Dresdner Kleinwort Ltd & Anor v Attrill & Ors

This appeal arises from two sets of claims by 104 employees (“the claimants”) who worked for Dresdner Kleinwort Investment Bank (“DKIB”). They successfully contended before Owen J, following a sixteen day trial, that they had been wrongfully denied their contractual entitlement to certain discretionary bonuses for the calendar year 2008 promised by their employers. At stake are sums totalling more than €50 million.

Dresdner Bank AG (“DBAG”) was, until May 2009, an international bank incorporated in Germany comprising retail, corporate, private client and investment banking businesses. It was wholly owned by Allianz SE (“Allianz”), a German insurance company.

On 31 August 2008, it was announced that Allianz had agreed to sell DBAG to Commerzbank AG (“Commerzbank”). The sale was originally intended to take place in two stages: the first was to be in January 2009 and was to involve the transfer of 60% of the shares in DBAG to Commerzbank; the second was to take place towards the end of 2009 and was to involve the transfer of the remaining shares. However, in November 2008, an accelerated timetable was agreed under which Commerzbank would acquire all the shares in one tranche in January 2009. The sale was completed on 12 January 2009. From that date, Commerzbank was the sole owner of DBAG. Commerzbank and DBAG formally merged under German law in May 2009, whereupon DBAG ceased to exist as a separate legal entity.

DBAG’s investment banking business, which was not a separate legal entity and did not directly employ any staff, had originally been known as Dresdner Kleinwort Wasserstein and, subsequently, DKIB. Dr Jentzsch was its Chief Executive. He was also a director of DBAG.

DKIB, the investment banking arm, was broken down into a Front, Middle and Back Office. The Front and Middle Offices were directly concerned in the business of DKIB. The Back Office comprised the Back Office “support functions”, such as IT, Legal, HR, Payroll, and Security, all of which supported DKIB’s business operations.

Dr Jentzsch was responsible for the Front and Middle Offices of DKIB. These comprised some 3,000 employees worldwide, with approximately 1,300 of these in London. There was a slightly larger number of staff worldwide employed to perform the support functions and they reported not to Dr Jentzsch, but to another DBAG board member.

All DKIB Front and Middle Office staff (including all the claimants), and all Back Office staff who worked for DKIB in the support functions had employment contracts with, or were seconded to, DKL. This was a service company wholly owned by DBAG. The HR function was responsible for employee relations and terms and conditions. The global head of HR for DKIB, who was responsible for HR issues for the Front and Middle Offices and for the support functions, was Mr Mark Hindle.

The contracts of employment between DKL and each of the claimants stated (in terms varying from employee to employee) that the employee would be eligible for consideration by the employer for such discretionary awards as in the employer's absolute discretion it saw fit to award.

Sections 1 to 20 of the Employee Handbook were incorporated in the contracts of employment. Section 1.4 conferred upon the employer the power unilaterally to vary the contract in the following terms:

“The Company reserves the right to vary the terms and conditions described in this handbook and the terms and conditions of your employment generally. Such changes can only be made by a member of the Human Resource Department and must be communicated to you in writing. When the change affects a group of employees, notification may be by display on notice boards or Company Intranet.”

The proper construction of this clause is one of the issues in this case.

Section 33 deals with “Discretionary Bonus Awards”. So far as relevant it provides:

“33.1

You may be eligible to be considered for the payment of an annual discretionary award if, but only if, on the day bonus awards are actually paid (which is usually in the first quarter of the calendar year), you are both employed by the Company and not within a period of notice of termination of your employment.

33.2

Whether an award is made and the amount of such award, if any, is at the absolute discretion of the Company.”

The clause goes on to identify a series of factors which would influence the way in which the discretion to make the award would be exercised. It states that awards are made to support the Bank’s commercial objectives at the given time and are likely to reflect many different and competing factors. These include the financial performance of the Company, the need to incentivise employees, the strategic needs and priorities of the business, and the need to retain employees.

Similarly there is no dispute that the established procedure, followed in many previous years, was to allocate a bonus pool, award individual bonuses in November, communicate the allocation and award in December and pay the cash element in January, provided that the employee was still employed at that time and not under notice. There was no prescribed size of bonus pool.

