance) which was contrary to his initial explanation, that "no further contractual relationships between the parties (investor, arranger, LVB)" had taken place, and that "no further such trips on invitations of the business partners had taken place". The judge found that this was deliberately misleading in view of the continuing relationship with Value Partners and a planned safari trip they had arranged for Mr Heininger and Dr Schirmer. The judge found that Dr Schirmer knew that what Mr Heininger was saying about the absence of further contractual relationships involving Value Partners was misleading but took no steps to correct him.

On 5 October 2006 Mr Blatz had spoken to Mr Bracy to request his help to get Mr Heininger out of trouble. The help sought was for Mr Bracy to find somebody who would provide a false explanation for Mr Heininger's 2003 trip to Dubai. Mr Bracy responded with suggestions about people who might be able to help, commenting that they might be "uncomfortable" about doing so.

On 16 October 2006 Mr Blatz sent an email stating:

"any news on the letter for K? we should target to have something in hand by mid week. Please give us a call in order to discuss status quo".

The "letter for K" was a letter for Mr Heininger from somebody who would purport to explain the business justification for Mr Heininger's presence in Dubai.

Mr Bracy had agreed to use his contacts to find somebody who would be willing to provide such a false explanation. Later Mr Bracy suggested that if he was to persuade one of his contacts to provide the desired letter exonerating Mr Heininger, he might need to bribe them. The judge found that he appeared "to have been quite willing to do so, at any rate if provided by Value Partners with the necessary wherewithal". Ultimately no letter was obtained, despite Mr Bracy's best efforts.

In relation to this episode the judge concluded as follows at [465]:

"….[Mr Bracy's] relationship with Value Partners was indeed such that Mr Senf and Mr Blatz felt able to reveal their own dishonesty to him, confident that he would not report them, and to ask for his active assistance, equally confident that this would be forthcoming. Their confidence was well founded. The facts that Mr Bracy responded as he did, apparently without surprise or qualms of conscience, and that he was the one who suggested that those he had approached to provide false evidence might need an incentive to co-operate, strongly suggests that so far as he was concerned, and even though he was not aware of the bribe paid to Mr Heininger, the dishonesty of Mr Senf and Mr Blatz was not news. I so find."

The publicity caused by the expenses scandal caused LBBW to reconsider its relationship with KWL, and it eventually pulled out of any further transaction on 7 January 2007.

Meanwhile in November 2006 Depfa had been introduced as a possible intermediary. Following extensive discussions on 28 March 2007, KWL entered into a CDS with UBS in respect of the Merrill Lynch and MBIA CBLs, under which it purchased credit protection from UBS for a combined premium of US$5m. KWL also sold credit protection to Depfa, as intermediary bank, a 10-year term on two Aa3-rated single tranche STCDOs (with a tranche width of 1.5% covering losses between 4.2% and 5.7% of the portfolio) in return for a combined premium of US$12m. The KWL-Depfa swaps have been referred to as the "Depfa Front Swap" and the Depfa-UBS swaps as the "Depfa Back Swaps". UBS paid Depfa's intermediation fee of €1.3m.

Each of the STCDOs was subject to an ISDA Master Agreement governed by English law, save for the LBBW Front Swap, which was subject to a German standard form agreement, a Rahmenvertrag für Finanztermingeschäfte, governed by German law.

The total net premium paid to KWL under these transactions was US$28.1m plus €6.4m (i.e. the combined total of the STCDO premiums received by KWL from the banks minus the combined total of the CDS premiums paid by KWL to UBS). Only €4.5m of this sum was retained by KWL. The remainder was siphoned off by Value Partners pursuant to the engagement letter dated 14 June 2006, under which Value Partners was entitled to all proceeds of the STCDO transactions in excess of €4.5m. From the premium received by Value Partners, Mr Senf and Mr Blatz paid a bribe of around $3m to Mr Heininger.

UBS was not aware of this bribe. Its interest in the transactions derived from the profit it could make on the STCDOs by selling equivalent protection to the market. These collateral hedging contracts generated the revenue required to fund the premium paid to KWL, pay UBS's own costs, and ensure a profit for UBS. These profits were initially anticipated to be around US$20m plus UBS GAM's management fees of around US$7.5m. After the LBBW STCDO but before the Depfa STCDOs had been concluded, Mr Bracy reported to his boss that a profit of US$26 million had been booked by the investment bank so far, with a further profit of US$6 million for UBS GAM, and that there was more to come when the final STCDOs were concluded [129]-[134].

KWL did not obtain Supervisory Board approval for any of these transactions.

The significant level of risk to which KWL was exposed eventuated following the global financial crisis. In September 2008, KWL terminated the Balaba CDS. At the time, the Balaba CDS was 'in the money' for KWL, and UBS accordingly paid KWL a total of £3.3m in relation to its unwinding. In December 2009, the defaults on the portfolios underlying the Balaba and Depfa STCDOs breached the relevant attachment points. KWL did not respond to UBS's request to pay the sums purportedly due as a result. On 18 January 2010 UBS and UBS GAM issued proceedings in the English High Court to enforce the Balaba STCDO. On 26 February 2010 KWL commenced proceedings against UBS in Germany seeking a declaration that the Balaba STCDO was void, shortly followed by similar claims against LBBW and Depfa.

Further defaults exhausted KWL's tranche in the Balaba STCDO, causing the Balaba STCDO to terminate automatically. In April and June 2010 Depfa and LBBW purported to terminate their Front and Back Swaps. Finally, UBS purported to terminate the outstanding single-name CDSs, which were at that stage all 'in the money' for KWL.

The LBBW Front Swap is subject to the jurisdiction of the German courts. On 3 June 2013, the Leipzig Regional Court held that the LBBW Front Swap was valid and binding on KWL (albeit quantum remains to be determined). KWL is currently appealing this decision. All of the remaining transactions were found to be subject to the jurisdiction of the English courts (see the judgment of Gloster J at [2010] EWHC 2566 (Comm)).

UBS's primary claim was to recover around US$138m from KWL, that being the net sum which it claimed was due under the Balaba STCDO after giving credit for sums due to KWL under the remaining CDSs.

KWL defended that claim on the grounds that:

In the event that the Balaba STCDO was held to be void or avoided, UBS claimed return of the premiums paid under each of the STCDOs. The judge held that the rescission of the Balaba STCDO required KWL to repay the net premium of US$28.1m plus €6.4m less the aggregate of: i) the bribe paid to Mr Heininger and; ii) the sums received by KWL which were used to purchase additional subordination. KWL cross appeals on this issue.

UBS's secondary and alternative claim was for an equivalent sum from KWL by way of damages for breach of warranty and fraudulent misrepresentations made in connection with the transaction (broadly to the effect that KWL was entering the transaction in good faith, had obtained all necessary approvals, was not relying on UBS for investment advice, and had assessed the risks of the transaction for itself). KWL argued that this claim could have no independent existence from UBS's primary claim, but must stand or fall with it. The judge dismissed this claim on the basis that UBS's losses were caused by its own fraudulent conduct, which rendered the STCDO unenforceable. UBS appeals on this issue.

Finally, UBS also sought the return of the £3.3m paid to KWL on the unwinding of the Balaba CDS in September 2008. The judge held that the rescission of the Balaba STCDO required KWL to repay this sum. KWL cross appeals on this issue.

KWL claimed some US$66.9m from UBS, being the sum due to it on the early termination of the GECC, MBIA and Merrill Lynch single-name CDSs. The judge held that the rescission of the Balaba STCDO required him to dismiss this claim. KWL cross appeals on this issue.

UBS sought to recover some €75.5 from LBBW, that being the sum due on the early termination of the LBBW Back Swap. LBBW defended that claim on the basis that:

The judge held that UBS was not entitled to enforce the LBBW Back Swap, and that LBBW had validly rescinded the Back Swap on the ground of fraudulent misrepresentation by UBS. As a result: i) LBBW was required to return its €2m intermediation fee to UBS; ii) LBBW was required to undertake not to enforce its Front Swap against KWL, and; iii) LBBW was entitled to the return of the collateral lodged by it with UBS. Although not required to determine the point, the judge would have held that the liability of LBBW to UBS was dependent on KWL being liable to LBBW, but not on actual payment being made by KWL. UBS does not challenge the order for rescission in limine, but appeals against the judge's decision that rescission should be ordered on terms that LBBW be required not to enforce the Front Swap.

UBS sought to recover around US$83.3m from Depfa, that being the sum due on the early termination of the Depfa Back Swaps (Depfa having already paid around US$32.6m to UBS upon termination). Depfa resisted that claim on the basis that:

The judge held that UBS was not entitled to enforce the Depfa Back Swap. His analysis is substantially identical to that set out above in relation to the LBBW Back Swap. In addition, the judge held that Depfa was entitled to the return of the US$32.6m paid to UBS, less its intermediation fee of €1.3m. Although not required to determine the point, the judge would have held that the liability of Depfa to UBS was dependent on KWL being liable to Depfa, but not on actual payment being made by KWL. UBS appeals against the judge's decision that rescission should be ordered on terms that Depfa be required not to enforce the Front Swaps.

Depfa sought to recover some US$116m from KWL which it claimed was due on the early termination of the Depfa Front Swaps. This claim is contingent on Depfa being held liable to UBS on the Depfa Back Swaps. KWL resisted this claim on essentially the same grounds on which it resisted UBS's claim to enforce the Balaba STCDO, contending that: i) the conduct and knowledge of UBS were attributable to Depfa, and; ii) Depfa itself was grossly negligent in failing to recognise that KWL's directors did not have authority to enter into the Front Swaps.

Although it was not necessary for him to decide this point, the judge stated that if required to do so, he would have held that KWL had capacity to enter into the Depfa Front Swaps which, subject to Depfa's undertaking not to enforce, were valid and enforceable against KWL.

In the event that the Balaba STCDO, the LBBW Front Swap or the Depfa Front Swaps were held to be binding on KWL, KWL sought to recover damages from UBS GAM for the negligent management of the STCDO portfolios. Although not required to determine the point, the judge would have held that UBS GAM's management of the portfolios failed to meet the required standard, so that KWL would have been entitled to recover damages from UBS GAM to indemnify it against any liability to LBBW or Depfa under the STCDOs.

The parties are agreed that for the purposes of the appeal, and the matters raised by Respondent's Notice or cross appeal, the following issues arise for determination:

The judge found that, in procuring that its client KWL entered into the Balaba STCDO with UBS, Value Partners acted as the agent of UBS, and that the bribe which it paid to Mr Heininger was something done by Value Partners within the scope of that agency, so that UBS was responsible for it in law, even though unaware of it. He held that this agency arose not from any contract, but from the corrupt arrangement whereby Value Partners undertook to deliver its "captive" clients to UBS for the purpose of making STCDO transactions with them, and UBS undertook to assist Value Partners in that endeavour, knowing that Value Partners was the fiduciary (and advisory) agent of its captive clients, and that Value Partners intended to abuse those fiduciary relationships by encouraging those clients to enter into STCDOs with UBS, regardless of their best interests, for the mutual financial benefit of Value Partners and UBS. KWL was the first (and in the event the only) captive client of Value Partners which entered into an STCDO with UBS by reason of that corrupt arrangement.

The judge's finding that this corrupt arrangement amounted in law to an agency of Value Partners for UBS may be found encapsulated in the following extracts from the judgment:

"596.

On the findings which I have already made in the narrative section of this judgment (see e.g. [188] to [201] above), UBS and Value Partners did agree to a relationship which as a matter of law amounted to an agency relationship. From as early as April 2006 there was in place an arrangement between Mr Bracy and Value Partners whereby Value Partners would advise clients which it had represented in connection with cross-border leasing transactions to conclude STCDOs with UBS. It described this arrangement in terms of "delivering the client" to UBS. Indeed, Mr Bracy described Value Partners to his boss in May 2006 as having "captive clients", including KWL, who would in practice do what Value Partners recommended (see [223] and [224] above). He later made clear his understanding that in practice Value Partners would exercise a degree of "control" over clients in directing business to UBS (see [399] to [403] above). It was quite clear to Mr Bracy and to Value Partners that the purpose of their arrangement was to make money for Value Partners and (indirectly) for Mr Bracy, regardless of whether the transaction was in the clients' best interests. All this went well beyond an understandable desire to do business or further business together if suitable opportunities arose.

…598. It is clear too that throughout the period between April 2006 and the conclusion of the Balaba transaction, UBS (primarily but not exclusively in the person of Mr Bracy) and Value Partners did work together in order to ensure the conclusion of the transaction, regardless of KWL's interests. This working together can be seen in the request, albeit not acceded to, for payment in return for recommending UBS GAM as portfolio manager (see [214] above); the preparations for the 9 May 2006 "kickoff" meeting ([225] and [226]); the silencing of Mr Cox ([251] to [256]); the e-mails exchanged between Mr Bracy and Messrs Senf and Blatz in connection with the visit to Wilmington Trust which show a clear alignment between UBS and Value Partners on the one hand and the "client" on the other ([262] to [265]); the briefing of Mr Heininger in preparation for the 30 May 2006 due diligence meeting ([286] to [290]); the care taken to ensure that Dr Schirmer did not have a full understanding of the financial aspects of the transaction ([314] and [315]); and the rather shocking efforts made by Mr Bracy and Mr Senf to remove language from the draft Freshfields opinion which might represent an obstacle to the conclusion of the transaction ([344] to [351])."

The judge acknowledged that some of the matters referred to in his paragraph [598] occurred after the conclusion of the Balaba STCDO, but he considered that those matters constituted evidence of the nature of the arrangement between Mr Bracy and Value Partners which existed from the outset. He found, and this is not challenged on appeal, that Mr Bracy acted throughout within the scope of his employment by UBS, so that his participation in the corrupt arrangement was the participation of UBS, and his knowledge about it was to be attributed to UBS [597].

Apart from noting (correctly) that parties may conclude what the law regards as an agency between them without either of them regarding it as an agency relationship, and that a person may be constituted agent for both parties to a contemplated transaction, the judge did not set out a precise legal analysis of why the corrupt arrangement which he found to have existed amounted in law to an agency. It may be that the way in which the matter was argued before him did not require him to do so, but this question of legal analysis became the central focus of the contest between the parties about this issue on appeal.

At the very beginning (paragraph 1-001) of Bowstead and Reynolds on Agency (20^th Ed) the authors set out the following well known definition of agency:

"(1)

Agency is the fiduciary relationship which exists between two persons, one of whom expressly or impliedly manifests assent that the other should act on his behalf so as to affect his relations with third parties, and the other of whom similarly manifests assent so to act or so acts pursuant to the manifestation. The one on whose behalf the act or acts are to be done is called the principal. The one who is to act is called the agent. Any person other than the principal and the agent may be referred to as a third party.

(2)

In respect of the acts to which the principal so assents, the agent is said to have authority to act; and this authority constitutes a power to affect the principal's legal relations with third parties.

(3)

Where the agent's authority results from a manifestation of assent that he should represent or act for the principal expressly or impliedly made by the principal to the agent himself, the authority is called actual authority, express or implied. But the agent may also have authority resulting from such a manifestation made by the principal to a third party; such authority is called apparent authority.

(4)

A person may have the same fiduciary relationship with a principal where he acts on behalf of that principal but has no authority to affect the principal's relations with third parties. Because of the fiduciary relationship such a person may also be called an agent."

At paragraph 1-019 the authors provide a commentary upon the scope of their paragraph (4) as follows:

"Incomplete agency: internal relationship only – the "canvassing" or "introducing" agent. Article 1(4) seeks to achieve completeness by taking in a well-established type of intermediary who makes no contracts and disposes of no property, but is simply hired, whether as an employee or independent contractor, to introduce parties desirous of contracting and leaves them to contract between themselves. In effecting such introductions he is remunerated by commission, which he may sometimes take from both parties. Such a person is a common figure in most western legal systems and may well be referred to as an agent. The most obvious example of such an intermediary in the English cases is the estate agent, who introduces purchasers to vendors and tenants to lessors of houses, and vice versa. Such persons are sometimes also referred to as brokers, and indeed in some English speaking countries the estate agent is referred to as a "real estate broker": but this may be misleading since the current practice, at any rate in England, is to use the term "broker" for persons who go beyond introductions and certainly do make contracts for their principals, e.g. commodity brokers, insurance brokers and stockbrokers. Canvassing agents are on the fringe of the central agency principles used by the common law, since their powers to alter their principals' legal relations are at best extremely limited. They often, however, have authority to receive and communicate information on their principals' behalf, and in so doing have the capacity to alter their principals' legal position. They also usually act in a capacity which may involve the repose of trust and confidence, and hence may be subject in some respects to the fiduciary duties of agents towards their principals. They are also subject of typical rules, largely developed in estate agent cases, as to entitlement to commission, which are normally regarded as part of agency law and are relied on also by agents who have greater powers to bind their principals. They may sometimes hold money (e.g. deposits) for their principals. The rules applicable to the internal relationship between principal and agent will therefore apply as appropriate, and for this reason such persons should certainly be treated in a work on agency even though they lack the external powers of the agent. It is an advantage of the formulation of basic agency principle in Article 1, which selects the internal relationship between principal and agent as a distinguishing feature of agency, that it can be taken to cover such persons. Canvassing agents are persons to whom the internal parts of agency law may apply, but who, because of the limited nature of their external powers to affect their principals' legal positions, are not agents in the full sense of the word. They may therefore be said to provide an example of "incomplete agency"."

Lord Falconer for UBS submitted that the corrupt arrangement between UBS and Value Partners identified by the judge exhibited none of the settled characteristics of agency. Thus:

While generally supporting the judge's analysis, Mr Tim Lord QC for KWL submitted that, leaving aside the fact that it was corrupt, the arrangement fell squarely within the fourth of the Bowstead and Reynolds categories, being a form of sales or canvassing agency akin to that of an estate agent for a vendor of real property. He submitted that an agent of that kind need not have authority to affect the principal's relations with third parties, and that fiduciary duty was not an essential prerequisite of agency. He submitted with real force that, if the only reason why the arrangement was not an agency was because it constituted dishonest assistance by UBS in an abuse by Value Partners of its fiduciary relationship with KWL, then this could hardly be a sound reason for exonerating UBS from responsibility for the bribe paid by Value Partners which would naturally flow from the recognition of an agency relationship.

Lord Falconer referred us to four authorities on this question. We need say little about the first, Garnac Grain Company Incorporated v HMF Faure & Fairclough Limited [1968] AC 1130 (Note). It merely reinforces the judge's conclusion that the question whether or not a particular relationship is that of agency depends upon what the parties have in substance agreed, rather than the label which they choose to place on it.

Branwhite v Worcester Works Finance Limited [1969] 1 AC 552 was a heavily fact-based majority decision of the House of Lords about the relationship between a car dealer and a finance company in relation to a hire-purchase transaction. Apart from the affirmation by Lord Wilberforce at page 587 of the principle which we have described as to be found in the Garnac case, it is difficult to derive any clear statements of legal principle from the speeches, save perhaps for these:

Plevin v Paragon Personal Finance Limited [2014] 1 WLR 4222 was a decision of the Supreme Court (in part) about the meaning of "on behalf of" in section 140 A(1)(c) of the Consumer Credit Act 1974, in connection with the omission by a broker to disclose to its client the amount of commission received by the provider of payment protection insurance, which it shared between itself and the lender, Paragon. The question was whether the broker was acting "on behalf of" the lender. In rejecting the Court of Appeal's analysis that the broker was so acting, the Supreme Court found that the broker was the agent of its client Mrs Plevin. Paragraph [33] reinforces the warning to be found in the Branwhite case against forcing into an agency analysis a relationship better explained in some other way, in particular where the supposed agent is already an agent of another party to the contemplated transaction.

There is an extended analysis of the legal principles relevant to the question whether a business person introducing transactions to a potential lender may be regarded as the lender's agent, in Tonto Home Loans Australia PTY Limited v Tavares & Others [2011] NSWCA 389, a decision of the New South Wales Court of Appeal, at [170] to [195], in the judgment of Allsop P. After quoting in full the extract from paragraph 1-001 of Bowstead, the judge continues, at [177], as follows:

"[177] These expressions of the central characteristics of the relationship reveal the closeness of identity that is required for the relationship to exist. Not every independent contractor performing a task for, or for the benefit of, a party will be an agent, and so identified as it, or as representing it, and its interests. Agency is a consensual relationship, generally (if not always) bearing a fiduciary character, in which by its terms A acts on behalf of (and in the interests of) P and with a necessary degree of control requisite for the purpose of the role. Central is the conception of identity or representation of the principal: Colonial Mutual at 48–9; Seavey op cit at 859. Examples and contexts may be infinite, and any arrangement must be understood and characterised by reference to its legal terms in context. In McKenzie v McDonald [1927] VLR 134 at 144 Dixon AJ, in saying that not every agent stands as a fiduciary, was recognising that the word "agent" is used in many senses and is apt to mislead, citing Kennedy v De Trafford at 188. That is, however, no more than to say that the word "agent" has a potentially wide and varying meaning in life and business and that, on some occasions, the business description will be given to someone who is not a fiduciary. See also Hospital Products Ltd v United States Surgical Corporation (1984) 156 CLR 41 at 71–2; BC8400480 (per Gibbs CJ), compare at 96–7 (per Mason J), Boardman v Phipps [1967] 2 AC 46 at 127; [1966] 3 All ER 721, F E Dowrick, "The Relationship of Principal and Agent" (1954) 17 Modern Law Review 24 and R P Meagher, J D Heydon and M J Leeming (eds) Meagher, Gummow and Lehane's Equity: Doctrines and Remedies (4th ed, LexisNexis, 2002) at 191–192 [5–195]. It is sufficient to recognise that the essential characteristic is that one party (A) acts on the other's (P's) behalf, and that this will generally be in circumstances of a requirement or duty not to act otherwise than in the interests of P in the performance of the consensual arrangement. Bowstead and Reynolds on Agency, the Restatement and Seavey op cit at 863 include in the conception of agency the characteristic of fiduciary duty. The duty will, of course, conform with the extent and scope of the agency and thus be of potentially varied content, recognising that context (in particular, perhaps, a market or commercial context) may attenuate the rigour or content of the fiduciary duty: Birtchnell v Equity Trustees, Executors and Agency Co Ltd (1929) 42 CLR 384 at 408; BC2900019; Re Goldcorp Exchange Ltd [1995] 1 AC 74 at 98; Meagher, Gummow and Lehane (4th ed) at 161–2 [5–010]); Finn J in South Sydney v News at [136], and in his text Fiduciary Obligations (LawBook Co, 1977) at 201. The necessary good faith implicit in a fiduciary character in the relationship reflects the character of identity or representation that the relationship essentially carries."

At [178] the judge noted that this fully accommodated Bowstead's paragraph (4) so as to apply, where appropriate, to canvassers and those seeking to bring business to another party, where the essential characteristics of agency were shown to be present.

Lord Falconer sought to derive from the Tonto case a principle that a relationship could never be identified as one of agency if none of the main characteristics, namely authority to affect the principal's relationships with third parties, fiduciary duty or control by the principal, was present. We would not be minded to go quite that far, but the absence of any of these main characteristics must nonetheless be a significant pointer away from the characterisation of a particular relationship as one of agency, even though there may be rare exceptions.

Mr Lord took us to Halton International Inc v Guernroy Limited [2005] EWHC 1968 (Ch) per Patten J at [138]-[9], and to Tigris International NV v China Southern Airlines Company Limited [2014] EWCA Civ 1649, per Clarke LJ at [155], in support of his submission that the existence of a fiduciary duty was by no means an essential characteristic of agency. We agree. There are no doubt many forms of non-fiduciary agency, just as there are forms of fiduciary agency in which the agent has no authority to affect the principal's relations with third parties.

Neither Bowstead and Reynolds, in the passages quoted, nor any of the authorities cited, was concerned with anything like the corrupt arrangement which the judge found to have existed between UBS and Value Partners. At its core was an understanding that Value Partners would abuse its fiduciary duties to its clients, and in particular to KWL, for the purpose of bringing them captive to UBS as counterparties in STCDO transactions, regardless of their interests, and that UBS would, behind those clients' backs, assist Value Partners in doing so, in both cases for the mutual benefit of Value Partners and UBS. At [701] the judge described this as "the whole point of [the] arrangement". The active participation in the arrangement by UBS was, as Lord Falconer frankly acknowledged, a form of dishonest assistance to Value Partners in the abuse of its fiduciary duties as agent for KWL. It was dishonest because it was both deliberate (i.e. carried out in the knowledge that Value Partners was abusing its fiduciary duty to KWL) and secretive, in the sense that UBS knew that KWL was ignorant of the corrupt arrangement. Using the language of Lord Hutton in Barlow Clowes International ltd v Eurotrust International Ltd [2005] UKPC 37, [2006] 1 WLR 1476, at [16], Mr Bracy, and therefore UBS, was aware that those elements of the corrupt arrangement were such as to make participation in it by UBS transgress ordinary standards of honest behaviour.

We see no reason to force that corrupt arrangement into an agency analysis for any purpose, let alone the purpose of deciding whether UBS is legally responsible for the bribe paid by Value Partners to Mr Heininger. Our reasons follow.

First, Value Partners was, and was known by UBS to be, the fiduciary agent of its captive clients before the corrupt arrangement with UBS was made. It is therefore a case, like the Plevin case, where a pre-existing agency relationship with another party is hostile to, albeit not necessarily irreconcilable with, an agency relationship with another party seeking to transact with the intermediary's pre-existing principal.

Secondly, the substance of the corrupt arrangement was that UBS would secretly assist Value Partners in abusing its pre-existing fiduciary relationship with its captive clients, including KWL. This could not be achieved either by UBS holding out Value Partners as its agent, or by Value Partners asserting an agency for UBS in dealings with KWL. On the contrary, the arrangement would only achieve its intended results if, from start to finish, Value Partners purported in its dealings with its captive clients to act loyally on their behalf.

Thirdly, the arrangement did not involve, or authorise, Value Partners to affect legal relations between UBS and any third party, even by the making of representations, or the receipt of information, as sometimes occurs within the context of a canvassing agency or an estate agency.

Fourthly, the judge did not find, and in our view the facts which he did find do not admit, the existence of a fiduciary duty by Value Partners to UBS. Each of the participants in the corrupt arrangement was pursuing its own interests, for its financial benefit. Neither was paying or otherwise remunerating the other.

We have not in our analysis ignored the submission by Mr Lord that, if this relationship fell short of agency only because of its corruption and dishonesty, this should not adversely affect the judge's finding that UBS was responsible for the bribe paid by Value Partners to Mr Heininger. Indeed it is largely, although not entirely, because of the corruption and dishonesty inherent in the arrangement between Value Partners and UBS that we consider that it did not amount to an agency: see our first and second reasons above. But we consider that this submission is better addressed later in the issue by issue analysis under which this appeal has been argued.

Nor are we unmindful of certain similarities between the role of Value Partners under the corrupt arrangement, and the role of a canvassing, introducing or estate agent. Nonetheless, in our view the differences which separate this particular relationship from those familiar business structures easily outweigh the similarities. In particular, the fact that Value Partners was already a fiduciary agent of its captive clients, including KWL, that it was pursuing its own interests side by side with those of UBS, and that the arrangement was corrupt and dishonest, places it in a separate category of its own. Of course it was part of the arrangement that UBS requested Value Partners to achieve a certain result. But, as is pointed out in the passage cited above from the Tonto case, this of itself is insufficient to lead to a categorisation of agency.

If we had found the corrupt relationship to be one of agency we would have affirmed the judge's view that the bribing of Mr Heininger fell within the scope of it. This is because, as the judge said, the bribe was paid for the purpose of assisting in the bringing about of the objective of the corrupt arrangement, namely bringing KWL as a captive client of Value Partners into an STCDO with UBS, regardless of KWL's best interests.

Nonetheless, for those reasons we would reverse the judge's conclusion on Issue 1.

The only basis upon which, at trial, KWL sought to make UBS legally responsible for the bribe paid to Mr Heininger was by means of the agency argument, which the judge accepted, but which we have rejected.

A broader basis for making UBS responsible for the bribe, so as to make the Balaba STCDO voidable at KWL's election, was canvassed but not pursued at trial. It was based on the following dictum of Robert Goff LJ in The Ocean Frost [1986] 1 AC 717 at 745C:

"I wish to reserve the question whether a party to a contract induced by the bribery of his servant by a stranger, or indeed a party to a contract induced by the fraud of a stranger, should not be entitled to rescind the contract on the discovery of the bribery or the fraud, on the basis that it would be inequitable for the other party, though innocent, to hold him to a contract so procured."

The point was not pursued at first instance because KWL anticipated that the judge would be inclined to follow the rejection of Robert Goff LJ's analysis by Andrew Smith J (again obiter) in Donegal International Limited v Zambia [2007] EWHC 197 (Comm) [2007] 1 Lloyd's Law Reports 397, at [496]. Nonetheless KWL gave notice that they might pursue the argument on appeal.

In this court counsel on both sides were substantially agreed that the question whether it would be inequitable for a party to resist rescission of a contract procured by a bribe of the other party's agent would depend upon whether the first party's conscience was sufficiently affected by the payment of the bribe. On this appeal Mr Lord did not pursue with any vigour a submission that wherever a contract is procured by payment of a bribe, the other party's conscience is inevitably affected, however innocent that party might be. Rather, he submitted that where (as in the present case) a party dishonestly assists the other party's agent in the abuse of its fiduciary duty to that other party and a contract ensues then, if that abuse of fiduciary duty included the payment of a bribe, the conscience of the dishonest assister will be affected by it even if it was unaware that a bribe had been paid.

Lord Falconer's main submission on this issue was that, if the answer to the question depended upon whether UBS's conscience was affected by payment of the bribe, then the judge had not been asked to find facts, and had not done so, on the basis of which that question could be determined on appeal. His fall-back submission, developed mainly as part of his argument under Issue 3, was that a contracting party's conscience could not be affected by some inequitable conduct on the part of the other party's agent (including the payment of the bribe) unless the first party knew of it, knowledge for that purpose being actual knowledge or wilful blindness ("Nelsonian knowledge"), which amounts to the same thing in law. Since the judge had found that UBS did not know that Value Partners had bribed Mr Heininger, that was the end of the matter.

We will address Lord Falconer's fall-back submission first. He deployed sufficient authority to show that, generally, the relevant criterion in this context is knowledge rather than, for example, notice, negligence, or knowledge of some other impropriety: see for example (but on very different facts) Ultraframe (UK) Limited v Fielding & Ors [2005] EWHC 1638 (Ch) per Lewison J, at [1500]-[1507], in the context of liability for dishonest assistance.

More to the point is the following dictum of Millett J in Logicrose Limited v Southend United Football Club Limited [1988] 1 WLR 1256. The football club had granted a licence to Logicrose to use its land for the holding of a market, unaware that its director and principal (but nominee) shareholder had required Logicrose to make a £70,000 payment to an offshore company under his control. The director had also concealed adverse advice from the football club's solicitors about the wisdom of granting the licence. On learning of the payment, the football club rescinded the licence. At page 1261-2 Millett J said this:

"In order to obtain the payment he committed S.U.F.C. to a transaction which, unknown to them, their own solicitors strongly advised against. S.U.F.C. have been deprived of the right to decide for themselves or by a disinterested agent whether to accept or reject that advice. That in itself is not sufficient to entitle S.U.F.C. to a fresh opportunity, as against the plaintiffs, to consider whether it is in their interests to affirm the transaction. For this purpose they must establish that they were deprived of the disinterested advice of their agent by or at least to the knowledge of the plaintiffs. Is this condition satisfied? I have no doubt that it is. It is, of course, immaterial whether the initiative for the agent's taking an interest of his own came from the agent himself or from the other party to the transaction. It must also be immaterial whether the other party provided it directly or knowingly assisted the agent to obtain it, for example by diverting to himself or an associate a payment intended for his principal. In all the reported cases, the other party has provided it himself and has been fully aware of the agent's personal interest. There is, accordingly, no direct authority on the degree of knowledge which he must possess of the existence of the agent's personal interest. With one reservation to which I shall come in a moment, however, and which goes only to the facts of which knowledge must be proved, I accept the submission made on behalf of the plaintiffs that nothing less than actual knowledge or wilful blindness will suffice. In particular, constructive notice will not do. Parties to negotiations do not owe each other a duty to act reasonably, but only to act honestly. In the present context, the principal's right is a right to rescind for fraud, not negligence. There is in my judgment a close parallel with the cases on knowing assistance in a breach of trust. The same facts may give rise to different remedies, and as the present case demonstrates it will often be impossible to distinguish between the payment of a bribe or secret commission properly so-called and the diversion of the principal's money into the agent's pocket. There cannot in truth be any real difference between the secret payment to the agent of a sum additional to the purchase price and the payment to him of part of the purchase price of which his principal is unaware. In my judgment, the difference between the two lines of authority (that is to say the "bribery" cases and the "knowing assistance" cases) lies not in the factual background but in the remedy sought; and the state of mind necessary to make the other party liable ought to be the same whether the claim is for an account of the money which he helped the agent to misappropriate or rescission of the transaction itself.

My one reservation, which I make for the sake of completeness, is this. It is clear that, where one party to a transaction takes what Collins L.J. described as "the hazardous course" of making a payment for the personal benefit of the other's agent, and does not disclose it to the principal, he cannot afterwards defend the transaction by claiming that he believed the agent to be an honest man who would disclose it himself: Grant v. Gold Exploration and Development Syndicate Ltd. [1900] 1 Q.B. 233, 249-250. Where, therefore, knowing that the agent has an interest of his own he does not himself disclose it to the other party, then in the words of Collins L.J., at p. 249: "he must at least accept the risk of the agent's not doing so." In my judgment, the converse must equally apply: if a man deals secretly with another's agent behind the back of his principal, knowing that the agent intends to conceal the dealing from his principal and that he may be intending to obtain some private advantage for himself, he takes the risk that he does intend to do so. The two are only different aspects of the same general principle, expressed in varying terms and contexts but always forcibly and to the same effect: "any surreptitious dealing between one principal and the agent of the other principal is a fraud on such other principal, cognizable in this court": see the Panama case, L.R. 10 Ch.App. 515, 526 per James L.J.; and "the real evil is not the payment of money, but the secrecy attending it": Shipway v. Broadwood [1899] 1 Q.B. 369, 373 per Chitty L.J."

Millett J's analysis broadly confirms the general rule that, for the conscience of the contracting party against whom rescission is sought to be affected, he must know about the bribe or other breach of fiduciary duty committed by his counterparty's agent. His "one reservation" is we think obiter, since he later concluded that Logicrose's director and principal shareholder knew perfectly well both of the relevant facts about the payment and of the intention that it be concealed from the football club's board. But the question remains whether Millet J's "one reservation" is correct in law as a limited exception to the requirement for actual knowledge and, if so, whether it is applicable to the present case.

In each of the three cases to which Millet J refers under his "one reservation" heading, namely Grant v Gold Exploration and Development Syndicate Limited [1900] 1 QB 233, Panama and South Pacific Telegraph Co. v India Rubber, Gutta Percha, and Telegraph Works Co. (1875) LR 10 Ch App. 515 and Shipway v Broadwood [1899] 1 QB 369, the surreptitious dealings in question consisted of or included the payment of a sum of money by the contracting party to the other contracting party's agent. There are also dicta by Clarke LJ in Tigris International NV v China Southern Airlines Company Limited [2014] EWCA Civ 1649, at paragraphs [175]-[178], that the term "surreptitious dealings" as used in those earlier cases was mainly concerned with payments of money, although that was a case where the party resisting rescission was found by the trial judge to be wholly innocent of any fraud or bad faith. But the principle is expressed in perfectly general terms, both by James LJ's reference to "any surreptitious dealing between one principal and the agent of the other principal" and Millett J's dictum: "If a man deals secretly with another's agent behind the back of his principal…", and by Chitty LJ in saying "the real evil is not the payment of money, but the secrecy attending it".

The mischief which the principle is aimed at preventing is the secret deprivation of the principal of the disinterested advice which he is entitled to expect from his fiduciary. The principal thinks he is getting the loyal and disinterested advice of his fiduciary when in truth he is not. This abuse may be achieved by a secret payment to the fiduciary by the other party to the contemplated transaction, but this is not the only way in which it can be achieved. The fiduciary may be disabled from giving disinterested advice by a multitude of surreptitious means. In the present case, UBS well knew of, and both secretly and dishonestly assisted, a wholesale abuse by Value Partners of its fiduciary duty to KWL, in bringing KWL to the STCDO as a counterparty with UBS regardless of KWL's best interests. The judge found that, although UBS made no payment to Value Partners for that corrupt purpose, it had "at least a good idea" of what Value Partners was being paid in connection with the transaction, namely "a very substantial fee which bore no real relationship to the services or advice that it had actually provided". UBS also knew that Mr Senf and Mr Blatz of Value Partners had been anxious to keep Dr Schirmer in the dark about the premium being earned from the deal, "and must therefore have realised that it was because of their relationship with Mr Heininger that they were able to extract such a high fee" [405]. UBS actively assisted Value Partners in keeping Dr Schirmer in the dark about the amount of the premium to be derived from the proposed transactions [314]-[315].

In our judgment, the general principle to be derived from the Logicrose case and the earlier cases cited by Millett J in the passage quoted above, which is applicable to the present case, is as follows. Where a party to an intended transaction deals with the other party's agent secretly and behind his back, and dishonestly assists that agent to abuse his fiduciary duties to the other party so as to bring that transaction about, then the first party's conscience may be affected not merely by the particular form of abuse by the agent of which it actually knew, but also by any other abuse which the agent chose to employ to bring about the transaction with the first party.

On the findings made by the judge we consider that this principle applies so as to affect UBS's conscience in the present case. The judge described the abuse of Value Partners' fiduciary duty to KWL, for the purpose of bringing KWL into the Balaba STCDO, as the "whole point" of the corrupt arrangement [701]. There is a useful analogy to be had with the question whether the bribe was, as the judge found, within the scope of Value Partners' agency for UBS. He said, at [617], that the bribe was paid for the very purpose of directing STCDO business to UBS from Value Partners' clients (here KWL), regardless whether it was in the clients' interests. The fact that in our view the corrupt arrangement was not in law an agency does not detract from the conclusion that the bribing of Mr Heininger was within the scope, or "point" of it.

As is well known, the essence of a fiduciary obligation is a duty of undivided loyalty to the person to whom that obligation is owed. UBS, having dishonestly assisted KWL's fiduciary agent to abuse its duty of loyalty in order to procure a contract by illegitimate means cannot in our view be heard to say that its conscience is clear if that agent, which was known to be dishonest, includes as part of its course of abuse, other illegitimate means such as the payment of a bribe to a decision-maker within the management structure of its principal.

For that reason, and subject only to Lord Falconer's primary submission under Issue 2, that the requisite facts were not found by the judge, we would conclude that the conscience of UBS was affected by the bribe which Value Partners had paid to Mr Heininger, so that it would be inequitable for UBS to resist rescission of the Balaba STCDO on that ground.

There is undoubtedly some force in Lord Falconer's submission that it is too late for KWL to advance this alternative basis for making UBS, in equity at least, responsible for the bribing of Mr Heininger for the first time in the Court of Appeal, since it plainly was not advanced before the judge. Nonetheless the facts about the corrupt arrangement between UBS and Value Partners and its causative significance in relation to KWL's entry into the STCDOs were the subject of intense investigation and fact-finding by the judge at the trial. It is difficult to imagine, and Lord Falconer did not explain, what further forensic analysis or fact-finding might have been needed for the purpose of deciding whether, apart from agency, the conscience of UBS was sufficiently affected by the bribe paid to Mr Heininger to make it inequitable for UBS to resist rescission of the Balaba STCDO.

Once all relevant facts have been found, the question whether the bribing of Mr Heininger entitled KWL to rescind the Balaba STCDO becomes essentially a question of law. The legal aspects of Issue 2 were fully argued by counsel in this court, and accordingly we consider that there is no injustice caused to UBS by entertaining KWL's submission, as we have described it, under Issue 2 despite it having not been pursued in the court below.

We have read in draft Gloster LJ's dissenting judgment on this issue.  She objects that our approach imposes the duties of a trustee on a counterparty (UBS), so as to mis-apply the moral standards of the vicarage to a commercial transaction.  In our view there is nothing uncommercial at all in holding that UBS's conscience was affected by Value Partners' bribing of Mr Heininger, where (as is common ground) UBS not merely knew of, but dishonestly assisted Value Partners in, its abuse of its fiduciary duties to KWL so as to bring a transaction about and the bribe was paid for that purpose.   As we have said, the underlying principle which the law seeks to maintain, in commercial dealings as much as elsewhere, is that a contracting party is entitled to the undivided loyalty of its fiduciary adviser, so that a counterparty which deliberately and dishonestly sets out to undermine that fiduciary relationship in relation to an intended transaction cannot expect to be able to resist a claim by the counterparty for rescission, if the transaction is brought about by the fiduciary's misconduct, and that misconduct included the payment of a bribe.

This is not a case where the contracting party merely harboured suspicions about the fiduciary's conduct, where the Logicrose requirement for knowledge operates as a salutary restraint against rendering contracts unduly vulnerable by the intervention of equity.  Provided that contracting parties act honestly, they will not be affected by what they do not know (provided they do not turn a blind eye to the truth).  Here UBS set out dishonestly to assist Value Partners in the very course of abusive conduct which led to the making of the transaction and the payment of the bribe in order to facilitate it.

For the reasons given above, we would answer Issue 2 in the manner contended for by KWL.