On 17 March 2008 Mr Diekmann, Chairman of the DBAG supervisory board and CEO of Allianz, announced the intention to separate DBAG’s commercial banking division from its investment banking division. This was a prelude to DBAG leaving investment banking, either by selling DKIB, drastically cutting back its operations, or winding it down.

Notwithstanding that announcement, staff were told that the possibility of a sale was mere speculation; there was, nonetheless, a general belief that this might occur and it led to widespread uncertainty about the future of staff. As Dr Jentzsch said in his evidence at trial, there was a serious risk of a mass exodus of staff. The Head of “Fixed Income, Currency and Commodities (“FICC”)” within DKIB confirmed in early May that key individuals were leaving and he expressed the view that unless active steps were taken to retain staff, the business could collapse in just a few weeks.

On 23 May 2008 the Financial Services Authority (the “FSA”) wrote to Dr Jentzsch as the Chief Executive Officer of DKIB notifying him that the Bank’s UK regulated entities had been placed on the FSA’s “Firm Watch List” because of the risks posed by the proposed restructuring. The FSA expressed the concern that there was no comprehensive strategy for the future in place and that DBAG faced an uncertain future. The FSA referred in particular to:

“…the continuing uncertainty among management and staff [which] could lead to a significant number of key individuals leaving or becoming disaffected…”

The FSA said that the Bank would not be taken off the Firm Watch List until there had been an assessment of the key risks and mitigation plans had been put in place.

At a meeting of the Management Operating Committee of DBAG on 11 June 2008, this concern was addressed. It was mooted that there should be “some kind of staff retention scheme” to prevent the damaging departure of disaffected staff. In his reply to the FSA dated 30 June 2008, Dr Jentzsch acknowledged that one of the key risks of the DBAG restructuring scheme was loss of staff. After setting out certain statistics he accepted that:

“the risk of defections has risen considerably. A retention programme is therefore being discussed internally as well as with Allianz and we expect its terms to be finalised shortly.”

Dr Jentzsch also sent a paper to the DBAG Board in which he noted that any prospective purchaser would run the risk that the investment banking division would be a withering asset.

A number of possible retention programmes were considered and a proposal was formulated on 25 July 2008 which was in essence the scheme later adopted. It proposed:

“for 2008 a de minimus [sic] pool of €400m (75% of 2007) to be set aside for DKIB.”

At a meeting with the FSA on 5 August 2008 Dr Jentzsch indicated that he would shortly announce a minimum bonus pool for DKIBwith funding to be guaranteed by Allianz. A further version of the bonus pool scheme dated 12 August 2008 was discussed at a meeting of the board of directors of DBAG on the same day. The relevant paper noted that:

“In order to stabilise the DK business it is essential and necessary to formally communicate the pool that has been secured for the staff base.

If agreed it would be our intention to formally communicate to the entire staff base of [DK] via a business update that Allianz have recognised the uncertainty and impact the strategic discussions are having on the staff base and have agreed to set aside a de minimus [sic] overall [DK] bonus pool for 2008 at €400m.”

The retention pool was formally approved at the board meeting. The minute recording it is in the following terms:

“A minimum bonus pool, which should be announced in the coming week, is needed to ensure employee stability. All risk policies must be strictly observed in order to prevent the pool being 'generated' by taking excessive risks. The minimum bonus pool should be guaranteed and allocated on a discretionary basis. Revenues of 2,327.5 million formed the basis of the cash pool of 400 million, with corresponding increases. Independently of the minimum pool, but related to it at a content level, approximately 60 front office staff should receive individual bonus guarantees.”

On 18 August 2008 Dr Jentzsch announced the creation of the minimum bonus pool at what was termed a Town Hall meeting. He was in Frankfurt but the announcement was simultaneously broadcast on the Company’s intranet and could be seen on desk top computers or large screens in other locations around the world. The announcement was not recorded but contemporaneous notes were made by two members of staff. It is not necessary to set out their records of what was said. There was a dispute about precisely what Dr Jentzsch had said before the judge, and he made certain findings about that which have not been challenged on appeal. He concluded that the announcement included statements to the following effect that:

Dr Jentzsch took steps to ensure that senior managers spread the message to those who were under their supervision. He initiated what the judge described as an “information cascade”. Furthermore, there were answers given on the Frequently Answered Questions (“FAQs”) site on the Company Intranet, which confirmed the creation of a guaranteed retention pool and this was also confirmed at meetings by individual managers. The parties disputed before us how many employees would have heard the 18 August announcement, but whatever that number, there was a deliberate and conscious attempt to disseminate it widely throughout the affected staff – it could hardly have achieved the objective of retaining staff unless they knew about it – and it was obviously of great interest and concern to the employees affected. The judge concluded on ample evidence that the message spread like wildfire.