The judge found that KWL was entitled to rescind the Balaba STCDO both because UBS was legally responsible for the bribe and (separately) because its corrupt arrangement with Value Partners made it aware of, and party to, Value Partners' breach of its fiduciary duty to KWL, being disabled from providing disinterested advice to its client by a conflict of interest. Applying the analysis of Millett J in the Logicrose case at [1998] 1 WLR 1256, pp 1260-1, he concluded that Value Partners was subject to that conflict of interest, that UBS both knew of and assisted in creating it, that KWL did not know of or consent to it, and that it had a sufficient connection with KWL's entry into the Balaba STCDO for KWL to have a right of rescission: see [621]-[641] and [697]-[708] of the judgment.

Both in this court and below, UBS resisted the claim for rescission based on conflict of interest on two main grounds:

In this court Lord Falconer sensibly acknowledged that the main building bricks for a right of rescission arising out of Value Partners' conflict of interest had been found by the judge and could not be challenged. Thus, he accepted that UBS had dishonestly assisted Value Partners in abusing its fiduciary duties to KWL and, in any event, knew of Value Partners' conflict of interest (although not of the bribe).

Lord Falconer founded his submission about KWL's knowledge upon the undoubted fact, found by the judge, that Mr Heininger knew at all material times that Value Partners was not providing disinterested advice to KWL. Rather, Mr Heininger and Value Partners were dishonestly collaborating in a scheme for their mutual enrichment at KWL's expense, and in complete disregard of KWL's best interests. At trial, UBS also asserted that Dr Schirmer knew that Value Partners was not providing disinterested advice to KWL before the Balaba STCDO was made.

As to Mr Heininger's knowledge, the judge concluded that it was not to be attributed to KWL, on the basis of the law on attribution as then set out in Bilta (UK) Limited v Nazir [2013] EWCA Civ 968 by the Court of Appeal, affirmed in the judgment of Lord Walker in the Hong Kong Court of Final Appeal in Moulin Global Eyecare Trading Limited v Commissioner of Inland Revenue [2014] HKCFA 22, at [106] and [113]. By contrast, he rejected the claim that Dr Schirmer knew that Value Partners was not providing disinterested advice, on the facts [640].

In this court UBS challenges both the judge's legal analysis of the attribution issue, and his negative factual finding about Dr Schirmer's knowledge.

It is convenient to deal with the factual questions first, namely Dr Schirmer's knowledge and the question whether the corrupt arrangement between UBS and Value Partners, and its consequential conflict of interest, had any causative effect on KWL's entry into the Balaba STCDO.

Dr Schirmer's participation in and knowledge about the matters in issue was subjected to intense forensic scrutiny during the trial, during which he gave oral evidence and was cross examined at length. The judge described him at [89] as "an unreliable and in important respects untruthful witness". He had been grossly negligent by not enquiring as to the net proceeds of the Balaba transaction [92], although he did not know what they were. He wrongly denied being invited to take part in a safari in South Africa organised by Value Partners to celebrate the completion of the Balaba transaction. He deliberately misled KWL's Supervisory Board in November 2006 about an investigation into allegedly inappropriate expenses and perks provided by a predecessor of Value Partners, Global Capital Finance, in one of which he had participated [94-98]. He cooperated with Mr Heininger in a sustained and dishonest cover-up which included the transactions in issue, in September 2009 [101]-[102]. Above all, he falsely denied having signed a letter of engagement of Value Partners dated 31 May 2006 which provided for Value Partners to receive any "interest rate advantage" to be earned by KWL from the proposed transaction in excess of €4.5 million [92].

Nonetheless the judge concluded that Dr Schirmer did not in fact appreciate the impropriety of the relationship between Value Partners and Mr Heininger (and therefore the fact that Value Partners was not providing disinterested fiduciary advice to KWL) before September 2006, i.e. after the conclusion of the Balaba STCDO.

It is worth quoting the judge's findings about this in full, from [105] of the judgment:

"105.

There is no sufficient basis to conclude, and no party suggested, that Dr Schirmer was aware of the monetary bribes paid to Mr Heininger, in connection with either the cross-border lease or the STCDOs, or of the fact that most of the premium generated by the STCDOs would be diverted to Value Partners. But I find that a time must have come when Dr Schirmer must have been aware that Value Partners or its predecessor had adopted a practice of providing extravagant and inappropriate perks to Mr Heininger which can only have meant that it was making lucrative profits from its role in connection with the cross-border leases and the STCDO transactions. That ought to have caused him to question whether Value Partners was giving disinterested advice to KWL. By the time of the 7 September 2006 board meeting it must have been apparent to him that something was wrong, and by the time of the expenses scandal in October 2006 he must have realised that the whole relationship with Value Partners was improper. From then on, for reasons which are not clear but which probably arose from a concern for his own position, he was prepared to collude with Mr Heininger in a cover up. The most likely explanation is that having begun to tell lies about the 2002 Concorde trip and knowing of the much more recent dealings with Value Partners, he felt unable to allow the truth to be known. However, although he could and should have asked some searching questions, I am not persuaded that Dr Schirmer in fact appreciated the impropriety of the relationship with Value Partners before September 2006."

UBS mounted a sustained and carefully focussed challenge to this finding, on appeal. Lord Falconer did not seek to go behind the judge's conclusion that Dr Schirmer knew nothing of the bribing of Mr Heininger or of the diversion of most of KWL's premium from the STCDOs to Value Partners. Nor is it suggested that Dr Schirmer (or for that matter Mr Heininger) knew anything of the corrupt arrangement between UBS and Value Partners. The challenge is only about the judge's conclusion as to the time when Dr Schirmer first became aware that there was an improper relationship between Mr Heininger and Value Partners. UBS's case on appeal is that Dr Schirmer knew this before the completion of the Balaba STCDO on 7 June 2006.

For this purpose Lord Falconer relied almost entirely on other findings by the judge of primary fact about Dr Schirmer, against which he submitted that the judge's conclusion on this one timing issue was arbitrary, unreasoned, and wrong.

In addition to the findings about Dr Schirmer summarised at paragraph 131 above Lord Falconer relied upon the following:

More generally Lord Falconer categorised Dr Schirmer's approach to the transactions as someone apparently determined not to make even the most basic enquiries (such as reading and if necessary having translated the key documents) and acting throughout as someone who did not want to know, i.e. as having Nelsonian knowledge.

Forcefully though these points were made, we have not been persuaded that they justify this court reversing the judge on an important question of primary fact about the knowledge of a person who gave evidence before him. Taken together they plainly constitute material upon which the judge could have found that Dr Schirmer knew that (or turned a blind eye to the fact that) his colleague Mr Heininger had an improper relationship with Value Partners, prior to the completion of the Balaba STCDO. But it is clear that this experienced judge had all those matters in mind when reaching his conclusion about the time when Dr Schirmer came to have that knowledge. He set out most of them in his detailed analysis of Dr Schirmer's testimony at [89]-[105], which concluded with the finding now challenged. Every one of them features somewhere in his detailed judgment.

There are cases in which facts found at trial or documents presented on appeal lead the appellate court to conclude that a particular finding of fact must have been wrong, all the more so where the appellant can point to matters relevant to that finding which the judge appeared to have ignored. But this comes nowhere near such a case. More generally, the judge displays a comprehensive and masterly understanding of the facts of this complicated matter, which only increases the respect which an appellate court ought to show for findings of fact about the knowledge of a person who gave evidence at trial, which cannot be shown to have been vitiated by any clear error of analysis or omission.

At the centre of Lord Falconer's submissions in relation to Issue 3 (and indeed underlying UBS's case on many of the issues raised in this appeal) was his presentation of what, at trial, he called (and the judge referred to as) "the bigger picture", which we must now describe. Lord Falconer said that the driving force behind KWL's entry into all three disastrous STCDOs was the criminal conspiracy between Mr Heininger and Value Partners to use KWL's entry into the STCDOs as the means of making for themselves large sums of money at KWL's risk and expense, and which manifested itself in the payment of the bribe and Value Partners' misappropriation of the bulk of the premiums. The corrupt arrangement between Value Partners and UBS played, he said, no causative part in that story. The STCDOs would have been entered into by KWL, and the bulk of the proceeds misappropriated, whether or not Mr Bracy and Value Partners had made their corrupt arrangement. Value Partners needed no incentive from UBS to abuse their fiduciary duties to KWL. Thus the creation of a conflict of interest in Value Partners by means of the corrupt arrangement with UBS was (although Lord Falconer did not use these precise words) no more than an irrelevant sideshow. Since UBS knew nothing of what Lord Falconer called "the big fraud" involving Mr Heininger and Value Partners, it would be unjust and inequitable to subject it to rescission, and therefore the consequential losses sustained under the hedging transactions which followed the STCDOs, when its participation in the corrupt arrangement, although regrettable, was of no causative effect in the disaster for KWL which then unfolded.

The judge did not accept this analysis. He described it, in his own words, and then rejected it, in [698]-]699] of the judgment in the following terms:

"698.

Broadly speaking the "bigger picture" on which UBS relies is that the real reason why this disastrous transaction took place was the corrupt relationship between Mr Heininger and Value Partners, of which Dr Schirmer had knowledge, not in the sense that he knew the detail of that relationship, but in the sense that he knew that the relationship was inappropriate; that this corrupt relationship was between three individuals who were all acting on behalf of KWL; that there was no proper oversight of them by KWL or its Supervisory Board; and that UBS should not be held responsible for the fact that it had the misfortune to deal with criminals whose interest in personal financial gain and not anything said or done by UBS was at all times the driving force behind their actions.

699.

I accept that there is something in this view of the matter, but I have found that in fact this "bigger picture" is not painted in quite such bright primary colours, or at any rate that the picture painted by UBS is not the whole canvas. Thus I do not accept that Dr Schirmer was actually aware of the impropriety of Mr Heininger's relationship with Value Partners before September 2006 even if he ought to have investigated this relationship more closely before then; I have found that Value Partners was also the agent of UBS as well as of KWL, and that the bribe paid to Mr Heininger was paid within the scope of that agency relationship as well as being a breach of Value Partners' duties owed to KWL; and that while the transaction was driven in part by the dishonest desire for personal gain of individuals who were for most purposes on the KWL side of the line, it was also driven by the corrupt relationship between Mr Bracy and Value Partners and the eagerness of UBS, including at the highest level, to conclude an unusual deal which despite its rejection by UBS's own internal control function was just too profitable for UBS to turn away."

The judge's clear conclusion in those two paragraphs that the corrupt arrangement between UBS and Value Partners was one of the drivers of KWL's entry into the STCDOs is reinforced by reference to other parts of the judgment. At [279] he quotes Mr Bracy as saying, in December 2006:

"As you know, this transaction was originated by and brought into the firm by me utilising as advisers Value Partners who were former colleagues of mine at CSFB. We collectively worked on the original transaction for KWL. "

At [280] the judge continued:

"This even goes so far as to claim that it was Mr Bracy who "utilised" Value Partners as his advisers to "bring [the transaction] into the firm" – a claim which, in my judgment, was substantially accurate."

At [701], addressing the submission that rescission would be unfair and disproportionate, the judge said:

"Indeed the whole point of Mr Bracy's arrangement with Value Partners was that the latter should abuse its position as a trusted adviser in order to deliver its clients to UBS."

At [714] the judge said:

"The only issue is whether the bribe and the conflict of interest continued to affect the LBBW and Depfa transactions. In my judgment it is clear that they did. Although Mr Bracy and indeed Value Partners played a less prominent role in the negotiation of the LBBW and Depfa STCDOs, they continued to play some part, while other members of the UBS deal team such as Mr Czekalowski and Mr Sanz-Paris were fully involved. In any event, these later contracts were, as between UBS and KWL, all part of what was seen as a single transaction, with UBS taking the lead in finding banks which were prepared to act as intermediaries, stepping into a structure which had already been established. As KWL submits, it would make no sense for KWL to have a remedy against UBS in relation to the Balaba STCDO but not in relation to the other STCDOs, when in substance all four STCDOs formed part of a single transaction and the bribery and conflict of interest which gave rise to a remedy in the case of Balaba continued to operate in the case of the other STCDOs." (our underlining)

In our view a better view of the "bigger picture" than that presented by Lord Falconer and rejected in part by the judge is as follows. Value Partners was, in substance, the ringleader of a sophisticated fraud of which KWL was the intended victim, and from which Value Partners derived the largest benefit, namely the lion's share of the premiums payable to KWL under the STCDOs, less the €4.5 million received by KWL, the amounts used to fund the CDSs and the bribe of about $3 million paid to Mr Heininger. Value Partners' ill-gotten gains were only exceeded in amount by the profit booked by UBS for re-hedging (or re-selling) the credit protection constituted by the STCDOs in the market.

To achieve its criminal objective, Value Partners enlisted the assistance of Mr Heininger, who both committed KWL to the transaction and set up the structure which enabled Value Partners to misappropriate most of the premium, for which he was handsomely bribed.

UBS also dishonestly assisted Value Partners in securing the same objective, since a comprehensive abuse of its fiduciary obligations to KWL lay at the heart of the corrupt arrangement between Value Partners and UBS. Neither Mr Heininger nor UBS knew of the dishonest aspects of the participation of the other in the overall scheme.

Against that factual background, we turn to the central legal questions under Issue 3, namely attribution and causation. The law about attribution has moved on from where it was left by the Court of Appeal in the Bilta case and by the Hong Kong Final Court of Appeal in the Moulin Global case. It is now to be found in the decision of the Supreme Court in the Bilta case [2016] AC 1. It has not however substantially changed, since the Supreme Court specifically approved the lead given by Patten LJ in the Court of Appeal and by Lord Walker in the Moulin Global case: see [208] per Lords Toulson and Hodge, [22] per Lord Neuberger, [46] per Lord Mance and [86], per Lord Sumption).

It can now be taken as settled law that, where a company claims against a third party in respect of that person's involvement as an accessory to a breach of fiduciary duty by one of its directors, the state of mind of the director who was in breach of his fiduciary duty will not, as a matter of policy, be attributed to the company: see per Lords Toulson and Hodge at [207].

Thus, on the facts of the present case, if KWL was claiming against Value Partners in respect of its participation with Mr Heininger in the fraud which led to KWL's entry into, and losses from, the STCDOs, then Mr Heininger's undoubted knowledge of the aspects of that fraud in which he and Value Partners participated, including his knowledge that Value Partners was not delivering disinterested advice to KWL, could not be attributed to KWL, as the victim of the fraud in which both Mr Heininger and Value Partners participated. Nor in our view would it matter who, as between Mr Heininger and Value Partners, was the ringleader and who was the accomplice. The policy basis upon which attribution to KWL of Mr Heininger's knowledge is refused is because "it would be absurd and unjust to permit a fraudulent director or employee to be able to use his own serious breach of duty to his corporate employer as a defence": see per Lord Walker in the Moulin Global case at [131]. That absurdity and injustice is fully applicable in a case where the defendant to the claim is an accomplice of, or a co-conspirator with, the fraudulent director: see again per Lords Toulson and Hodge at [207] of Bilta, where they equate conspiracy and dishonest assistance as an accessory within the same principle.

The present claims do not, as Lord Falconer accurately submitted, fall squarely within that analysis. UBS is claiming against KWL under the Balaba STCDO, and KWL is claiming rescission of it. For the purposes of testing KWL's claim for rescission, the case may fairly be described as a claim by the company (KWL) against a third party (UBS). But the question of attribution relates to the knowledge of Mr Heininger rather than the knowledge of Value Partners. UBS were undoubtedly accomplices of Value Partners in the abuse of its fiduciary duties to KWL, but UBS did not know of what Lord Falconer calls "the big fraud" (i.e. the bribe and misappropriation of the premium) and was not, in that sense, a direct accomplice of Mr Heininger. Rather, they were each dishonestly involved in conspiring with or in assisting Value Partners, without knowing the precise parts of the other's dishonest participation.

Thus the legal question is whether the rule against attribution which applies where the company is the victim of a breach of fiduciary duty in which the third party defendant is an accomplice also applies in the present context, where the director and the third party are both accomplices of Value Partners, albeit in different aspects of the fraud upon KWL.

In our view the principle which prevents attribution does apply to this case. The question is best addressed by asking whether the underlying policy ground for refusing attribution is equally applicable, and in our view it is. It is no less unjust or absurd to attribute to the corporate victim of a sophisticated fraud the knowledge of one of its participant directors in a claim against another, different, participant in that fraud than it would be in connection with a claim against the ringleader (here Value Partners) or against any accomplice of the director. In all those cases the company is the victim of the fraud and it would be absurd to attribute to the company knowledge of any aspect of that fraud which is held by one of its directors, for the purpose of constructing an entirely artificial theory that the company is to be taken as having consented to it.

This must be a fortiori the case where, as the judge found, the corrupt arrangement between Value Partners and UBS remained one of the driving forces behind KWL's disastrous entry into the STCDOs, from start to finish.

Thus, while Lord Falconer's skilful submission may demonstrate that the present case does not, for attribution purposes, fall squarely within any of the categories described in the Bilta and Moulin Global cases, we are satisfied that the principle which those cases establish, and in particular the underlying policy rationale from which that principle takes its force, is equally applicable to the present, albeit more complicated, facts. We would therefore concur with the judge's rejection of the defence of UBS, based on attribution.

Likewise, we consider that Lord Falconer's submission based upon a lack of any causative link between the corrupt arrangement between Value Partners and UBS and KWL's entry into the STCDOs is wrong, for two reasons. The first is that the judge found, as a fact, that the corrupt arrangement remained a driving force behind KWL's entry into all four STCDO structures, from start to finish. Causation, if it were necessary, is therefore established as a fact. It is in that context no answer for UBS to say that the "but for" causation test is not satisfied. It may be that, even if there had been no corrupt arrangement, KWL would still have entered into the STCDOs, driven there by the fraudulent design shared by Mr Heininger and Value Partners. But where, as here, there are concurrent causes, it is no defence to say that the transaction would or might have proceeded even if there had only been one cause, for which the defendant is not responsible.

The second reason is that causation is not a requirement of the equitable right to rescind, where the decision-making of the party seeking rescission has been undermined by a conflict of interest on the part of its fiduciary adviser which the counterparty resisting rescission either assisted, or of which it knew. As Millett J put it in the Logicrose case at [1988] 1 WLR 1256, 1260:

"It is immaterial whether the agent's mind has been affected or whether the principal has suffered any loss as a result: "the safety of mankind requires that no agent shall be able to put his principal to the danger of such an enquiry as that:"… the principal having been deprived by the other party to the transaction of the disinterested advice of his agent, is entitled to a further opportunity to consider whether it is in his interests to affirm it."

For those reasons, we would affirm the judge's conclusion under Issue 3. Regardless of whether UBS is or is not responsible for, or whether its conscience is affected by, the bribe paid to Mr Heininger, its dishonest assistance in Value Partners' abuse of its fiduciary duty to KWL is, separately and on its own, sufficient to entitle KWL to rescind the Balaba STCDO, subject to any other defences, which we analyse under Issue 4.

A claim (at least of this kind) for rescission is a claim for an equitable remedy. Where a prima facie right to rescission is demonstrated, the court nonetheless retains what is traditionally called a discretion to refuse it where it would be unfair or disproportionate, or to afford some other more suitable remedy, such as equitable compensation or an account. Further, the claim is subject to the usual equitable defences including laches and lack of clean hands.

In the present case UBS defended the claim to rescission of the Balaba STCDO both because it would be disproportionate and unfair, and because, so it alleged, KWL did not come to the court seeking equity with clean hands.

The judge rejected the clean hands defence, and exercised his discretion in favour of rescission, mainly because of his view that KWL's entry into the Balaba STCDO was in part driven by the corrupt arrangement between UBS and Value Partners [698]-[701]. He also relied upon the fact that UBS had been legally responsible for the bribe [701]. We have reached a slightly different conclusion from the judge in that respect, to the effect that UBS's conscience was affected by the bribe, but not that it was in all respects liable in relation to it as a joint tortfeasor with Value Partners.

Lord Falconer advanced UBS's appeal in relation to this issue on the express basis that:

We will first address this issue strictly on the basis proposed by Lord Falconer. For that purpose we will assume, contrary to our finding under Issue 2, that the conscience of UBS was unaffected by the bribe, so that KWL's case for rescission depends entirely on UBS's dishonest assistance, by the corrupt arrangement, in Value Partners' breach of fiduciary duty. Since that differs from the basis upon which the judge exercised his discretion, we will approach the discretionary aspect of this issue as if we were exercising it afresh.

The undoubted discretion to refuse to give effect to a prima facie right to rescind in equity where to do so would be unfair or disproportionate was addressed by counsel by reference to two authorities, namely Hurstanger Limited v Wilson [2007] 1 WLR 2351 and Johnson v EBS Pensioner Trustees Limited [2002] Lloyd's Law Reports 309. In the Hurstanger case the defendants obtained a loan from the claimant through the services of a broker, to whom the claimant paid a commission. The terms of the loan agreement notified the defendants that a commission might be paid, but not its amount. At [45] Tuckey LJ treated the case as a half-way house between the payment of a secret commission and the payment of a fully disclosed commission. It was neither secret nor fully disclosed. Relying on the statement of principle in the Johnson case (which we will shortly describe) the Court of Appeal declined to order rescission of the loan agreement, because its terms were fair and (from the rigorous perspective of the Consumer Credit Act 1974), enforceable. Rather, the defendants received an order for an account in relation to the commission.

In the Johnson case the second defendant Mr O'Shea was the surety under a lease granted to his property development company, the reversion in which was assigned to his pension trustees who included the claimant Mr Johnson, a solicitor and partner in the firm which had arranged the lease. The development company had also received a loan arranged by Mr Johnson from clients of his firm. The lenders paid what was described as a modest "service charge" to the firm in connection with the loan, which Mr Johnson did not disclose to Mr O'Shea. He sought equitable rescission of his surety obligation by reason of that non-disclosure. The Court of Appeal granted him an account of the service charge, but refused him rescission. Dyson LJ pointed out, at [69], that, leaving aside the service charge, there was nothing remotely unfair about Mr O'Shea's surety obligation, which he had undertaken as an astute and experienced businessman.

Under the heading "Rescission" Dyson LJ continued as follows:

"76.

The next question is whether, once he found that there was a breach of the duty to disclose, the judge was obliged to hold that Mr Johnson was not entitled to enforce the surety covenant. Mr Darton submits that Mr O'Shea did not need to obtain an order rescinding the surety covenant: the abuse of confidence simply disentitled Mr Johnson as a matter of law from enforcing the covenant. Alternatively, if an order of rescission was required, the court either had no discretion to refuse it in a case of an abuse of confidence, or, if it had a discretion, it was obliged to make an order in such a case.

77.

I cannot accept Mr Darton's submission that a contract entered into following an abuse of confidence is unenforceable simply because it is "tainted" (to use Mr Darton's word) by the abuse. Such a contract is not illegal, nor is it analogous to an illegal contract. Until it has been set aside, it remains in being for the benefit of, and is enforceable by, both parties.

78.

In relation to rescission, in my view the judge was right to say (paragraph 46) that, whatever the position in relation to a claim to rescind based on misrepresentation, the right to rescission on grounds of undue influence, abuse of confidence or breach of fiduciary duty depends on the exercise of the discretion by the court to intervene in the enforcement of legal rights. I refer to the passage in the judgment of Millett LJ in Dunbar Bank Plc v Nadeem [1998] 3 All ER 876, 884 H-J which Mummery LJ has already cited.

79.

When exercising its equitable jurisdiction, the court considers what fairness requires not only when addressing the question of the precise form of relief, but also when considering whether the remedy should be granted at all. As Sir Donald Nicholls V-C said in Cheese v Thomas [1994] 1 WLR 129, 137C:

"It is important not to lose sight of the very foundation of the jurisdiction being invoked. As Lord Scarman observed in the Morgan case [1985] AC 686, a court in the exercise of this jurisdiction is a court of conscience. He noted, at p. 709:

"There is no precisely defined law setting limits to the equitable jurisdiction of a court to relieve against undue influence ... Definition is a poor instrument when used to determine whether a transaction is or is not unconscionable: this is a question which depends upon the particular facts of the case."

As with the jurisdiction to grant relief, so with the precise form of the relief to be granted, equity as a court of conscience will look at all the circumstances and do what fairness requires. Lord Wright adverted to this in Spence v Crawford [1939] 3 All ER 271, which was a misrepresentation case. He said regarding rescission and restitution, at page 288:

"The remedy is equitable. Its application is discretionary, and, where the remedy is applied, must be moulded in accordance with the exigencies of the particular case."

80.

In my judgment, the judge was entitled in the present case to hold that fairness required an order for an account of the service charge, but not rescission of the surety covenant. The transaction was fair. Mr O'Shea was an experienced business man who, the judge found at paragraph 19, "remained focused throughout on competitive rates on interest and would not have dealt with Churchers if it was not in his commercial interests to do so". He held (paragraph 27) that if the service charge had been disclosed to Mr O'Shea, it is possible that he would have sought a reduction in the rate of interest, but if this had been refused, he would still have gone ahead. There is no challenge to any of these findings. In these circumstances, the breach of duty by Mr Johnson was marginal."

Lord Falconer's submission that to grant rescission of the Balaba STCDO would be unfair and disproportionate may be summarised as follows:

Lord Falconer's first point was, of course, a re-iteration of the "bigger picture" argument presented to the judge and in this court, which both he and we have rejected. We have already quoted from the passages in [699] and [701] of the judgment (specifically addressed to this submission) in which the judge concludes both that the whole point of the corrupt arrangement between UBS and Value Partners was that Value Partners should abuse its position as a trusted advisor in order to deliver KWL as a counterparty with UBS in the STCDOs, and that KWL's entry into the Balaba STCDO was indeed driven, in part, by that corrupt arrangement.

In stark contrast with the Hurstanger and Johnson cases, the Balaba STCDO was neither a fair and ordinary transaction (as in Hurstanger) or a transaction entered into by KWL with its eyes open and on the basis of its wide experience in transactions of the relevant kind (as in Johnson). It was a highly unusual type of complex derivative, normally traded only between investment banks and similar financial institutions, undertaken by a water company wholly outside the confines of the business activities in which it was experienced. Furthermore, as the judge said, the whole purpose of the corrupt arrangement between Value Partners and UBS was to draw KWL into the transaction by depriving it of the loyal advice as to its best interests which it was entitled to receive from its financial advisers Value Partners. Both UBS and Value Partners entered into the corrupt arrangement for financial gain. For UBS the gain was the immediate booking of a profit from re-hedging or re-selling the credit protection provided by the STCDOs, which was (after allowing for payment of the premium to KWL) broadly equivalent to that premium. It was a risk-free profit save only for UBS's exposure to KWL's credit risk. In stark contrast, KWL was wholly exposed to downside risks broadly equivalent to those which UBS undertook by re-hedging, without recourse against anyone other than the Leipzig taxpayers, but for which it nonetheless received a broadly equivalent reward in the form of premium.

For those reasons, which broadly equate with those given by the judge (apart from UBS's responsibility for the bribe), we reject Lord Falconer's "unfair and disproportionate" submission, and would exercise our discretion in favour of the grant of rescission, regardless of any question whether UBS was liable in relation to the bribe or whether its conscience was affected by it.

We have however concluded that UBS's conscience was affected by the bribe. The discretion whether or not to grant equitable rescission is exercised by the court as a court of conscience: see the dictum of Sir Donald Nichols VC in Cheese v Thomas [1994] 1 WLR 129 at 137, cited by Dyson LJ in the passage quoted above from the Johnson case. It follows that, on our conclusion as to the answer to Issue 2, there is in substance no real distinction between the judge's attribution to UBS of legal responsibility for the bribe, and our view that UBS's conscience was affected by it, for the purpose of the exercise of this equitable discretion.

There remains the question whether KWL is barred from obtaining rescission by coming to equity without clean hands. In addressing this submission at trial, the judge directed himself as to the law by reference to the following passage in the judgment of Aitkens LJ in Royal Bank of Scotland Plc v Highland Financial Partners LP [2013] EWCA Civ 328, [2013] 1 CLC 596, at [158]-[159]:

"158.

There is no dispute that there exists in English law a defence to a claim for equitable relief, such as an injunction, which is based on the concept encapsulated in the equitable maxim "he who comes into equity must come with clean hands". Mr Nicholls accepted that the doctrine applies to a claim for an anti-suit injunction where the claim is based on an allegation that the defendant has started proceedings in a foreign jurisdiction in breach of contract because the claimant and defendant had agreed to an exclusive jurisdiction clause in favour of the English courts. It is clear from the speech of Lord Bingham in Donohue v Armco Inc that this defence is distinct from that of there being "strong reason" not to grant an anti-suit injunction.

159.

It was common ground that the scope of the application of the "unclean hands" doctrine is limited. To paraphrase the words of Lord Chief Baron Eyre in Dering v Earl of Winchelsea the misconduct or impropriety of the claimant must have "an immediate and necessary relation to the equity sued for". That limitation has been expressed in different ways over the years in cases and textbooks. Recently in Fiona Trust & Holding Corporation and others v Yuri Privalov and others Andrew Smith J noted that there are some authorities in which the court regarded attempts to mislead it as presenting good grounds for refusing equitable relief, not only where the purpose is to create a false case but also where it is to bolster the truth with fabricated evidence. But the cases noted by him were ones where the misconduct was by way of deception in the course of the very litigation directed to securing the equitable relief. Spry: Principles of Equitable Remedies, suggests that it must be shown that the claimant is seeking "to derive advantage from his dishonest conduct in so direct a manner that it is considered to be unjust to grant him relief". Ultimately in each case it is a matter of assessment by the judge, who has to examine all the relevant factors in the case before him to see if the misconduct of the claimant is sufficient to warrant a refusal of the relief sought."

Neither Lord Falconer nor Mr Lord criticised the judge's reliance upon this statement of principle, and we are content to adopt it. The key part of it, which dates back to the 18^th century, is that the misconduct or impropriety of the claimant must have "an immediate and necessary relation to the equity sued for", and that it must be shown that the claimant is seeking "to derive advantage from his dishonest conduct in so direct a manner that it is considered unjust to grant him relief". As the last five lines of the quoted passage show, this is one of those multi-factorial assessments to be conducted by the trial judge, with which an appellate court will be slow to intervene, unless the judge's conclusion was clearly wrong, or based upon some evident failure of analysis.

Lord Falconer based his submission at trial on two main limbs:

The judge rejected the first point because it was in his view a re-run of the "bigger picture" argument about causation which he had already rejected. We agree with the judge's analysis. We would add that, if Mr Heininger's knowledge of the fraud is not to be imputed to KWL as its victim, we find it difficult to understand how his fraud can sully KWL's hands for the purpose of the grant or refusal of rescission in equity.

The judge rejected Lord Falconer's second point for two reasons:

Taking those reasons in turn, our view about the outcome of Issue 2 means that the conscience of UBS is affected by the bribe, which played a part in inducing Mr Heininger to make his fraudulent misrepresentation to UBS about the bona fides of the transaction. For reasons already given, this must in the context of equitable relief be a question of conscience, rather than liability at common law for the bribe as a joint tortfeasor.

As to the judge's second reason, we deal with, and agree with, the judge's conclusion about causation in relation to UBS's losses on its hedging contracts, in our treatment of Issue 5 below. Moreover in our view Mr Heininger's lie was not even the effective cause of the making of the Balaba STCDO although it might be said that, but for the lie, the transaction might not have gone ahead. The effective cause was, or at least included, the corrupt arrangement between UBS and Value Partners. But even if the lie was an effective cause, it was perpetrated in pursuance of Mr Heininger and Value Partners' fraud on KWL, of which KWL was the victim. For the reasons given above we would not regard it as sullying KWL's hands.

Either of the reasons given by the judge for rejecting the clean hands defence is, in our view, sufficient to justify his conclusion. We consider that the clean hands defence fails, substantially for the reasons given by the judge. In any event, its assessment was pre-eminently a matter for him and no basis has been shown which would justify this court in departing from it. We would therefore dismiss UBS's appeal on this Issue.

UBS did not embark upon this litigation by claiming damages for deceit from KWL. Rather it amended (at the beginning of the trial) to plead such a claim in response to KWL's defence to enforcement of the Balaba STCDO, upon the basis that, if that transaction was rescinded, UBS would not be able to look to KWL to make good its liabilities under its hedging or re-selling transactions. As the judge said, at [677], the deceit claim was made by UBS in the alternative to its primary claim to recover the sum due pursuant to the Balaba STCDO.

UBS claimed that Mr Heininger had made a number of false statements to UBS about the contemplated transaction, first at a due diligence meeting on 30 May 2006, secondly in an Engagement Letter dated 24 May 2006 and thirdly in a Risk Disclosure Letter of the same date. The judge found that most of these statements, to the extent that they were untrue, were known at the time to be untrue by Mr Bracy (whose knowledge was, by common consent, to be attributed to UBS) so that their falsity availed UBS nothing. But he found that one untrue statement by Mr Heininger, namely that KWL was entering into the transaction in good faith and in accordance with its normal business activity, was not known by UBS to be untrue, at least at the time of the Balaba STCDO [688].

The deceit claim was put both in terms of breach of warranty and fraudulent misrepresentation. In relation to breach of warranty, the judge held that the basis for measuring loss would be the contractual measure of damages, namely that necessary to put UBS in the same position as it would have enjoyed if the warranty had been true. He found that, in those circumstances, UBS would still have entered into the Balaba STCDO and hedged or re-sold it into the market. UBS would still have been vulnerable to the rescission of the Balaba STCDO either because of bribery or conflict of interest, so that the falsity of the warranty caused UBS no loss [690].

The judge rejected the alternative claim in deceit for two reasons. The first was that, applying the tortious measure of damages (designed to put the defrauded party in the position which it would have enjoyed if the representation had not been made) it was likely that the absence of a good faith statement would have been a red flag sufficient to bring the Balaba STCDO (and its subsequent re-hedging) to a halt. But he held, at [692], that even in this event, the true cause of UBS's loss was not the making of the Balaba STCDO and its re-hedging but rather the rescission of the Balaba STCDO, which deprived UBS of what would otherwise have been a full recourse against KWL for its hedging liabilities. It is worth quoting the judge's concise reasoning in full:

"692.

However, I accept KWL's submission that the loss suffered by UBS on its hedging contracts was not caused by its entry into the Balaba STCDO. The real and effective cause of the loss was not that UBS entered into its hedging contracts or even that it entered into the Balaba STCDO. Rather it was that the Balaba STCDO is unenforceable by UBS as a result of matters for which UBS is responsible -- the bribery of Mr Heininger by UBS's agent Value Partners and/or its knowledge of the conflict of interest to which Value Partners was subject. Accordingly its claim for damages fails as a matter of causation. A similar analysis applies to the hedging contracts concluded by UBS in order to generate the premium and profit required for the LBBW and Depfa transactions. If those Back Swaps are valid and enforceable, UBS will have suffered no loss. If they have been rescinded for misrepresentation by UBS, an issue addressed below, it is UBS's own misrepresentation which is the cause of its loss on those hedging contracts."

The judge's second reason for ruling against UBS on its deceit claim was that Mr Heininger's fraudulent statement about the good faith of the transaction formed part of a set of routine statements and warranties which were all part and parcel of a transaction tainted both by the bribe, for which he regarded UBS as responsible, and by the conflict of interest. He said, at [693]:

"If the STCDO itself is voidable by reason of bribery and/or conflict of interest, the representations and warranties in these ancillary documents can have no independent life."

At [694] he continued:

"Similarly, if UBS knew that KWL was not being properly advised as to entry into the STCDO itself, it cannot have thought that it was receiving proper advice about ancillary agreements such as the Engagement Letter and the Risk Disclosure Letter."

In this court UBS challenged both limbs of the judge's reasoning for concluding that the deceit claim failed. While accepting that if UBS was responsible for the bribing of Mr Heininger, then the deceit claim was bound to fail, Lord Falconer submitted that if UBS was only responsible as a dishonest assister for depriving KWL of the fiduciary advice of Value Partners, then both limbs of the judge's analysis were wrong in law. As to the first he said that the ordinary policy reasons for limiting the effect of a tenuous chain of "but for" causation, such as the scope of duty limitation identified in the SAAMCO case [1997] AC 191 do not apply to a claim based in fraud: see Smith New Court Securities Limited v Citibank NA [1997] AC 254 at 266-7 and 279-80. As to the judge's second reason, Lord Falconer submitted that if he could trace a causative link between the fraudulent misrepresentation of Mr Heininger and UBS's entry into the re-hedging transactions, then it mattered not that the Balaba STCDO was tainted by some form of illegality, or that UBS had itself contributed to that taint. While warranties might stand or fall with the transaction itself, so as to cease to operate after rescission, the same could not be said of a fraudulent misrepresentation.

For his part Mr Lord for KWL supported both limbs of the judge's analysis. He made two additional submissions. The first was that UBS could not be heard to say that it was induced to make the hedging transactions by having given any credit to Mr Heininger's statement that the STCDOs were, from KWL's perspective, bona fide when UBS itself lied to both LBBW and Depfa in asserting that it knew of no reason to doubt the honesty of Value Partners or Mr Heininger. This was said to be all the more so since, by the time of the Depfa transaction, UBS knew of and had actively participated in covering up Mr Heininger's dishonesty in the course of the "letter for K" incident, described at paragraphs 51-55 above. Secondly he submitted that to acknowledge UBS's claim for damages for deceit would give rise to circuity of action because of KWL's equal and opposite claim for equitable compensation on account of UBS's dishonest assistance of Value Partners in its abuse of fiduciary duties to KWL.

We admit to some discomfort at the outset of our analysis of this issue, arising from the fact that it appears to have been common ground at trial, and continued to be on appeal, that KWL was vicariously liable for Mr Heininger's fraudulent assertion about the bona fides of the transaction. Both we and the judge have concluded that, because KWL was the victim of the transaction, Mr Heininger's knowledge of the fraudulent aspects of the means whereby it was committed to the Balaba STCDO is not to be attributed to KWL. Why, it might be asked, is a company vicariously liable for a fraudulent statement by its director in circumstances where the director's knowledge of the truth is not to be attributed to it? It may be that the law as to vicarious liability and attribution run in different channels, or that the full implications of the Supreme Court's decision in the Bilta case have yet to filter through into the ordinary law of tort. However that may be, we received no submissions on this point, and are constrained therefore to proceed upon the basis that KWL is vicariously liable for Mr Heininger's fraudulent misrepresentation to UBS.

We accept Lord Falconer's submission that policy-based limitations on the scope of liability for negligence are not to be applied to a claim in deceit. The deceiver is liable, on the tortious basis of analysis, for all the loss directly caused to the representee by the fraudulent misrepresentation, without limits derived from the law as to foreseeability or scope of duty. We also accept that a claim of this kind is not defeated merely because the transaction entered into in reliance upon the fraudulent misrepresentation is itself tainted by illegality or impropriety.

Nonetheless we are not persuaded that this undermines the first limb of the judge's reasoning, namely that Mr Heininger's fraudulent misrepresentation did not cause UBS's loss. We think that he was right, or at least entitled on the evidence, to conclude that the cause of UBS's loss was the vulnerability of the Balaba STCDO to rescission. Had the STCDO been valid and enforceable, UBS's re-hedging or re-selling of the credit protection which it provided in the market would have caused it no loss since, as the judge accepted as common ground, the STCDO liability of KWL (but for rescission) was broadly the same as UBS's hedging losses. As we have earlier explained, UBS's only risk, on the assumption of a valid Balaba STCDO, was the KWL credit risk, and it has not been suggested that this is sufficient to play any part in the assessment of causation and loss under this issue.

On the face of it, entry by UBS into the Balaba STCDO followed by re-hedging, even if induced in any real sense by Mr Heininger's fraudulent statement as to bona fides, was not something which exposed UBS to the risk of being left without recourse in the event that the hedges proved to be, as against the market, loss making. The propensity for such hedging losses to cause UBS loss has arisen entirely from the rescission of the Balaba STCDO which was intended to provide full protection to UBS against them.

The vulnerability of the Balaba STCDO to rescission at the election of KWL was in no sense a consequence of Mr Heininger's fraudulent statement about the bona fides of the transaction. It arose in part from UBS's dishonest assistance in Value Partners' abuse of its fiduciary duties to KWL, and in part from the bribe which, although UBS was not a joint tortfeasor in relation to it nonetheless affected its conscience, for the purposes of rescission in equity.

As the judge said, at [692], the position is even more stark in relation to the losses occasioned by UBS in re-hedging the LBBW and Depfa STCDOs. There its vulnerability to hedging losses was the result of its own fraudulent misrepresentations to LBBW and Depfa, as the result of which (as we explain below) both those Back Swaps have been rescinded.

While we agree with the judge's analysis about the lack of the requisite causation, it is in our view sufficient for us to say that nothing has been demonstrated on appeal which would enable us to depart from his analysis. Causation, in a complicated fact-intensive case such as this, is another of those mixed questions of fact and law, from which an appellate court will be slow to depart, unless persuaded that the judge was clearly wrong, or that he made an error of principle. We have not been so persuaded.

We are less confident that the judge's second ground for rejecting the deceit claim is well founded, viewing it as a claim in tort for deceit rather than a claim for breach of warranty. Similarly, although we can see some force in Mr Lord's submissions about the unreality of inducement, in the context of UBS's subsequent conduct in relation to the LBBW and Depfa transactions, we do not consider it necessary to reach any conclusions about them, and we are inclined to think, but without deciding, that the circuity of action analysis would not, if it mattered, be a sufficient basis for rejecting UBS's claim in deceit.

But the judge's first reason, based on a lack of causation, for rejecting the deceit claim is self-standing and sufficient, if not set aside, to dispose of this issue against UBS. In particular, that reason did not depend upon the judge's view that UBS was legally responsible for the bribe paid to Mr Heininger. Accordingly we would dismiss UBS's appeal under Issue 5.