On 31August the acquisition of DBAG by Commerzbank was announced. On 15 September Lehman Brothers collapsed, precipitating a worldwide crisis in the banking sector, the ripples of which are still being felt. On 3 November Commerzbank received some €8.2 billion from the German Government’s Stabilisation Fund and that was subsequently increased so that by January 2009 the figure was in excess of €18 billion. Notwithstanding these developments, both domestic and international, staff were continually told both before and after these developments that the bonus pool remained in place. For example, in the FAQs the message was reiterated on 12 September 2008 and 3 October 2008. On 20 October 2008 Mark Hindle, the Head of Human Resources of DKsent an email to all employees stating:

“The communication date for bonus awards for both [DK] (Front Office employees) and Aligned Functions (support employees) will be Friday 19 December 2008.

The bonus pool for the Front Office has already been communicated by Stefan Jentzsch in his updates. The bonus pool for Functions will be comparable with last year's, adjusted for headcount movements. Individual bonus awards continue to be discretionary and determined by the relevant management, and last year’s award should not be taken as any indication of the level of any bonus award for 2008.

Bonuses communicated on 19 December 2008 will be paid in full with January salaries through the January payroll. The 2008 bonus awards will be awarded in cash and subject to statutory deductions ...”

On 12 November at a Town Hall meeting Mr Blessing, the CEO of Commerzbank, was specifically asked whether he could give an assurance that Commerzbank would not interfere in the awarding of the bonus pool which had been set aside for the employees and he replied that it was a decision for DBAG’s management. On 17 November, following a meeting of the DBAG Steering Committee, Dr Jentzsch sent a memorandum to various DBAG managers in which he expressed the view that the purpose for which the bonus pool had been established had been achieved:

“…. That this measure has fulfilled its purpose can already be seen on the basis of significantly reduced personnel fluctuations.”

He went on to add that in his view it would be dishonourable for the bonus pool now to be reduced.

However, in late November/early December although senior line managers and compensation committees reviewed each employee’s performance and decided on the level of bonus there was active discussion at Board level in DBAG and Allianz as to whether the bonus pool should be retained and the commitment made by Dr Jentzsch should be honoured. I return to consider some of the detail relating to this development later in the judgment. The minutes of a meeting of the Board of DBAG on 26 November raised the question whether modifications could be made to the bonus pool without involving a breach of contract. The minutes record that this needed to be clarified. It was recognised that “changes to the bonus pool will raise not only the question of how these can be explained to employees but also questions of credibility, fairness and reputation”.

On 4 December 2008 there was a meeting of the board of DBAG at which the Board resolved to introduce the “material adverse change” (“MAC”) clause into the bonus letters. The Board had already resolved that these letters should be sent out on 19 December. Those present considered a draft of the bonus letter and discussed changes to it. They resolved:

“... to leave the volume of the [DK] bonus pool resolved on August 12, 2008 unchanged at present. However, this resolution is subject to the proviso that a clause is included in the bonus letter stating that the bonus payment will be adjusted if material negative deviations in DKIB’s revenue and earnings as against the existing forecast for the months of November and December 2008 are established during the preparation of the annual financial statements for 2008. The review of this issue will be conducted in January 2009 under the leadership of Stefan Jentzsch.”

On 8 December 2008 Dr Jentzsch wrote to his fellow directors of DBAG expressing his strong opposition to reducing the pool. He said:

“I expect and apprehend that a retraction of the commitments will destroy the trust of the employees in the executive management fundamentally, finally and irrevocably, with the risk of high operating instability and corresponding consequences for the income and capital of the bank.”