The judge's decision that KWL was entitled to rescind the Balaba STCDO did not of itself (and without more) mean that KWL had any similar entitlement to rescission as against LBBW or Depfa in relation to the STCDOs ("Front Swaps") it made with them, even though all four STCDO structures were part of a single composite commercial transaction. By the beginning of the trial LBBW had succeeded in a claim to enforce its Front Swap against KWL in proceedings in Germany, although its judgment was subject to an appeal, stayed pending the outcome of these proceedings. Depfa initially sought to enforce its Front Swaps against KWL in these proceedings. UBS sought to enforce both of its Back Swaps against LBBW and Depfa.

KWL claimed to be able to rescind the Depfa Front Swaps, on substantially the same grounds as in relation to the Balaba STCDO. For reasons which do not now matter the judge rejected that claim, holding that, if it had mattered, the Depfa Front Swaps would have been enforceable against KWL. Following the emergence of the "Letter for K" episode during oral evidence at the trial, both LBBW and Depfa amended to claim rescission of the Back Swaps by reason of fraudulent misrepresentation by UBS, and the judge upheld these claims, granting rescission of all three Back Swaps on terms which included undertakings by LBBW and Depfa not to enforce the Front Swaps against KWL: see Issue 7 below. The result was that, on the judge's analysis, KWL had no liabilities to LBBW or Depfa on the Front Swaps. Nonetheless he addressed the (fully argued) issue whether KWL should be indemnified by UBS in relation to any such liabilities, and concluded that it should, because KWL's entry into the Front Swaps was caused both by the bribing of Mr Heininger, for which UBS was responsible, and separately because it was also caused by UBS's dishonest assistance in Value Partners' abuse of its fiduciary duties to KWL.

In this court UBS does not challenge in limine the judge's order that the Back Swaps be rescinded, but it does appeal the judge's decision to take an undertaking by LBBW and Depfa not to enforce the Front Swaps, submitting that, by various alternative methods, the Front Swaps should be made available for its benefit: see again Issue 7. The result is that the question whether KWL is entitled to an indemnity from UBS in respect of any liability under the Front Swaps remains live.

The judge dealt with this issue quite briefly, as can we. He had already determined that UBS was responsible for the bribe and that it had dishonestly assisted Value Partners by implementing the corrupt arrangement. He concluded that both bribery and what he called "knowing assistance" (now more accurately called dishonest assistance) sounded in damages or equitable compensation [712]-[713], so that the only issue was "whether the bribe and the conflict of interest continued to affect the LBBW and Depfa transactions" [714]. He continued:

This finding about causation is the only part of the judge's analysis of this issue which is challenged on appeal. The challenge consisted of a repetition of the "bigger picture" argument which the judge had already rejected, and which we have also rejected under Issues 3 and 4. Like the judge, we accept Lord Falconer's submission that, for this purpose (if not for rescission), KWL must establish causation: see Gwembe Valley Development Co. v Koshy (No 3) [2004] 1 BCLC 131, at [145]-[147]. But causation was established, separately in relation to the bribery and the dishonest assistance, for the reasons given by the judge, with which we agree. As in relation to Issue 2, Lord Falconer submitted that UBS did not have sufficient knowledge of the broad terms of the dishonest design, relying on Lewison J's dictum in the Ultraframe case, approved by this court in Goldtrail Travel Ltd v Aydin [2016] 1 BCLC 635, at [46]. For this purpose it is enough that UBS plainly knew of, and dishonestly assisted, the abuse by Value Partners of its fiduciary duty to KWL, even if it did not know of the bribe.

UBS did not suggest that anything happened between the making of the Balaba STCDO and the making of the Front Swaps to break the chain of causation linking them to the bribe and to the dishonest assistance, and we regard the judge's reasoning for concluding that there was no such break in the chain, in the passage quoted above, as unassailable. The Balaba STCDO and the Front Swaps were clearly all part of a single commercial transaction, even though separated in time, and involving different parties.

We would therefore dismiss UBS's appeal under Issue 6.

The judge found at [740] and [747] that UBS impliedly represented to Depfa and LBBW that:

The judge found that these representations were false [751]-[753].

The judge further found that the representations were fraudulent [754]-[772].

In relation to the no knowledge of dishonesty representation he found that Mr Bracy knew of Mr Heininger's dishonesty from the "letter for K episode" but that he knew of Value Partners' dishonesty before then. As he stated at [764]:

"…while this episode demonstrated the dishonesty of Value Partners, it was not the original source of Mr Bracy's knowledge of that dishonesty, but merely a further confirmation of what was by then already apparent to him (see [465] above). As Mr Nicholas Peacock QC for LBBW put it, vividly but also accurately, "Mr Bracy did not know of the true depth of Value Partners' corruption (nor of Mr Heininger's participation in that corruption) but he did know, to put it bluntly, that Value Partners was willing to sell its clients' interests (including the interests of KWL) down the river if there was money to be made". That was so well before the request for a "letter for K"".

The judge further found that Mr Sanz-Paris had knowledge of Value Partners' dishonesty [771]. He found that both Mr Bracy and Mr Sanz-Paris intended the representation made to be relied upon and that their knowledge and intention was to be attributed to UBS.

The judge found at [772] that the same analysis applied to the "no knowledge of taint representation":

"…Mr Bracy and for that matter Mr Sanz-Paris were aware of the existence of such a conflict and must have known and intended that any intermediary bank in the position of Depfa would understand that in presenting the intermediation opportunity UBS had no knowledge of such a taint…."

The judge found that Depfa relied on both representations in entering into the Back Swaps [785]-[786].

In relation to LBBW the judge found that the same fraudulent representations were made, were false, were made dishonestly (so far as Mr Bracy was concerned) with the intention that LBBW should rely upon them and were relied upon [798]. The LBBW transactions pre-dated the "letter for K episode" and so it was Mr Bracy's prior knowledge and intent which mattered.

None of these findings are challenged on appeal.

In considering the issue of rescission the judge directed himself as follows at [720]:

"The purpose of rescission or avoidance is to restore the parties, so far as possible, to the position in which they were before the contract in question was concluded. There may come a point at which such restitution in integrum is impossible, such that the remedy of rescission is no longer available, but the court has a degree of flexibility in order to ensure that practical justice is achieved: see the summary at Chitty on Contracts, (31st Edn, 2012), Volume 1, paras 6-120 and 6–121."

The judge found that Depfa and LBBW were prima facie entitled to rescind the Back Swaps [787]. That, however, left the question as to what was to happen to the Front Swaps. As the judge observed, if the Back Swaps were rescinded but the Front Swaps were enforceable (as the judge found with regard to Depfa and the German courts found with regard to LBBW) that would leave them with "an unintended and unjustified windfall".

The judge's solution was that "rescission is an equitable remedy" and "the court has a degree of flexibility in order to ensure that practical justice is achieved". In this case he considered that that would be achieved by requiring Depfa and LBBW to give an undertaking not to enforce the Front Swaps, an undertaking that they were both willing to give. As the judge explained at [791]:

"That gives effect to [Depfa/LBBW]'s right to rescind as the victim of a fraudulent misrepresentation by UBS. It gives effect to the substance of the position which is that, regardless of the correct legal analysis, the …. Front and Back Swaps undoubtedly comprised a single commercial package. It ensures that [Depfa/LBBW] does not receive an unjust windfall…."

UBS argued that rescission was not an available remedy because it could not be restored to the position in which it would have been if the transaction had not been concluded because, come what may, it has suffered losses on the hedging contracts into which it entered in order to provide the premium, as well as to cover its costs and make its profit. The judge accepted that UBS had suffered such losses but concluded that it did not mean that rescission is not available [792]. The judge referred to and relied upon Halpern v Halpern [2007] EWCA Civ 291, [2008] QB 195 as support for the proposition that for the purposes of practical justice absence of prejudice to the representor is a secondary consideration. As the judge explained at [793]:

"….[Depfa/LBBW] is the victim of a fraudulent misrepresentation by UBS and the primary requirement of practical justice is that it should be restored to its pre-contractual position. Achieving the absence of prejudice to UBS is a secondary consideration. The hedging contracts into which UBS entered have provided no benefit to [Depfa/LBBW], at any rate once [Depfa/LBBW] repays its intermediation fee. The fact that UBS will be left with losses on these contracts should not be allowed to prevent rescission of the …..Back Swaps. That would be to allow the tail to wag the dog."

UBS only pursues its appeal on this issue if KWL fails in its damages claim against UBS in respect of the LBBW and Depfa swaps (there being no point if KWL succeeds because KWL would then be entitled to be indemnified by UBS for any claim advanced on the Front Swaps). We have upheld the judge's decision on KWL's damages/indemnity claim and therefore the appeal on this issue does not strictly arise. Nevertheless since it has been fully argued by the parties it shall be briefly addressed.

The hypothesis upon which the issue arises is that KWL has no claim against UBS in relation to the LBBW and Depfa STCDOs. In circumstances where KWL does not succeed in its bribery/dishonest assistance damages claim against UBS in respect of the LBBW and Depfa STCDOs, Lord Falconer submitted that the proper way of achieving practical justice is not to impose upon LBBW and Depfa, as a condition of rescission, that they undertake to the court not to enforce the Front Swaps, but rather that they effectively permit UBS to enforce the Front Swaps against KWL at no risk to themselves. It was submitted that this reflects practical justice because unless UBS is permitted to enforce the Front Swaps against it by making their assignment (or its equivalent) a condition of the Back Swaps being rescinded, KWL ends up in the same position as if it did have a claim against UBS, and UBS shoulders all of the loss as a result of its losses on its hedging contracts.

Lord Falconer further submitted that in the case of the two Depfa Front Swaps, that can be done by making it a condition of rescission that Depfa assign to UBS its claims against KWL in respect of those Front Swaps. In the case of the LBBW Front Swap, the position is a little more complicated because, under German law (which governs that Front Swap) it cannot be assigned; and KWL have also appealed the decision of the Leipzig court that the LBBW Front Swap is enforceable, and LBBW does not wish to incur the time and expense of contesting KWL's appeal for the benefit of UBS. However, Lord Falconer submitted that LBBW can be practically protected from the difficulties they fear, simply by allowing UBS to continue to contest KWL's appeal in its own existing capacity as a third party intervenor in the proceedings; or (if LBBW prefer) by granting UBS a power of attorney to contest the appeal on their behalf.

Depfa, LBBW and KWL all oppose UBS's appeal.

In our judgment there are numerous reasons why UBS's appeal should be dismissed. These include the following.

First, as Lord Falconer stressed in relation to Issues 9 and 10, the decision being challenged as to the conditions to be imposed in granting rescission involves the exercise of discretion. It cannot be shown that the judge has erred in principle in his approach since he has correctly considered how practical justice is best achieved. In effect UBS therefore needs to establish that the judge's decision is outside "the generous ambit within which a reasonable disagreement is possible" – see G v G (Minors: Custody Appeal) [1995] 1 WLR 647 HL at 652; Tanfern Ltd v Cameron MacDonald (Practice Note) [2000] 1 WLR 1311 at [32].

Lord Falconer submitted that the exercise of the judge's discretion was somehow vitiated by reason of the fact that he would have been exercising his discretion on the basis that KWL's claim succeeded. We do not agree. The judge does not refer to or rely upon that as a factor. He was understandably concentrating on the position as between UBS and Depfa/LBBW, not that as between UBS and KWL.

Secondly, within Lord Falconer's argument is the assumption that the court should exercise its discretion to achieve practical justice in relation to contracts induced by fraudulent misrepresentation so as to ensure that no loss is suffered by the dishonest representor as a result of contracts entered into by it in reliance on the contracts it had fraudulently induced and in order to fund them. No principled reason has been advanced as to why the court should exercise its discretion to assist the dishonest representor in this manner. Nor has any authority been cited in which any such approach has been adopted or indeed argued for. As KWL put it in its skeleton argument: "UBS can point to no authority that would justify the court in refusing to permit LBBW and Depfa to rescind the contracts that they were duped by UBS into entering merely because to do so leaves UBS out of pocket as a result of arrangements it made to finance its fraud".

Thirdly, UBS's argument depends fundamentally on what it says would be the practically just result as between it and KWL. But the achievement of practical justice in the rescission of a contract made between UBS and Depfa/LBBW is concerned with what is practically just as between those parties, not as between UBS and KWL.

This is borne out by the authorities on restitutio in integrum which make it clear that it is concerned with restoration of the position as regards the rights and obligations created by the contract in question, not some other contract(s). It is in the context of seeking to achieve as full and fair restitutio in integrum as possible that the issue of practical justice arises.

As the High Court of Australia stated in Brown v Smitt [1924] 34 CLR 160, at pp.165-6:

"[P]utting the parties in the position they were in before the contract, replacing them in statu quo, does not involve replacing them in the same position in all respects, but only in respect of the rights and obligations created by the contract which is rescinded".

The fact that a party has incurred losses to third parties by relying on the validity of a contract does not mean that restitutio in integrum cannot be achieved, as is illustrated by the decision of the Privy Council in MacKenzie v Royal Bank of Canada [1934] A.C. 468. In that case Mrs MacKenzie had provided shares to the defendant bank by way of security for the indebtedness of her husband's company. She had done so on the basis of a misrepresentation by the bank. In deciding whether she was entitled to rescind the transaction, Lord Atkin (giving the judgment of the Privy Council) considered that it was irrelevant that the bank had advanced money on the faith of the security, stating at page 476:

"There is no difficulty as to restitutio in integrum. The mere fact that the party making the representation has treated the contract as binding and has acted on it does not preclude relief. Nor can it be said that the plaintiff received anything under the contract which she is unable to restore".

In the present case there is no difficulty in achieving restitutio in integrum. Everything received by Depfa/LBBW under the Back Swaps has been returned and the undertaking given ensures that there is no unjust enrichment. The hedging contracts entered into by UBS did not confer any benefit on Depfa/LBBW of which it might be required to give counter restitution. UBS's entry into these contracts reflected its independent commercial decision and was neither a contractual pre-requisite nor a necessary consequence of the Back Swaps.

Further, the right to payment which Depfa/LBBW received under the Front Swaps was not something which ever belonged to UBS or could be returned to it. Indeed, it is not a benefit which UBS could ever have obtained for itself since it was precluded from contracting with KWL directly. The effect of the order it now seeks is therefore for UBS to be given the fruits of a contract it could never have obtained at the time.

Fourthly, as submitted by Depfa and LBBW, there are legal, practical, reputational and financial risks for them in the proposal that UBS in effect take over and enforce their rights under the Front Swaps. There is no good reason why they should be exposed to any such risks. Further, the legal terms upon which UBS would take over and enforce Depfa and LBBW's rights have not been clearly and satisfactorily set out and defined. UBS's proposals have involved constant adaptation, including at the appeal hearing, and would leave much for the parties and indeed the court to do.

For the reasons given above we would affirm the judge's conclusion under Issue 7.

As the judge noted at [864] this claim did not arise on the findings he had made and so he stated his reasoning and conclusions more briefly than might otherwise be the case.

Each portfolio was managed by UBS GAM pursuant to a Portfolio Management Agreement ("PMA"). The portfolio manager responsible for the management at all material times was Mr Sunil Dattani.

The standard required of Mr Dattani in his management of the portfolios was addressed by the judge at [867]-[875] of the judgment. Having referred to various terms of the PMA, the judge summarised UBS GAM's contractual obligation as follows at [869]:

"It had to exercise a standard of care consistent with the practices and policies of prudent institutional portfolio managers of international standing. It had to apply in the case of KWL the same standard of care as it applied to other customers. And it had to follow its own customary standards, policies and procedures."

At [870] the judge stressed that these obligations had to be considered by reference to KWL's particular portfolios and pointed out that, as KWL had received its premium up front, unlike some other investors, "KWL had only one interest so far as the management of the portfolios was concerned, and that was to avoid defaults reaching the attachment point of its tranches." The judge found that: "What Mr Dattani had to do, therefore, was to focus exclusively on taking steps to minimise the risk of defaults in the portfolios."

On 7 June 2006 Mr Dattani met with Mr Heininger and explained the management strategy which KWL could expect by reference to UBS GAM's standard "pitch book". This emphasised the following points under the heading "Investment Philosophy":

"Selection philosophy

▪ UBS Global AM integrated research to identify credits

— UBS Global AM credit analysis integrated with UBS Global AM equity research

▪ Avoidance of companies with

— poor transparency or accounting

— rapidly changing fundamentals

— outlier spreads

▪ Exposure to low-risk HY [High Yield] companies rather than high-risk IG [Investment Grade] companies

▪ Maximise diversification, minimise risk concentration: no big bets

Monitoring philosophy

▪ Continuous surveillance to identify potential problem credits early (e.g. WorldCom, Enron, Parmalat, Delphi)

▪ Early exit strategy for problem credits

Execution philosophy

▪ Integration of UBS Global AM functions for execution decisions

— portfolio managers, research analysts, traders".

The judge found that both Mr Dattani and UBS GAM's expert, Mr Powell, agreed that the approach described in the pitch book was the approach which KWL was entitled to expect from UBS GAM in managing the STCDO portfolios [874].

The judge further held that this informed the standard of care required of UBS GAM and that it meant that it had "to follow the particular conservative strategy described in the pitch book — specifically, to maximise diversification, minimise risk concentration and avoid what would reasonably be regarded as "big bets" on any particular sector; and to ensure continuous monitoring of potential defaults in order to implement when necessary an early exit strategy" [875].

The issue of breach was addressed by the judge at [876]-[907]. He found that Mr Dattani's management of the portfolios failed to meet the standards required in a number of respects. These were referred to by the judge under various headings and may be summarised as follows:

In the light of these specific findings of breach the judge accepted KWL's "essential complaint" that "the management approach adopted by Mr Dattani was not as set out in the pitch book, but involved an undue and unnecessary concentration of risk which was not properly monitored as the situation developed" [906].

In relation to causation, the judge found that "there is ... no real doubt that Mr Dattani's "concentrated bet" on high-risk financial entities, which constituted a breach by UBS GAM of its obligations under the Portfolio Management Agreements, caused loss to KWL" [908].

In relation to assessment of loss, the judge found that it would be impossible to reconstruct the trades which Mr Dattani would have carried out if he had been exercising the required standard of care [909].

At [910] he concluded that a simpler approach was appropriate, which he set out as follows:

"I accept KWL's submission that a much simpler approach is appropriate. That is to say that any competent manager should (and on the balance of probabilities would) have done at least as well as if there had been no management at all. Having trumpeted the advantages of portfolio management as a means of minimising the risk, and taken the fees payable for managing the portfolios, UBS can hardly suggest otherwise. The Balaba and LBBW STCDO portfolios would have suffered no loss at all if they had not been managed. In the case of the Depfa portfolios, with no management there would have been a loss of some US $92 million but only as a result of the presence in the original portfolios of two entities which Mr Dattani removed. However, I accept KWL's submission that if Mr Dattani removed them, the probability is that any competent manager exercising an appropriate standard of care would also have done so. Necessarily this represents something of a broad brush approach, but in my judgment it is realistic and does not involve the making of unduly generous assumptions in favour of KWL (cf. the reservations expressed by Hamblen J in Porton Capital Technology Funds v 3M UK Holdings Ltd [2011] EWHC 2895 (Comm) at [237] to [245] concerning earlier cases in which a "generous" approach had been applied)."

UBS GAM challenged each of the judge's findings of breach other than that relating to quarterly reporting.

It further contended that on the judge's findings the only causative breach was the concentrated bet on financials.

In relation to assessment of loss, Lord Falconer on behalf of UBS GAM submitted that the judge erred in using as his comparator a portfolio which had not been managed at all. The comparator he should have used was a portfolio which was not negligently managed in the respects which the judge found Mr Dattani to have been negligent. In circumstances where the only relevant breach was the concentration of financials, what the judge should have done was to make an assessment of what loss would have been suffered by the portfolios if Mr Dattani had managed them in a way which was not negligently over concentrated in financials. Had the judge considered, as he should have done, what a permissible level of financials would have been, he should have concluded that the Balaba and Depfa portfolios would have sustained the same loss as they actually did, and the LBBW portfolio would have suffered only 63.24% of the loss it did.

UBS GAM's appeal involves a challenge to various findings of fact made by the trial judge. As emphasised in a number of recent cases, an appellate court will only interfere with such findings in limited circumstances. As Lewison LJ explained in Fage UK Ltd v Chobani UK Ltd [2014] EWCA Civ 5, [2014] FSR 29 at [114]-[115]:

"114.

Appellate courts have been repeatedly warned, by recent cases at the highest level, not to interfere with findings of fact by trial judges, unless compelled to do so. This applies not only to findings of primary fact, but also to the evaluation of those facts and to inferences to be drawn from them. The best known of these cases are: Biogen Inc v Medeva Plc [1997] R.P.C. 1; Piglowska v Piglowski [1999] 1 W.L.R. 1360; Datec Electronics Holdings Ltd v United Parcels Service Ltd [2007] UKHL 23; [2007] 1 W.L.R. 1325; Re B (A Child) (Care Proceedings) [2013] UKSC 33; [2013] 1 W.L.R. 1911 and most recently and comprehensively McGraddie v McGraddie [2013] UKSC 58; [2013] 1 W.L.R. 2477. These are all decisions either of the House of Lords or of the Supreme Court. The reasons for this approach are many. They include:

i.

The expertise of a trial judge is in determining what facts are relevant to the legal issues to be decided, and what those facts are if they are disputed.

ii.

The trial is not a dress rehearsal. It is the first and last night of the show.

iii.

Duplication of the trial judge's role on appeal is a disproportionate use of the limited resources of an appellate court, and will seldom lead to a different outcome in an individual case.

iv.

In making his decisions the trial judge will have regard to the whole of the sea of evidence presented to him, whereas an appellate court will only be island hopping.

v.

The atmosphere of the courtroom cannot, in any event, be recreated by reference to documents (including transcripts of evidence).

vi.

Thus even if it were possible to duplicate the role of the trial judge, it cannot in practice be done.

115.

It is also important to have in mind the role of a judgment given after trial. The primary function of a first instance judge is to find facts and identify the crucial legal points and to advance reasons for deciding them in a particular way. He should give his reasons in sufficient detail to show the parties and, if need be, the Court of Appeal the principles on which he has acted and the reasons that have led him to his decision. They need not be elaborate. There is no duty on a judge, in giving his reasons, to deal with every argument presented by counsel in support of his case. His function is to reach conclusions and give reasons to support his view, not to spell out every matter as if summing up to a jury. Nor need he deal at any length with matters that are not disputed. It is sufficient if what he says shows the basis on which he has acted. These are not controversial observations: see Customs and Excise Commissioners v A [2002] EWCA Civ 1039; [2003] 2 W.L.R. 210; Bekoe v Broomes [2005] UKPC 39; Argos Ltd v Office of Fair Trading [2006] EWCA Civ 1318; [2006] U.K.C.L.R. 1135."

Since that decision the circumstances in which an appellate court is entitled to interfere with a trial judge's findings have been addressed by the Supreme Court in Henderson v Foxworth Investments Ltd [2014] UKSC 41, [2014] 1 WLR 2600 and by the Court of Appeal in Grizzly Business Ltd v Stena Drilling Ltd & Anor [2017] EWCA Civ 94 and in Frank Perry v Raleys Solicitors [2017] EWCA Civ 314. In summary such interference will only be justified where a critical finding of fact is unsupported by the evidence or where the decision is one which no reasonable judge could have reached.

The main criticisms made by Lord Falconer of this finding were as follows:

As Mr Simon Salzedo QC for KWL pointed out, the fundamental fallacy underlying most of these criticisms is the false assumption that the judge's finding of concentration of risk was concerned solely with the amount of financial entities in the portfolio. It is clear, however, that the judge's finding is also directed at the nature and degree of risk taken on thereby. Thus at [878] the judge finds that Mr Dattani had been wrong to "concentrate on highly correlated financial entities, bond insurers, risky consumer lenders, and banks that were known to have high and rising levels of leverage". As he pointed out, this high degree of correlation meant that "defaults in one of these areas were likely to trigger defaults in others also". Similarly, at [906] the judge refers to the "undue and unnecessary concentration of risk" and at [908] to Mr Dattani's breach in making a concentrated bet on "high risk" financial entities.

As the judge found at [878], Mr Dattani accepted that the summary from the report of KWL's expert, Mr Hale, which was put to him in cross examination was factually correct. That summary was:

"The explanation for the substantially higher default rates in the UBS GAM portfolios compared to the investment grade universe is that the approach adopted by UBS GAM was not a diversification strategy. Essentially, UBS GAM took a concentrated bet on financials, bond insurers, risky consumer lenders, and banks that were known to have high — and rising — levels of leverage before the credit crisis struck. These sectors were all highly correlated with each other. By July 2008 exposure to these sectors had been ramped up to above 40 per cent in all three portfolios. The subsequent defaults as the crisis intensified came from within these sectors and the very high overlap between the portfolios meant that total loss was suffered (or virtually total loss priced in via unwind) across all four STCDOs".

Although Lord Falconer sought to criticise the finding that Mr Dattani accepted that he had made such a "concentrated bet", it was clearly a finding open to the judge to make in the light of the evidence cited by him at [877]. As the above passage makes clear, that bet related to the nature of the financials bet upon, not just the fact that they were financials.

Even if one has regard only to concentration in amount, the fact that there was a significant concentration in financials from the outset, and that this was explained by Mr Dattani as reflecting his preference for being overweight, does not justify significantly increasing that concentration in circumstances where what was required was to "maximise diversification" and "minimise concentration of risk", as the judge found. As Mr Hale stated in the passage accepted by Mr Dattani as being factually correct, this was not "a diversification strategy". Nor could the failure to adopt such a strategy be justified by the absolute financial sector limits. These outer limits on what the manager was entitled to do tell one nothing about whether substitutions within the limits were compliant with the standard of care contractually required of him.

Lord Falconer submitted that a concentrated bet on financials was not a breach in itself and could be justified if, for example, Mr Dattani reasonably considered that it would reduce risk or was otherwise an appropriate strategy for the client. As the judge's other findings of breach make clear, however, the substitutions were not made by Mr Dattani in order to reduce the risk of default but rather to improve the Moody's Metric regardless of whether or not it increased the risk of defaults. That was, as the judge found, an inappropriate and wrongful strategy.

As to the reliance on the expert evidence, the judge was entitled to find that the approach followed by portfolio managers "in general" was of little assistance in determining the appropriateness of the approach followed by Mr Dattani having regard to his contractual obligations in relation to this particular portfolio. There was no evidence that other managers were concerned with tranches of the same seniority as KWL's, had the sole objective of minimising the risk of defaults, or owed a standard of care which reflected the specific requirements of the pitch book.

The judge found at [883] that Mr Dattani's view was that his "prime function was to protect the rating" and that his evidence was "symptomatic of a focus on rating to the exclusion of all else". He further found at [885] that Mr Dattani struggled with the concept that KWL's interest - and his only priority - was to minimise default and that his approach involved "a potentially dangerous misunderstanding of what his real task was". His real task was to minimise defaults. That is what his exclusive focus was meant to be.

Although the judge sets out in his judgment at [882] and [884] passages from the evidence which support these findings, Lord Falconer sought to challenge the findings made and suggested that the fair view of Mr Dattani's evidence was that the rating was only one of the considerations to which he had regard, and that he did not focus on rating to the exclusion of all else. To that end we were referred to various paragraphs in Mr Dattani's witness statement and to passages of his evidence.

Nothing we have been shown comes close to justifying this court in interfering with findings made by the judge following consideration of all the written and oral evidence. The reference to Mr Dattani's view of his "prime function", for example, is a direct quotation from his oral evidence. These were his own words. In another passage from his evidence he explained that "the rating is fundamental" and that he "would achieve the client's target by maintaining the rating".

Lord Falconer emphasised that it was Mr Dattani's evidence that Mr Heininger had stated that the rating was important to him. No doubt he did wish the rating to be maintained, if possible, but that does not alter UBS GAM's contractual obligations, or amount to an instruction to maintain the rating regardless of the risk of default.

The judge found that Mr Dattani understood that "he was required when substituting Reference Entities into the portfolio to maintain or improve" the Moody's Metric. As the judge explained at [886]:

The judge found that Mr Dattani did not understand this, that he described the Moody's Metric as the "contractually agreed measure of risk" when it was no such thing, and that he appeared to have believed that all his trades were required to improve the Moody's Metric and, conversely, that any trade which did improve the Metric was desirable whether or not it increased the risk of defaults [886].

Lord Falconer sought to challenge these findings and again referred us to various passages in Mr Dattani's witness statement and evidence which he submitted showed that Mr Dattani did in fact understand that the Moody's Metric was a Hurdle MM. There are, however, many passages in Mr Dattani's evidence which support the judge's findings. The description of the Moody's Metric as "the contractually agreed measure of risk" came from Mr Dattani's own witness statement. He confirmed in terms in evidence that at all times it was his view that it was "the contractually chosen measure of risk". He explained this on the basis that "the only way I can maintain the rating is if I maintain the Moody's Metric". He described it in evidence as "the only measure of risk" and said that there was no need to do a trade unless it was "maintained or improved". Further, save for some error trades, the evidence was that all Mr Dattani's substitutions did in fact improve the Moody's Metric. Again no proper basis for overturning the judge's findings has been made out. Further, even if Mr Dattani did understand that the Moody's Metric was only a Hurdle MM, the evidence was that he carried out his substitutions by reference to it regardless of whether the Hurdle was engaged.

The judge found that "Mr Dattani's focus on the rating tended not to give proper weight to the relevance of spreads as a measure of risk". As the judge explained, a widening spread could be a signal of deteriorating credit quality and it was therefore important to consider the reason for the widening. Further, ratings tend to lag the market and "what might appear to be an attractive spread for a given rating might in reality represent an increasing risk with which the rating agency had not yet caught up" [887].

The judge further found that "Mr Dattani accepted that he did not pay enough attention to attempting to catch names whose credit quality was falling" and cited an answer given in cross examination where he acknowledged that history showed this to be right [888].

Lord Falconer submitted that the finding of negligence made by the judge was unfair, that it was informed by this hindsight answer and that substituting with wider spreads is not negligent in itself. Again, no proper basis for challenging the judge's findings has been made out. The judge was not finding that substituting with wider spreads is in itself negligent, but rather that it is important to have regard to the reason for the widening spread before doing so. Indeed Mr Dattani expressly accepted that it is important to take care over what is causing spreads to be wide when substituting names with wide spreads. On the judge's findings, however, Mr Dattani failed to give proper weight to this when making his substitutions. He was unduly focused on ratings and the arbitrage to be made from substituting portfolio entities for names with wider spreads.

The judge found that Mr Dattani did not carry out any ongoing assessment of the impact of different market scenarios on the KWL tranches and that, if he had, he ought to have been alerted to the risks of the concentrated position in financials he was running [889].

Lord Falconer submitted that this finding was unjustified as the complaint had not been pleaded and was not put to Mr Dattani. This was a feature of UBS GAM management described in the pitch book, the significance of which was not appreciated at the time of pleading. The point was put to the experts and there was no disclosure of any such analyses. Mr Dattani's evidence was that the principal form of record keeping was to take a snapshot of the COS Toolkit before and after substitutions, but that is not the assessment referred to. The judge was best placed to determine the fairness of allowing reliance on this point and it was open to him on the evidence to make the finding which he did.

This breach was not disputed. The judge found that "Mr Dattani's reporting generally was poor"; that there was "no contemporaneous written analysis explaining the reasons for the various substitutions carried out or analysing their potential effects on the risk of loss to the KWL tranches", and that there was "no evidence of any review of the existing contents of the portfolios or of any strategy as to whether and why any particular entity should be retained or substituted out" [890].

With one exception quarterly reports were produced. That exception was deliberate and was done to cover up an error trade. Not only were KWL not told of the error but they were left to bear the transaction costs of both the error and replacement trades. Mr Dattani saw nothing wrong with this.

In December 2006 Mr Dattani was criticised for poor reporting by UBS GAM's internal auditors, who commented that his quarterly reports "contain insufficient detail to evidence regular monitoring over credit risks and the rationale for CDS substitution decisions taken" [891]. Mr Dattani did not follow the auditors' recommendations for improvement and the judge found that proper monitoring of credit risks was not taking place [896].

Lord Falconer submitted that this breach of duty was not causative and therefore does not matter. However, the judge found that Mr Dattani's poor reporting reflected a failure to carry out proper monitoring of credit risks. As with the failure to carry out impact assessments, this is likely to have contributed to the fact that Mr Dattani was not alerted to the risks of the concentrated position in financials he was running.

The judge found that Mr Dattani failed to follow the early exit strategy promised by the pitch book. The judge then cited an extensive passage of his evidence, from which he found that Mr Dattani appeared to be saying that "he would only consider removing a name when he became convinced that it was likely to default" [900].

The judge found at [902] that Mr Dattani's strategy was actually "a late exit strategy which, because of the high cost in terms of lost subordination of removing a name with a wide spread, was calculated either to lead to defaults (if the name was not removed) or to increase the vulnerability of the portfolios by reducing the number of defaults required to reach their attachment point (if it was removed, but only at the cost of a loss of subordination)." The judge noted at [905] that there was a dispute between the experts as to when Mr Dattani should prudently have sought to reduce the portfolio's overweight position in financials (July 2007 or some later date) but found that it was "not apparent that Mr Dattani ever considered this properly".

Lord Falconer complained that this point was not pleaded and that the judge's finding was contradicted by evidence as to what Mr Dattani actually did. It was also submitted that there was oral evidence, as referred to by the judge at [901], during which Mr Dattani said that he would look at exiting where there was a possibility of default rather than a likelihood of default.

The judge's finding that Mr Dattani was saying that he would only consider removing a name when he was convinced of a likelihood of default was open to him on the evidence, as borne out by the passage of evidence set out at [900].

As is clear from [904], the judge had regard to UBS GAM's case as to what they actually did, and to the table which was relied upon showing the large number of "credit impaired" names removed and the fact that 80% of them were removed when they were investment grade. He noted that this table had not been put to Mr Dattani or the experts and that it contradicted the evidence Mr Dattani had given. He further noted that there was no evidence of the judgment made at the time of their removal. He was, however, prepared to assume that the names removed could reasonably be regarded as credit impaired but found that it did not follow that Mr Dattani was following a strategy of removing names where he foresaw a risk of a future decline.

Whether or not this point was pleaded, it was addressed in the evidence and the findings made by the judge reflect Mr Dattani's own evidence. Again no proper or sufficient basis for challenging the judge's findings has been made out.

In summary, for the reasons outlined above, UBS GAM cannot meet the high hurdle required for this court to overturn the findings of breach made by the judge.

Although the judge found various breaches under seven different headings, the heading does not always reflect the full findings made. It is also helpful to consider those findings of breach in the round.

The judge found that Mr Dattani's sole objective should have been to take steps to minimise the risk of defaults. To that end he should have been maximising diversification and minimising risk concentration, as stated in the pitch book.

Mr Dattani did not, however, consider that his role was to minimise the risk of defaults. Had he so considered he would on his own evidence have made minimal substitutions. Instead he regarded his "prime function" to be to maintain the rating of the tranche. To that end he made substitutions to improve the Moody's Metric regardless of whether or not it increased the risk of defaults. That involved introducing entities with wider spreads without regard to whether that was a signal of deteriorating credit quality. Mr Dattani did not specifically apply his mind to default risk when making his substitutions and no proper monitoring of credit risks took place.

Further, Mr Dattani did not follow an early exit strategy and would generally only consider removing a name when he became convinced it was likely to default. By that stage it was difficult and costly to remove the name, so that either the name was left in, which increased the default risk, or it was taken out at the cost of a loss of subordination, which reduced the number of defaults required to reach the attachment point.

The result of following this approach, in serial breach of his duty as portfolio manager, was that KWL ended up with a portfolio containing a concentration of "highly correlated financial entities, bond insurers, risky consumer lenders, and banks that were known to have high and rising levels of leverage", which in itself was a breach of duty. It was the default of some of these entities which led to the total loss on the portfolios.

KWL advanced a case at trial that each of the individual names which defaulted were introduced negligently by Mr Dattani. The judge, however, made no findings in relation to individual substitutions. Mr Salzedo sought to persuade this court that it should make such findings and we were taken to some of the evidence and many of the submissions at trial relating to the individual names. The judge was in a far better position than we could ever be to make findings on these detailed issues, but declined to do so. We are not persuaded on the limited evidence we have been shown that this court should take the exceptional course of making its own findings or that the judge should have made the findings contended for by KWL.

The judge dealt with these issues succinctly. His essential reasoning was set out at [908]-[910].

In considering the judge's conclusions as to assessment of loss the evaluative nature of that exercise has to be borne in mind. As explained by Toulson LJ in Parabola Investments Ltd v Browallia Cal Ltd [2010] EWCA Civ 486, [2011] QB 477 at [22]-[23]:

"22…..Some claims for consequential loss are capable of being established with precision (for example, expenses incurred prior to the date of trial). Other forms of consequential loss are not capable of similarly precise calculation because they involve the attempted measurement of things which would or might have happened (or might not have happened) but for the defendant's wrongful conduct, as distinct from things which have happened. In such a situation the law does not require a claimant to perform the impossible, nor does it apply the balance of probability test to the measurement of the loss.

23.

The claimant has first to establish an actionable head of loss. This may in some circumstances consist of the loss of a chance, for example, Chaplin v Hicks [1911] 2 KB 786 and Allied Maples Group Ltd v Simmons & Simmons [1995] 1 WLR 1602, but we are not concerned with that situation in the present case, because the judge found that, but for Mr Bomford's fraud, on a balance of probability Tangent would have traded profitably at stage 1, and would have traded more profitably with a larger fund at stage 2. The next task is to quantify the loss. Where that involves a hypothetical exercise, the court does not apply the same balance of probability approach as it would to the proof of past facts. Rather, it estimates the loss by making the best attempt it can to evaluate the chances, great or small (unless those chances amount to no more than remote speculation), taking all significant factors into account: see Davies v Taylor [1974] AC 207, 212, per Lord Reid, and Gregg v Scott [2005] 2 AC 176, para 17, per Lord Nicholls of Birkenhead, and paras 67–69, per Lord Hoffmann."

As made clear in the Parabola case, an appellate court will be reluctant to interfere with evaluative judgments of this kind. As stated by Floyd LJ in Wellesley Partners LLP v Withers LLP [2015] EWCA Civ 1146, [2016] Ch 529 at [125]:

"125…..This court is notoriously reluctant to interfere with evaluative judgments of this kind in the absence of an error of principle: see for example per Lord Hoffmann in Biogen Inc v Medeva plc [1997] RPC 1, 45. Nowhere is this more so than in a judge's assessment of damages. Toulson LJ put it in this way in the Parabola case [2011] QB 477, para 24:

"The judge had to make a reasonable assessment and different judges might come to different assessments without being unreasonable. An appellate court will therefore be slow to interfere with the judge's assessment. As Lord Wright said in Davies v Powell Duffryn Associated Collieries Ltd [1942] AC 601, 616–617: 'An appellate court is always reluctant to interfere with a finding of a trial judge on any question of fact, but it is particularly reluctant to interfere with a finding on damages which differs from an ordinary finding of fact in that it is generally much more a matter of speculation and estimate. No doubt, this statement is truer in respect of some cases than of others … It is difficult to lay down any precise rule which will cover all cases, but … the court, before it interferes with an award of damages, should be satisfied that the judge has acted on a wrong principle of law, or has misapprehended the facts, or has for these or other reasons made a wholly erroneous estimate of the damage suffered.'"

Lord Falconer submitted that the judge's approach was flawed for three principal reasons:

Contrary to Lord Falconer's submission, in our judgment if the judge's reasoning at [910] is considered as a whole the judge is not using an unmanaged portfolio as his comparator. The question which the judge asks himself is what "any competent manager should (and on the balance of probabilities would) have done". This is borne out by his later reference to what "any competent manager exercising an appropriate standard of care" would have done. It is also borne out by his approach to the two Depfa names which he found any competent manager would have removed. The competent manager is the comparator which the judge adopts.

What would have been achieved if there had been no management is relied upon by the judge as evidence of what any competent manager would reasonably be expected to achieve. This was a portfolio which was to be managed with the sole priority of minimising the risk of default and UBS GAM were being paid millions of dollars in fees to perform this specific task. In those particular circumstances it was open to the judge to find that any competent manager would have done at least as well as no manager. This was a reasonable general inference to draw.

The drawing of such an inference is not contradicted by the generalised evidence relied upon by UBS that certain managed STCDO portfolios underperformed static portfolios during the relevant period. As Mr Salzedo pointed out, this evidence merely showed that a greater proportion of managed portfolios than static portfolios had been downgraded to Caa or below following the financial crisis. It did not say whether the tranches in those portfolios actually defaulted, which was the relevant question in making a comparison with KWL's tranches. Further, there was no evidence that these portfolios were set up or managed in the same way as KWL's portfolios, for similar types of clients, or with the sole priority of minimising defaults.

Whilst there may be some force in the criticism made that such a general inference does not in itself justify the conclusion that any competent manager would have avoided any loss, there was other evidence to support the finding made by the judge.

It was, for example, Mr Dattani's evidence that if you manage a tranche to avoid defaults you would not do many trades. This provides support for the view that the approach of a competent manager would be broadly similar to that for an unmanaged portfolio, save for names removed specifically for the purpose of minimising the risk of default.

Further, there was evidence that if UBS GAM had merely replicated the experience of the investment grade market as a whole then the portfolios would have experienced defaults of around 1.94% through to the end of 2012, far below the attachment point of KWL's tranches. This was borne out by a table in Mr Hale's report. Further, (i) Mr Hale pointed out that this was a conservative comparison because the investment grade market as a whole would include a large number of small and higher risk names; (ii) when considering the market as a whole there is no option of substituting entities away and (iii) given that UBS GAM's sole priority was to minimise the risk of default one would expect a more conservative strategy than the market average.