Notwithstanding his strong objections, the clause was included in the letters sent on 19 December 2008 to each employee of DKIB, cast in the following terms:

“Dear ,

A discretionary bonus for 2008 under the arrangements given below has been provisionally awarded at [x] EUR

The provisional bonus award stated above is subject to review in the event that additional material deviations in Dresdner Kleinwort's revenue and earnings, as against the forecast for the months of November and December 2008, are identified during preparation of the annual financial statements for 2008 i.e. that Dresdner Kleinwort's earnings position does not deteriorate materially in this period. This will be reviewed in January 2009 by Stefan Jentzsch. In the event that such additional material deviations are identified, the Company reserves the right to review the provisional award and, if necessary, to reduce the provisional award……

You will receive a detailed statement confirming your final bonus award in local currency in February 2009. This will be paid together with your salary in the February 2009 payroll.”

Dr Jentzsch addressed a Town Hall meeting the same day and sought to reassure employees that the bonus pool remained in place. He told staff that the bonus pool had served its purpose of stabilising staff and that he was not happy with the proviso. He emphasised that the review referred to in the letter would be conducted by him, and that Mr Blessing and Mr Reuther were honourable men who would stick to the bonus commitment publicly made. He concluded his statement with the observation that:

“…. I remain confident that in the end we will deliver on the overall bonus pool amount for [DKIB] and the individual promises made to each of you.”

The acquisition of DBAG by Commerzbank AG was completed on 12 January 2009 and Dr Jentzsch was replaced by Mr Reuther. Two days later there was a Town Hall meeting addressed by Mr Blessing, who was now both CEO of Commerzbank and Chairman of the Board of DBAG, to explain the problems which the bank faced.

On 27 January the minutes of the DBAG board meeting record that the Company would “undertake a more far reaching review of the bonus pool and a reduction in the direction of a target of EUR 300m”.

On 10 February 2009 Dr Jentzsch met members of the board of DBAG to discuss a document dated 9 January 2009 and entitled “Bonus Pool [DK] Business Performance”. The full board of DBAG met later that day with Dr Jentzsch in attendance by invitation, since he was no longer employed by the Bank. They discussed three possible alternatives relating to the guaranteed pool, namely payment according to the distribution mechanism as originally planned, payment on the basis of a reduced bonus pool, and payment on the basis of individually agreed legal obligations only.

On 17 February 2009 there was an extraordinary meeting of the board of DBAG. It started by discussing the bonus decisions taken in relation to Commerzbank. There was then an in depth discussion of the bonus pool for DK. The decision was taken to reduce the discretionary bonuses by 90%. Those who received individual guaranteed bonus received no discretionary bonus at all.

On 8 September and 27 November 2009 the proceedings to which these appeals relate were commenced. On 28 January 2010 DKIB and Commerzbank issued applications seeking orders under CPR Rule 24.2 summarily dismissing the actions against them on the grounds that the claimants had no real prospect of success. The applications were heard by Simon J who on 28 May 2010 summarily dismissed any part of the claims which relied on the announcement made on 18 August 2008 on the ground that it did not create any enforceable contractual rights on which the claimants could rely. However, the judge concluded that the claims based on the bonus letters raised issues of fact which should go to trial. His decision was successfully appealed to the Court of Appeal by the claimants in a judgment delivered by the Chancellor ([2011] EWCA Civ 229) and as a result there was the trial before Owen J. He heard the case over some three weeks in January and February 2012.

There were three principal issues before the judge, and they remain the issues in the appeal. They are:

The first issue had a number of overlapping sub-issues. The first was whether the 18 August announcement, whether considered independently or in conjunction with other communications, was an exercise of DKIB’s power of unilateral variation under section 1.4 of the Employment Handbook. The second was whether the announcement was intended to be legally binding in any event so as to be capable of having contractual effect. The third was whether there was any obligation on the part of the claimants to communicate acceptance of the appellants’ “offer” to guarantee a minimum bonus fund and if so, whether they had done so.

It is important to understand how these issues interlink. If the guaranteed bonus pool had been lawfully introduced pursuant to clause 1.4 then the question of acceptance did not arise. It would still, however, be necessary to establish that there was an intention to create legal relations. If there had been no variation pursuant to clause 1.4, it was conceded that it would still be possible for a variation to be made outside that procedure, although in those circumstances it would be necessary to establish that the offer was accepted or alternatively, that the need for acceptance had been waived.