In addition, the losses were the result of defaults among the "highly correlated financial entities, bond insurers, risky consumer lenders, and banks that were known to have high and rising levels of leverage" which were substituted into the portfolio and kept in there as a result of the negligent strategy of Mr Dattani. A competent manager would not have followed such a strategy.

As Mr Salzedo put it, a competent portfolio manager who did not build up a concentrated exposure to high-risk financial entities but instead sought to diversify exposure, and did not chase the Moody's Metric but rather sought to avoid defaults, monitored credit exposure and adopted an early exit strategy would on a balance of probabilities have avoided any defaults.

Although the judge did not expressly rely on this further evidence, it is clear that there was ample evidence before him to justify the conclusion that he reached that any competent manager of this portfolio should and would have avoided any loss.

Lord Falconer further submitted that in any event the relevant comparator was not any competent manager but rather this particular manager, Mr Dattani, acting competently. This was important because, for example, Mr Dattani had a stated preference for being overweight in financials and regard should be had to this in considering what would have been done had he acted non-negligently.

In support of this submission Lord Falconer relied in particular on the case of Bolitho (Deceased) v City and Hackney Health Authority [1998] AC 232. In that case a doctor was summoned to attend to a child admitted to hospital with breathing difficulties but failed to attend. The child suffered a cardiac arrest which would have been avoided if the child been intubated. The claim failed because it was found that although the doctor had been in breach of duty in failing to attend, it was also found that she would not in fact have intubated the child and such a decision would not have been negligent. The case therefore involved considering what the particular doctor would have done if she had not been negligent. Similarly, it is submitted that in this case the judge should have considered what the particular portfolio manager, Mr Dattani, would have done if he had not been negligent.

As Mr Salzedo submitted, the Bolitho case focuses on issues of causation rather than assessment of loss. It is likely to be of particular relevance where the negligence consists of an omission and there are different ways of doing what should have been done. In those circumstances it may well be appropriate to consider what the negligent person would have done. As pointed out, however, in Beary v Pall Mall Investments [2005] EWCA Civ 415, [2005] PNLR 35 in many cases the only relevant question in relation to a defendant's conduct is what should have been done rather than what he would have done if he had not been negligent. As Dyson LJ stated in that case at [30]:

"30.

In Bolitho, the claim would have succeeded either if the judge had found that the doctor who negligently failed to attend, would as a matter of fact have intubated if she had attended, or if it would have been negligent not to intubate. It was necessary on the facts of that case to consider what the doctor would have done if she had attended the child. But it does not follow that it is necessary in every case to ask what a defendant would have done if he or she had not been negligent. That question falls to be considered only where it is relevant on the facts of the particular case. In Bolitho it was relevant because the negligence lay in the failure to attend, and there was a causal link between that failure and the injury suffered by the child, because, if the doctor had attended and if she would have intubated, she would thereby have averted the injury. This causal link on the facts of that case was the hypothetical conduct of the defendant herself. In many negligence cases, the question is what would the claimant or some third person have done if the defendant had not been negligent. Usually, the only relevant question in relation to a defendant's conduct is: what should the defendant have done? It will not often be meaningful to go on to ask what the defendant would have done if he had not been negligent. It is tautologous to say that, if the defendant had not been negligent, he would not have acted negligently."

This is not a negligent omission case nor is it a case in which the issue on appeal is one of causation. The judge clearly found at [908] that loss was caused as a result of Mr Dattani's concentration on "high risk financial entities" consequent upon the breaches of duty found by him. The issue is the assessment of that loss. What Mr Dattani would have done is not relevant to that issue. That assessment falls to be made by reference to the standards of a reasonably competent manager, just as in the case of Banque Bruxelles S.A. v Eagle Star [1997] AC 191 the loss was assessed by reference to the standards of a reasonable valuer – see, in particular, the judgment of Lord Hoffmann at pp.221F-222A.

As to the positive case now advanced by UBS GAM that the proper approach to damages is to make an assessment of what loss would have been suffered by the portfolios if they had been managed in a way which was not negligently over concentrated in financials, this case was not advanced at trial and would require a proper evidential basis. For example, the table now relied upon was not produced at trial and involves and requires expert input. Further, as Mr Salzedo pointed out, it is internally inconsistent as it assumes that the level of subordination was that at the outset or in July 2007 but relies on the portfolio being comprised of entities that were there at the end rather than at the beginning or in July 2007, thereby ignoring the effect on subordination of the substitutions made.

In any event, UBS GAM's new positive case is premised on the false assumption that the sole concentration which the judge has found to be relevant was as to the amount of financial entities. For reasons already given, that is an oversimplification and a wrong interpretation of the judge's findings. The judge's findings focused on both the "high risk" of the financials concentrated upon and the extent of their concentration. In those circumstances the pro-rata quantitative approach adopted in UBS GAM's table is inappropriate, quite apart from the fact that it is not supported by, and has not been tested in, evidence.

For the reasons outlined above, UBS GAM cannot show that the judge has erred in the evaluative judgment he made as to the assessment of loss in this case.

In those circumstances it is not necessary to consider the argument raised by KWL in its Respondent's Notice as to whether the judge could and, if necessary, should have made generous assumptions in its favour in reliance by analogy on the case of Armory v Delamirie [1721] 73 ER 644.

Accordingly, we would dismiss UBS GAM's appeal on Issue 8.

As already noted, in considering the issue of rescission the judge directed himself at [720], by reference to the summary in Chitty on Contracts, (31st Edn, 2012), Volume 1, paragraphs 6-120 and 6–121, that the aim is restitutio in integrum but that the court has a degree of flexibility in order to ensure that practical justice is achieved.

At [721] the judge found that the Balaba STCDO constituted a single transaction:

"…the Balaba STCDO and the single name CDS constituted a single transaction, despite the fact that they were contained in two separate contractual documents. That was always the commercial reality. There was never any question of one being concluded without the other. Without the STCDO, KWL would have had no interest in entering into a CDS. Without the premium generated by the STCDO, it would probably not have had the readily available cash to pay the premium required for the CDS…"

The judge further found that this also applied to the transactions involving LBBW and Depfa as intermediaries, at [722]:

"…although the later parts were complicated by the intervention of LBBW and Depfa as intermediaries, as between UBS and KWL all four parts of this transaction were in reality a single transaction for the restructuring of the liabilities under KWL's single name bonds.

In considering what practical justice required in the circumstances of this case the judge held that given that the STCDOs and the CDSs represented a single transaction, any restitutio in integrum which allowed KWL to retain benefits derived from the CDSs would fail to give effect to the practical justice which the law requires [721].

The Balaba CDS had been unwound by agreement in September 2008 upon payment of a premium of £3.3 million by UBS to reflect the fact that KWL was "in the money" on this contract. Given his conclusion that the STCDO and the CDS comprised a single transaction, the judge held at [727] that the payment should be returned, observing that: "If KWL were to keep it, it would be better off than if the transaction had not been concluded in the first place."

In relation to the remaining CDSs KWL claimed that some US$67 million was due to it on their early termination. The judge held at [728] that for KWL to be able to claim this sum as well as being entitled to rescind the Balaba STCDO and to be indemnified by UBS for any liability which it may have under the LBBW and Depfa Front Swaps "would be a fairy tale ending for KWL to this otherwise rather sad story. It would mean that this transaction, in which KWL was defrauded by its own crooked managing director and corrupt agent, and which was disastrously affected by the worst financial crisis in living memory, had turned out to be massively profitable for it after all." He held that this would not accord with practical justice and that KWL could not rescind the Balaba STCDO as well as retaining or recovering sums under the CDSs.

KWL contended that the judge was wrong to conclude that it would be unjust to enforce the CDSs while allowing KWL to rescind the STCDOs. Mr Midwinter for KWL submitted that the contracts are capable of existing and being enforced independently and are legally independent of one another. They are contained in separate contractual documents. KWL paid UBS a premium for each CDS that represented UBS's calculation of the true cost of the CDS, just as it would if it had acquired the CDS from any other market participant. UBS paid KWL a premium for each STCDO, separately calculated by UBS as the price of the STCDO. The triggers for liability under the CDSs are separate from and independent of the triggers for liability under the STCDOs. The CDSs themselves were unobjectionable contracts under which KWL paid a premium for protection against the risk of insolvency of an entity to which it was potentially exposed. As the victim of bribery and/or a transaction procured via agents with a known conflict of interest of which UBS had notice, KWL was entitled to choose to rescind the objectionable STCDOs but maintain in place the unobjectionable CDSs. There is nothing unjust in requiring UBS to perform the CDSs according to their terms.

In support of this case that the STCDOs and the CDSs were independent contracts Mr Midwinter relied on the judgment of Coleman J in De Molestina v Ponton [2002] 1 QBD 587 at [6.9] where he stated as follows:

"6.9

The crucial issue in the present applications is, however, one of mixed fact and law and it is how one identifies the criteria for determining whether a number of separate contracts are part of a single overall transaction for the purposes of the rule against rescission of part of a transaction. On this point there is little or no help in the authorities, but application of general principles strongly suggests the necessary criteria. If a representee is induced to enter into separate contracts A & B by the same misrepresentation, it may be that performance of contract B depends on the prior performance of contract A. In that case one cannot rescind contract A without also rescinding contract B. To permit the survival of contract B would be inconsistent with the principles of restitutio in integrum. But there may be cases where, although both contracts were induced by the same misrepresentation either can be performed without performance of the other. In that case the representee may rescind unless the contract not sought to be rescinded would never have been entered into by the parties without also entering into the other. Thus, for example, in a case where the transaction is divided into different contracts simultaneously negotiated, it may be that the consideration for the whole bargain is written into one contract, leaving only nominal consideration in the other contract. In that event it would not be open to the representee to leave open the contract that gave him the main consideration while rescinding the other contract under which his primary performance obligation lay. Again, to do otherwise would not effect restitutio in integrum. Or there may be cases where it is clear from the terms of the contracts and the matrix evidence that the subject matter of the contracts is so interrelated that, although it would be theoretically possible to perform each separately, one would never have been entered into without that contract sought to be rescinded. However, in the absence of structural interdependence between separate contracts, the most usual determinant of inseparability is likely to be the distribution of consideration for the whole bargain between the separate contracts."

Mr Midwinter submitted that in this case there was separate and real consideration provided under the STCDOs and the CDSs and that there was no or no sufficient structural interdependence between the contracts. They should accordingly be treated as separate contracts for the purpose of rescission.

As already held, the decision being challenged as to the conditions to be imposed in granting rescission involves the exercise of discretion and it cannot be shown that the judge has erred in principle in his approach since he has correctly considered how practical justice is best achieved.

Mr Midwinter submitted that the judge had, however, erred in law in his conclusion that the STCDOs and the CDSs were a single transaction for the purpose of the issue of rescission. But this is not a pure question of law. As Coleman J observed in the De Molestina case it involves a mixed question of fact and law to which matrix evidence may be of particular relevance.

In the light in particular of the history to the transactions as fully set out by the judge, the fact that there was never any question of one transaction being concluded without the other, and the fact that he found that the commercial reality was that they were a single transaction, we consider that the conclusion of the judge that they were a single transaction for the purpose of the issue of rescission is clearly justified.

It is to be noted that in the De Molestina case Coleman J gives as an example of a single overall transaction for the purposes of rescission a case where "the subject matter of the contracts is so interrelated that, although it would be theoretically possible to perform each separately, one would never have been entered into without that contract sought to be rescinded." This could be said to be such a case.

If the judge was justified in finding that the STCDOs and the CDSs were a single transaction for the purpose of the issue of rescission then his conclusion that practical justice required that KWL could not rescind the Balaba STCDO as well as retaining or recovering sums under the CDSs is unassailable. As the judge observed, to conclude otherwise would mean that KWL could "pick and choose". It would enable them to achieve piecemeal rescission with a view to achieving a financial windfall. That would not be practical justice and on any view the judge was entitled so to conclude.

For the reasons given above we would affirm the judge's conclusion under Issue 9.

The judge found at [726] that UBS paid the STCDO premium in accordance with KWL's instructions. Having been so paid it was then misappropriated by Value Partners acting as UBS's agent. The judge found that this "was done in [Value Partners'] capacity as KWL's agent pursuant to a letter signed by both of KWL's two managing directors" (the further engagement letter – see [44] above).

Against the background of those findings the judge concluded that as a matter of practical justice the part of the premium siphoned off by Value Partners should be repaid to UBS as a condition of rescission of the Balaba STCDO.

Mr Midwinter submitted that the sums taken by Value Partners were never received by KWL but were stolen by Value Partners acting as dishonest middlemen who, as the judge found: (i) had been acting from the outset pursuant to a secret agreement with UBS to push UBS's unsuitable product onto KWL for the mutual benefit of UBS and itself; (ii) had an undisclosed conflict of interest known to UBS; (iii) were known by UBS to have concealed the amount of the premium from KWL, and; (iv) were engaged in a deceit on KWL as UBS knew from September 2006 at the latest. KWL had no idea that Value Partners was conducting a fraud on it; UBS did know and took advantage of that fact. In those circumstances, it was submitted that it is unjust to require KWL to make 'restitution' of sums that it never received (and moreover to pay interest on those sums) as a condition of rescission. The judge should have concluded that practical justice would be achieved by deducting those sums from the amount of the premiums that KWL is required to pay to UBS (or at the very least not requiring KWL to pay interest on them).

It is not contended that the judge erred in law or in his approach. The submission made is that the judge should have reached a different conclusion as to how practical justice was best to be achieved and as to the conditions upon which rescission was to be granted. In effect KWL accordingly needs to show that the conclusion reached by the judge was outside "the generous ambit within which a reasonable disagreement is possible".

In our judgment the findings made by the judge at [726] amply justify the conclusion which he reached.

The judge found that the siphoning off of the premium was carried out by Value Partners in its capacity as agent for KWL and was not done pursuant to any arrangement with UBS. The judge further found that it was the actions of KWL's two managing directors rather than anything done or not done by UBS which enabled the siphoning off to be carried out. Mr Midwinter's emphasis on various aspects of the arrangement between UBS and Value Partners was therefore essentially beside the point.

Further, as the judge found, the premium was paid by UBS in accordance with KWL's instructions. What happened to it thereafter as a result of actions taken by KWL's own agent is not UBS's responsibility.

Although Mr Midwinter asserted that the premium was never received by KWL, as Lord Falconer pointed out, that is not correct. Not only was it paid into an account in accordance with KWL's instructions, but that was a KWL account. The Wilmington Trust account was created in order to hold the STCDO premiums and was a KWL account. Value Partners was able to access the funds in that account because of a Power of Attorney which KWL had given to it.

In circumstances where the premium was paid by UBS in accordance with KWL's instructions, received into a KWL account, paid out to Value Partners pursuant to an arrangement which had been made by KWL's two managing directors, of which UBS was unaware and which had nothing to do with its arrangement with Value Partners, it is plain why, as a matter of practical justice, the judge should conclude that it should be returned as a condition of rescission. On any view that was a conclusion which was reasonably open to him.

For the reasons given above we would affirm the judge's conclusion under Issue 10.

In summary, our final conclusions are as follows.

In relation to Issue 1, we conclude that Value Partners was not acting as UBS's agent when it paid the bribe to Mr Heininger, and was therefore not responsible in law for the bribe on that ground. In relation to Issue 2, however, we consider that by virtue of its dishonest assistance in Value Partners' abuse of its fiduciary duty to KWL, UBS's conscience has been affected by the bribe so as to make it unconscionable for it to enforce the Balaba STCDO, which was a transaction procured by a bribe paid by Value Partners to Mr Heininger in the course of that abuse of duty. In any event we have also found, in relation to Issue 3, that the Balaba STCDO is unenforceable by reason of UBS's knowledge of, and dishonest assistance in, Value Partners' abuse of its duty to provide disinterested advice to KWL. The knowledge of Mr Heininger is not, for these purposes, to be attributed to KWL, so that KWL cannot be taken to have consented to that conflict of interest. Therefore whilst we would allow UBS's appeal on Issue 1, the Balaba STCDO remains unenforceable by UBS by reason of our conclusions on Issues 2 and 3.

In accordance with our conclusion on Issue 4, KWL is entitled to rescind the Balaba STCDO, as the grant of rescission is neither unfair and disproportionate, nor contrary to the principle that he who seeks equity must come to the court with clean hands. By reason of our conclusions in relation to Issues 9 and 10, such rescission requires that KWL cannot retain, or further receive, any sums paid or due under the CDSs, which were part of the same overall transaction as the STCDOs, and that KWL return to UBS a sum equivalent to that portion of the STCDO premiums which was siphoned off by Value Partners.

On Issue 5, we conclude that UBS's deceit claim against KWL must fail for the first of the two reasons given by the judge; namely, that the losses suffered by UBS on its hedging contracts were not caused by Mr Heininger's fraudulent misrepresentation, but rather by the rescission of the Balaba STCDO which, in accordance with our conclusions above, arises in part from UBS's dishonest assistance in Value Partners' abuse of its fiduciary duties to KWL.

On Issue 6 we conclude that, if necessary, KWL would be entitled to damages or an indemnity from UBS in the event that it was required to pay LBBW and/or Depfa the amounts due under the LBBW or Depfa Front Swaps.

In relation to Issue 7, the judge was correct to conclude that rescission of the Back Swaps should be contingent on Depfa and LBBW undertaking not to enforce the Front Swaps. In any event, in reaching that conclusion the judge was plainly within the reasonable bounds of his discretion.

In relation to Issue 8, we consider that in any event, UBS GAM is liable in damages to KWL for its negligent management of the STCDO portfolios. UBS GAM has failed to meet the high hurdle required for this court to overturn the judge's findings of breach. Nor has UBS GAM shown that the judge's analysis of causation and loss was flawed.

Accordingly, we would allow UBS's appeal on Issue 1, but dismiss its appeal on Issues 3, 4, 5, 6, 7 and 8. Further, we would accept KWL's additional ground for upholding the judge's decision under Issue 2, but dismiss its cross-appeal on Issues 9 and 10.

Whilst I would agree with the conclusions reached by Lord Briggs and Hamblen LJ on Issues 1 (subject to one point), 7, 9 and 10, I disagree with the majority on a number of other issues and, consequently, on the disposition of the appeal. Most significantly, in relation to Issues 2-4, I conclude that KWL was not entitled to rescind and UBS was therefore entitled to enforce the Balaba STCDO. I gratefully adopt the summary of the facts, enumeration of the issues and abbreviations of the majority judgment.

In my view, there is no basis on which to extend the "one reservation" by Millett J in Logicrose to the present facts so as to reach the conclusion that UBS's conscience was affected by the bribe paid by Value Partners to Mr Heininger ("the bribe").

The general principle articulated in Logicrose was that a principal ('A') was entitled to rescind a transaction where, to the actual or "Nelsonian" knowledge of the counterparty ('B'), A's agent was not providing A with disinterested advice and A was not aware of the agent's conflict. This general principle is relevant to Issue 3, in that KWL seeks to rescind on the basis that UBS had actual knowledge that Value Partners was not giving KWL disinterested advice as a result of the arrangement between UBS and Value Partners ("the UBS-VP arrangement").

The reservation in Logicrose was that A may also be entitled to rescind where B had made a payment to A's agent and not disclosed it to A, expecting the agent to do so, or where B had dealt secretly with A's agent knowing that the agent may intend to retain some private advantage from that dealing. In effect this reservation represented a very modest extension to the general principle. A can rescind even if B does not have actual knowledge that A's agent was not providing disinterested advice, because B must be taken to have accepted the blatant risk thereof due to something which B knowingly has done. KWL sought to argue that this reservation applied to the present facts, so as to entitle it to rescind on the basis that UBS's conscience was affected by the bribe.

There are three reasons why I would not significantly extend the reservation in Logicrose in this way.

First, I do not accept the cornerstone of the majority's reasoning, at paragraph 113, that UBS's conduct in relation to the UBS-VP arrangement has the effect that UBS's conscience was affected by the bribing of Mr Heininger. In my view, UBS's knowledge of the UBS-VP arrangement did not extend to or encompass knowledge of the bribery and nor does it have the consequence that UBS's conscience was affected by the bribery.

Certainly, UBS did not have a level of knowledge of the bribery equivalent to that required for dishonest assistance, per the approach set out in Ultraframe which I discuss below under Issue 3. This lack of knowledge is not an end to the matter in relation to Issue 2, since the Logicrose reservation is avowedly concerned with whether B's conscience is affected and with circumstances which may not involve knowledge. Nonetheless, to treat UBS's conscience as affected by the bribe fails to delineate the abuse which could fairly affect UBS's conscience. Whilst Value Partners played a role in both the arrangement with UBS and the bribery of Mr Heininger, and breached its duties to KWL through each, the two schemes were otherwise self-standing. Each scheme could have operated without the other and, indeed, the scheme to bribe Mr Heininger was hatched some time before Value Partners made contact with UBS: see [591] of the judgment below. Moreover, it could not even be said that UBS had constructive knowledge of the bribe. This factual distinction is underscored by the significant difference in the gravity of the two arrangements. Quite simply, it is one thing for UBS to know that Value Partners had a conflict of interest because it represented that it would be directing all clients' business towards UBS, but another thing entirely for UBS to know that Value Partners was engaged in a deliberate, calculated bribery of the counterparty, KWL.

Whilst I will return to the point under Issue 3, it is convenient to clarify at this juncture that whereas the UBS-VP arrangement did not encompass the relationship between Value Partners and Mr Heininger, I accept that Mr Heininger's knowledge that Value Partners was thoroughly abusing its position did extend to the UBS-VP arrangement. In that sense, the UBS-VP arrangement can be conceptualised as a smaller circle of impropriety, within the larger sphere of the improper relationship between Value Partners and Mr Heininger. UBS only knew about, and its conscience was only affected by, the lesser scheme: it knew that Value Partners was going to deliver clients to UBS, regardless of that client's best interests. By contrast Mr Heininger knew that Value Partners was comprehensively abusing its position in relation to KWL, and was in effect seeking to itself benefit from the STCDOs in any way possible, to the direct detriment of KWL. This greater scheme, of which Mr Heininger knew, subsumes the lesser scheme to which UBS was a party. Thus even if Mr Heininger might not have known of the UBS-VP arrangement in terms, this is eclipsed by his knowledge that Value Partners was seeking to benefit from the STCDOs to the fullest possible extent, largely by taking any benefit from KWL for itself.

I interpose briefly that, for related reasons, even if the UBS-VP arrangement had given rise to an agency relationship under Issue 1, I would have concluded that the bribing of Mr Heininger did not come within the scope of that agency. The short point is that, on these facts, the bribery did not arise out of any relationship between UBS and Value Partners, but out of the relationship between Value Partners and Mr Heininger. This is not a question of causation, but of whether the bribe took place as part of one or other (or both) relationship(s). On these facts, in particular that the scheme to bribe Mr Heininger pre-dated any UBS-VP arrangement and that any such agency that exists would be a very exiguous relationship, the bribe can only have taken place as part of the relationship between Value Partners and Mr Heininger.

The second reason why I would not extend the Logicrose reservation to the present facts is that any approach whereby UBS's conscience is affected by the bribe would constitute a significant inroad on the emphasis in Logicrose on actual or Nelsonian knowledge, which would be commercially and practically unworkable. UBS is, of course, responsible for its part in the UBS-VP arrangement. However, it is in my judgment commercially unreal to use this as a springboard to make UBS legally responsible for the bribery carried out by Value Partners, given the absence of knowledge on UBS' part and the conclusion that there was no agency relationship. In practice a contracting party might, for any number of reasons, harbour a degree of suspicion about the conduct of the other party's agent, not least the latter's stated enthusiasm for placing business the contracting party's way, no doubt in both parties' mutual commercial interest. But in order for the entitlement to rescind arising from Logicrose to be workable, it must be confined to what that party knows, subject only to very limited qualification. In my view it is impracticable and unreal to introduce into commercial transactions the moral standards of the vicarage – or, put in legal terms, to impose on counter-parties the obligations of a trustee.

The third reason is that there are three important distinctions between the circumstances contemplated by the Logicrose reservation and the present facts, which it is important not to overlook in order to maintain appropriate limits on the scope of the reservation. The first distinction is that Millett J conspicuously confined the reservation to A rescinding on the basis that B's conscience was affected by dealings between A's agent and B, whereas the general principle had extended to any corruption of A's agent of which B was aware. This accords with the cases cited in support of the reservation, all of which concerned B dealing with A's agent: Grant v Gold Exploration and Development Syndicate [1900] 1 QB 233, Shipway v Broadwood [1899] 1 QB 369 and Panama and South Pacific Telegraph Co v India Rubber, Gitta Percha and Telegraph Works Company (1875) LR 10 Ch App 515. By contrast, here KWL seeks to rely on the reservation to allow rescission on the basis that UBS's conscience was affected by dealings between the agent and a third party (namely Mr Heininger).

The second distinction is that whereas the reservation in Logicrose contemplated that B would know fully of the dealings of which A complains, UBS had no knowledge whatsoever of the bribe. For the reasons already given, I do not accept that one can equate UBS's knowledge of the UBS-VP arrangement with knowledge of the bribery.

The third distinction is that the reservation appeared to be confined to instances where A's agent had been bribed, or something very closely analogous, as distinct from any other form of corruption. Again this fits with the cases cited by Millett J and, moreover, aligns with how the reservation has been understood in the academic literature: for example Lewin on Trusts at 20-067. Yet the present position is fundamentally different: Value Partners had a conflict of interest but it was not bribed, whether by UBS or anyone else.

In sum, the Logicrose reservation concerned a starkly different scenario from the present. The function of the reservation in Logicrose, as distinct from the general principle, is to ensure that B does not escape the consequences of the specific evil of secretly paying or bribing A's agent by reason of not knowing for certain that the dealings did in fact corrupt the agent or were not disclosed to A. By contrast, what KWL has sought to do in the present case is to affix UBS's conscience with a bribe to a third party of which it knew nothing.

Accordingly, I would accept what the majority describe as Lord Falconer's fall-back submission. UBS's conscience was not affected by the bribing of Mr Heininger, so KWL is not entitled to rescind on this basis.

UBS obviously had actual knowledge of the arrangement between itself and Value Partners. The question is whether this entitles KWL to rescind under the general principle of Logicrose, to which I have already referred at paragraph 340 above. This turns on whether KWL knew of Value Partners' conflict of interest arising from that arrangement.

I approach this issue on the same premises as the majority. I too would reject UBS's challenge to the judge's factual conclusions in relation to Dr Schirmer's knowledge. There is no basis on which to overturn the judge's findings in relation to the knowledge of a witness who gave evidence before him. I also agree that Lord Falconer's submissions in relation to causation do not avail UBS on this issue, for the reason that whether the UBS-VP arrangement caused KWL to enter into the Balaba STCDO is irrelevant to the entitlement to rescind under the general principle of Logicrose. That said, as I will explain below, I would broadly accept Lord Falconer's submissions in relation to causation: the point is that they do not assist UBS on this issue.

However, unlike the majority, I would accept that KWL is not entitled to rescind, as it knew of, and consented to, Value Partners' conflict of interest by reason of Mr Heininger's knowledge of that conflict being attributed to KWL.

In my view, when considering the question of attribution it is necessary to give a fuller emphasis to the context in which the issue arises, in line with Bilta: see the statements of Lord Neuberger JSC at [9], Lord Mance JSC at [41], and Lords Toulson and Hodge JJSC at [181] and [191]. The overall legal context is of course that KWL is seeking to rescind a contract entered into with UBS. Here the question of attribution determines whether KWL may say, per Millett J in Logicrose at 1261A, that it "is entitled to a further opportunity to consider whether it is in his interests to affirm [the contract]".

More specifically, applying Bilta, a crucial contextual factor is whether UBS was an accessory to the breach of duty by the corporate agent, namely Mr Heininger as director. If so, Mr Heininger's knowledge will not be attributed to KWL. To that extent I agree with the proposition which the majority derive from Bilta at paragraph 145.

However, in my view this proposition does not apply to the present facts since UBS was not an accessory or co-conspirator to Mr Heininger's breach, for four reasons.

First, it cannot be said in any meaningful sense that UBS assisted Mr Heininger in his breaches of duty. UBS was, of course, party to the arrangement with Value Partners to deliver clients to UBS. And Value Partners undoubtedly assisted Mr Heininger in the scheme to pay bribes to Mr Heininger. But UBS assisting Value Partners is not transitive: it is unreal to say that UBS assisted Mr Heininger, when it had no knowledge of that second, independent scheme centred on the bribe. I have already explained, at paragraphs 344-345 above, why the UBS-VP arrangement cannot be equated with the bribery of Mr Heininger.

The second reason, and what might be seen as a more legalistic variant of that commonsense point, is that UBS would not be liable in dishonest assistance in relation to any breach of duty by Mr Heininger. The lens of dishonest assistance is particularly relevant insofar as Millett J appeared broadly to equate the level of knowledge B is required to have in order for A to be entitled to rescind, per the general principle of Logicrose, with the level of knowledge required of B in order for A to able to bring a claim for dishonest assistance.

As to the relevant threshold, in Ultraframe Lewison J (as he then was) held that "[a]lthough it is not necessary for the dishonest assistant to know all the details of the whole design, he must, I think, know in broad terms what the design is". This was cited with approval in the Court of Appeal in Goldtrail, at [46], where Vos LJ (as he then was) added that there would not be dishonest assistance "where the dishonest assister lacked vital pieces of information known only to the primary fraudster", at [48]. Applying this approach, it seems clear that UBS could not be liable in dishonest assistance in relation to the scheme to bribe Mr Heininger, of which it knew absolutely nothing. So in this very relevant sense, the law does not consider UBS to have assisted Mr Heininger in his breach.

Third, the underlying rationale for equating a third party accomplice who assisted the corporate agent's breach of duty is that they are in the same position in relation to the company. Yet UBS's position is in no way equivalent to Mr Heininger's in relation to the bribe.

Finally, a concept of assistance that extends beyond these confines would appear to produce problematic outcomes. For example, if the arrangement between Value Partners and UBS had instead been actively to deceive Mr Heininger into agreeing to something which benefitted the former two parties, it would be perverse nonetheless to conclude that this was a tripartite relationship of assistance because one of the parties was masterminding the whole scheme. Yet in my view there is no real difference between that situation and the present facts.

KWL's claim to rescind is therefore a claim against a third party who was not an accomplice to the wrongdoing of Mr Heininger. The relevant principles therefore include those set out by Lords Toulson and Hodge JJSC at [207] and Lord Sumption JSC at [91]. In essence, one reverts to questions of context and the nature of the claim.

In my view, and taking into account the context to which I have already referred, there is every reason to attribute Mr Heininger's knowledge that Value Partners was thoroughly abusing its position in relation to KWL, and certainly was not providing disinterested advice. In essence Mr Heininger, one of two managing directors of KWL, was aware of this fact from a source which was entirely independent from UBS.

Mr Lord's response to this was that attributing Mr Heininger's knowledge to KWL nonetheless did not amount to KWL knowing of Value Partners' conflict of interest, since Mr Heininger knew nothing of the UBS-VP arrangement. I do not accept this. For the reasons given at paragraph 345 above, the consequence of attributing Mr Heininger's knowledge is that KWL knew that Value Partners would abuse its position in any way it could. The critical point is that Mr Heininger's knowledge extends far beyond the bribe, and encompasses Value Partners simply having no regard for KWL's interests whatsoever. Mr Heininger knew that Value Partners, or more concretely Mr Senf and Mr Blatz, were primarily motivated by the opportunity for personal enrichment. Moreover, Mr Heininger knew that under the letter of engagement dated 31 May 2006 Value Partners was entitled to retain all proceeds of the transactions in excess of €4.5 million. In this respect, what Mr Heininger knew about Value Partners and what UBS knew about Value Partners were very different indeed.

The position is therefore tantamount to KWL knowing that Value Partners was abusing its position in any way in relation to the proposed STCDOs, or at least so long as it did not interfere with the scheme with Mr Heininger. The possibility of an arrangement between Value Partners and UBS was therefore entirely appreciated by KWL. In this sense – of what KWL knew, as opposed to the question of whether it had any causative role – the UBS-VP arrangement did not add anything to what KWL, through Mr Heininger, knew about Value Partners by virtue of other dealings.

For these reasons, I would accept that Mr Heininger's knowledge is attributed to KWL, with the consequence that KWL had knowledge of Value Partners' conflict of interest in relation to the UBS-VP arrangement. Accordingly, KWL is not entitled to rescind on the basis that UBS had actual knowledge of the same.

Whereas this is a live issue for the majority, it follows from my conclusions on Issues 2 and 3 that KWL does not have any right to rescind that might be refused on discretionary grounds.

However, if (contrary to my view) KWL did have a prima facie entitlement to rescind under Issue 3, I would nonetheless find that it was precluded from doing so on the bases advanced by Lord Falconer, namely that:

I confine this to the hypothetical where KWL is entitled to rescind under Issue 3 alone. This is because I accept that if UBS was responsible for the bribe, whether under Issue 1 or 2, it would not be appropriate for the court in its discretion to refuse rescission. The clear tenor of the authorities is that responsibility for bribery or secret commissions produces an almost (or perhaps completely) unfettered entitlement to rescind, given the specific evil which pertains to such acts. Conversely, however, UBS's knowledge that Value Partners was subject to a conflict does permit a refusal to rescind on the two equitable grounds set out above: Hurstanger Limited and Johnson.

Since the principles relevant to these equitable bars have already been set out by the majority, it is sufficient for me to state briefly how I disagree in their application, on the premise that the discretion is being exercised afresh.

Whereas the majority consider that this bar does not apply on the basis that the Balaba STCDO was not a fair and ordinary transaction and nor was it entered into by KWL with its eyes wide open, I reach the contrary conclusion by taking a broader view of the issue of fairness.

I accept Lord Falconer's submission that it would be disproportionate to allow rescission, given the limited relevance of UBS-VP arrangement to the entering into of the STCDOs, when compared to the relationship between Mr Heininger and Value Partners. In my view this is not inconsistent with the judge's conclusion that the UBS-VP arrangement did play a role in driving the Balaba STCDO. The point is that, more holistically, I accept Lord Falconer's submission that in the overall scheme the significance of the UBS-VP arrangement was "marginal", to appropriate the language of Johnson at [80]. Value Partners and Mr Heininger had been looking for a bank through which to achieve their objectives, and UBS fit that bill. The fact that, in the course of this, UBS came to an additional arrangement with Value Partners which further helped to bring the Balaba STCDO about does not change the fact that it would be unfair to allow KWL to rescind when the contract reached was exactly the sort which had been envisaged all along.

In my view there is also inequitable conduct on the part of KWL which is so closely connected with the relief sought (namely rescission) that relief should not be granted.

The inequitable conduct in question is the deceit by Mr Heininger in making the fraudulent misrepresentations identified by the majority at paragraph 177. It was not disputed that KWL was vicariously liable for this conduct, with the result that KWL's hands are, through Mr Heininger, relevantly sullied. Moreover, the judge accepted at [691] that the Balaba STCDO would not have been concluded without this representation, so the deceit is plainly connected to the contract. The fact that Mr Heininger's deceit was procured by the bribe is irrelevant to this enquiry, given the hypothesis that UBS is not responsible for the bribe under either Issue 1 or 2. Thus the position of KWL is that, despite having deceived UBS into entering into the contract, the contract ought to be rescinded at KWL's election. In my view equity debars KWL from doing so.

I agree with the essential conclusion reached by the majority, namely that UBS's deceit claim fails because the deceit caused no loss. Insofar as the claim relates to losses suffered by UBS when hedging the LBBW and Depfa STCDOs, I agree with the majority's reasoning at paragraph 190 entirely.

However, in relation to the losses suffered by UBS when hedging the Balaba STCDO, my reasoning differs. Since I have concluded that the judge erred in concluding that KWL was entitled to rescind, and instead UBS was entitled to enforce the Balaba STCDO, UBS has suffered no loss. This is because UBS's liability on the CDS hedges initially entered into was effectively equivalent to KWL's liability on the Balaba STCDO. The result is that the losses from those CDSs which resulted from the deceit would be extinguished by the recovery under the STCDO.

In light of the unanimous conclusion of this court in relation to Issue 7, the question whether KWL would be indemnified by UBS in respect of any liability to the intermediary banks does not strictly arise. However, my conclusion that KWL is not entitled to rescind the Balaba STCDO would give rise to a live question as to whether KWL is entitled to recover its liability to UBS under the Balaba STCDO through a claim for dishonest assistance against UBS.

In any event, I would reject all of KWL's claims in relation to dishonest assistance for the same two reasons. First, UBS's dishonest assistance did not cause KWL any loss. Second, it would follow from my conclusions on Issue 3 that KWL consented to Value Partners' conflict, in relation to the UBS-VP arrangement, and accordingly there is no claim for dishonest assistance at all. Thus if the dishonest assistance claims in respect of the LBBW and Depfa STCDOs had arisen, I would have rejected them for these reasons.

In my view, the UBS-VP arrangement did not cause KWL to suffer any loss in relation to the Balaba STCDOs or the subsequent LBBW or Depfa STCDOs, even if it helped to bring about those STCDOs in a looser sense. Accordingly, since causation is a prerequisite to a claim for dishonest assistance (see the citation of Gwembe Valley by the majority at paragraph 196), KWL's claim for dishonest assistance fails.

This conclusion emerges starkly as soon as one asks the question of what would have happened but for the UBS-VP arrangement. In my view it is absolutely plain on the findings of fact made by the judge that, due to the relationship between Value Partners and Mr Heininger, KWL would have entered into comparable STCDOs in any event. As the judge found at [591]: before any contact with UBS took place there was a corrupt relationship between Value Partners and Mr Heininger; these individuals were already looking to restructure the CBLs for their (as opposed to KWL's) benefit; and personal enrichment was throughout the primary motivation. As I have said, the true position is that Value Partners and Mr Heininger were looking for a bank through which to achieve their objectives, and UBS fitted that bill.

In that context, and put at its absolute highest, the only loss which the UBS-VP arrangement could theoretically have caused was if KWL entered the STCDOs on less favourable terms than it otherwise would have obtained. However, that any such loss was suffered is inconsistent with the judge's other findings and in any event extremely doubtful. For example, the amount which KWL was ever going to receive from the STCDOs was capped at €4.5 million, due to the letter of engagement dated 31 May 2006 but actually signed on 14 June 2006. This letter was the result of the relationship between Value Partners and Mr Heininger and could not have been caused by the UBS-VP arrangement. The capping of the amount which KWL would receive is simply not explicable by reference to Value Partners attempting to deliver clients to UBS. If anything, it runs directly contrary to that arrangement since Value Partners seeking to take more profit at the expense of the client would make any deal less attractive for the client. Conversely, it is entirely explicable in the context of the relationship between Value Partners and Mr Heininger, as indeed the judge appeared to recognise at [54]. Since KWL received this amount on the terms of the STCDOs ultimately entered into, its position would not have been any better without the UBS-VP arrangement.

In my view, the judge did not in fact purport to decide the question of what would have happened but for the UBS-VP arrangement or, alternatively, insofar as the judge did consider the issue of causation he erred in principle in his approach. What the judge accepted was that the arrangement helped to bring about the Balaba STCDO in a looser sense, but he failed to apply his mind to the question of what would have happened without the arrangement.

The closest the judge came to discussion of causation are his remarks, in the passages reproduced by the majority at paragraphs 137 to 140, that: (i) the Balaba STCDO was "also driven by the corrupt relationship between Mr Bracy and Value Partners"; and (ii) the UBS-VP arrangement "continued to affect" and "continued to operate in" the later STCDOs. However, I do not think either represents a finding that KWL's losses would not have been suffered without the UBS-VP arrangement:

Accordingly I would accept, overturning the judge's findings if necessary, that the only proper conclusion was that the UBS-VP arrangement did not cause the losses which resulted from any of the STCDOs.

Whilst my conclusion on Issue 3 avoids the inconsistency between the majority's conclusion on attribution and KWL's concession that it was vicariously liable in respect of Mr Heininger's deceit (see paragraph 183), I confess to a discomfort of my own. Namely, my view that KWL is not entitled to rescind because it had knowledge of Value Partners' conflict would appear to be inconsistent with Lord Falconer's concession that UBS was liable in dishonest assistance in relation to that arrangement. If Mr Heininger's knowledge is attributed to KWL, KWL must be taken to have consented to the breach of fiduciary duty by Value Partners in relation to the UBS-VP arrangement. In that light, I do not see how UBS could nevertheless be liable in dishonest assistance for the same.

I also differ from the majority on this discrete issue, which, in light of my conclusions above, is a live issue. Again the relevant submissions, legal principles and findings by the judge are set out in the majority judgment.

I agree with the majority, for the reasons given, that the judge did not err in finding that UBS GAM negligently managed the portfolio. However, I do not agree that the judge's conclusions in relation to causation are sustainable. In my view the judge improperly (though perhaps understandably, as this was, for him, a distantly obiter issue) short-circuited the necessary analysis.

The correct comparator was a portfolio which was managed by a competent portfolio manager, as indeed I understand the majority to accept at paragraph 285. However, in my view the judge did not use this comparator and instead erred by adopting an unmanaged portfolio as the comparator. This is supported by the judge's remarks at [910] that "a much simpler approach is appropriate" (than that suggested by UBS) and his open acceptance of the chosen method as "broad brush". In addition, I cannot accept that it should be assumed that the competent manager would have achieved at least as good an outcome as an unmanaged portfolio, either for the judge's reason that "UBS [GAM] can hardly suggest otherwise" or for the reasons given by the majority at paragraphs 286 to 294.

Accordingly, I therefore would have remitted this question of causation to a different trial judge.

For the above reasons, I would have allowed UBS's appeal. Since, however, the majority take a different view, the appeal will be dismissed.