There were two additional points which were raised as part of the argument on intention to create legal relations but which logically, if correct, would defeat the claimants’ case on their own. These are whether the bonus pool was appropriate for incorporation into the contract and the related issue whether it was sufficiently certain to be enforced. I will deal with those submissions after first considering the clause 1.4 issue.

I have set out the clause above, but will repeat it here for ease of reference:

“The Company reserve the right to vary the terms and conditions described in this handbook and the terms and conditions of your employment generally. Such changes can only be made by a member of the Human Resource Department and must be communicated to you in writing. When the change affects a group of employees, notification may be by display on notice boards or Company Intranet.”

The construction of the clause was strongly disputed before the judge. There were four points where the parties advanced different interpretations. They were as follows:

The claimants submitted before the judge, and before us (by way of a respondents’ notice where necessary) that the clause was complied with in three principal ways.

First, the announcement on 18 August itself satisfied the requirements. It was a notification on the Company Intranet, satisfying the method of communication to a group of workers. It was not necessary for the HR Department to be involved and it was not necessary for any separate written communication to each individual member of staff.

Second, and in any event, the combination of the announcement and the notification by Mr Hindle, the Global Head of HR, on 20 October, which was also communicated on the Intranet, together satisfied the requirements of the clause. If and insofar as HR involvement was required, this provided it.

Third, the claimants submit that there were frequent comments on the Intranet made by managers and in the answers to FAQs which constituted a display of the announcement satisfying the requirements.

In analysing the clause, the judge accepted a submission from the Bank that since the clause allowed changes adverse to an employee to be made, it should be read strictly. He also agreed with the defendants that HR involvement was required even in the case of a group notification. Accordingly, he rejected a submission that the announcement on 18 August was of itself enough to constitute an effective variation of the contract. It was a notification by Company Intranet but Dr Jentzsch was not a member of the HR department and accordingly the announcement did not of itself comply with all the requirements necessary to constitute a valid amendment.

However, the judge went on to find that the combination of the 19 August announcement and the 20 October communication from Mr Hindle together satisfied the requirements of the clause. The October communication constituted a confirmation by the global head of HR of the variation in relation to discretionary bonuses. It referred back to the 18 August announcement and was an endorsement of that announcement satisfying the requirements of clause 1.4. The judge therefore impliedly rejected both the submission that separate individual communications in writing were required and the submission that compliance could not be achieved by a combination of communications.

The judge, however, did not accept that the other communications made on the Intranet, such as the FAQs, could be relied upon. The judge referred to the evidence of Mr Hindle that FAQs are issued by DKIB corporate communications and although reviewed by HR, were not “a tool used by HR to amend or introduce new terms of employment”.

The battle lines established below have been renewed before this court. I agree with the submissions of Mr Linden that in accordance with well established principles the clause should, in the case of ambiguity, be construed in favour of a construction which limits the scope of the clause since it confers a right on the employer unilaterally to make detrimental changes to the terms and conditions of employment. Usually this means that it will be construed against the employer because he will be seeking to rely on it. The unusual feature of this case is that it is the employees who are seeking to rely on the clause and are therefore urging a broader construction, but of course the meaning of the clause cannot vary depending upon who is praying it in aid. The significance of the principle is, however, limited: it is relevant only where there is genuine ambiguity: see Chitty on Contracts, 39^th Edn para.12-083. The fundamental principle is that the clause should be interpreted objectively so as to give it a sensible and workmanlike construction.

Adopting that approach, in my judgment it is plain that the second and third sentences are intended to be read disjunctively. There would be no point in indicating that group communications may be made by notice board or Intranet if every change had to be effected by personal communication in writing. Consistent with that analysis, and contrary to the conclusion of the judge, I do not consider that group changes need to be either made or notified by a member of the HR Department if the notice board or Intranet option is used. The purpose of HR having a role is no doubt, as Mr Linden asserts, because of the real risk that individual managers will make promises to members of their team without proper authority. HR involvement ensures that this should not happen. However, where a change is effected by a communication made to a group of employees openly and in public, there is not the same risk. No doubt there will need to be authority to use the Intranet system but there is no obvious reason why HR should necessarily be involved.

But even if HR involvement is a requirement, I accept the submission of the claimants that it would be satisfied in this case. First, the communication was made with the approval of the Board of the Remuneration Committee and the HR Director was a member of that Board. It would be artificial, in my view, to require any other formal approval or endorsement by HR, and I do not read the clause as requiring HR actually to make the notification on the Intranet. Further and in any event, I agree with the judge that assuming that HR involvement was necessary, the clause was complied with by the combination of the August announcement and Mr Hindle’s 20 October message on the Intranet. In my view, there is no reason why compliance cannot be achieved by two documents. I would accept in those circumstances the variation would not be effective until the necessary endorsement or confirmation had been taken (and insofar as the judge suggested otherwise, I would not agree) but that is of no materiality in this case.

The one area that has caused me some difficulty is whether [the promise/the announcement/the communication] can properly be said to have been displayed on the Company Intranet as a result of it having been orally communicated. If there is no display, the clause is not satisfied whether or not HR involvement is necessary. I see force in the submission that normally the concept of displaying something envisages that it will be available at least for some period of time, as a point of reference. But this gives rise to its own difficulties: for how long must it be displayed in order for the condition to be met? There is no obvious answer to that question, and I have concluded that the broadcast itself would suffice, even if it could not be seen after the initial announcement. In fact, however, the evidence suggests that it could and if that is so it reinforces my conclusion that there was here an effective notification.

For these reasons, therefore, I conclude that the variation was effected in accordance with clause 1.4. The Bank submits that even so, there are other reasons why there is no contractually binding obligation.

I first turn to consider two preliminary issues which are raised, as I understand it, both in their own right and also as part of the contention that there was no intention to create legal relations with respect to the bonus pool. First, the Bank contends that the relevant term was not appropriate for incorporation into the individual contract of employment. The second and related point is that the term is too uncertain.

As to the aptness point, Mr Linden submits that the announcement was to the entire workforce and did not, and was not intended to, confer rights on any individual employee. It gave some indication of the minimum size of the cake but not of the size of individual slices. The employees had no right to any particular bonus since that depended on the exercise of management discretion. This was not the stuff of individual contracts.

Mr Hochhauser QC, one of the counsel for the claimants, points out that the right to be considered for a bonus is a valuable right which is perfectly capable of constituting a contractual term. It is then implied that the exercise of discretion shall not be carried out arbitrarily or in bad faith: see Clark v Nomura [2000] IRLR 766 and Horkulak v Cantor Fitzgerald International [2004] EWCA Civ 1287 [2005] ICR 402. He submits, and I agree, that in principle there is no reason why the promise that the Company will at least pay bonuses up to €400 million to the relevant employees should not also be enforced as a contractual term defining one of the contours regulating the exercise of that discretion.

The Chancellor, when hearing the appeal from Simon J, considered that the term was in principle capable of being incorporated (para 29):

“The essence of the Town Hall meeting held on 18^th August was that it quantified and qualified one uncertainty in the bonus procedure, namely the size of the bonus pool. Theoretically, it could have been any amount between nothing and the maximum the directors might properly so apply. Given the uncertainties which the directors of DBAG and the FSA had been discussing in May and June 2008, to set up a guaranteed minimum bonus pot of a specific amount was, and was intended to be, a substantial benefit to all those who might be eligible for the future award of bonuses and an inducement to stay. I see no reason why a promise of a guaranteed minimum bonus pool cannot be contractually binding even though individual employees cannot at that time point to an entitlement to a specific bonus payable out of it. At the very least each of them would be entitled to nominal damages for breach.”

Since the court was simply determining whether the claim was arguable, I do not think that this is strictly binding on us as part of the ratio. Nonetheless it was the considered view of the court after hearing legal argument and I would attach very great weight to it. I respectfully agree with it and in my judgment, there is nothing in this point.

The submission on uncertainty was based on the assertion that the announcement of the bonus pool left many problems not determined. For example, it is submitted that there is uncertainty as to whether the individual guaranteed fixed bonuses should come out of the fund; whether the bonus should be paid by way of shares or cash; and what proportion of the fund could be held back for contingencies, it being accepted that an element of the fund could be dealt with in that way. In my judgment, these problems are largely dealt with by the finding of the judge that the fund would be dealt with “in the usual way”. For example, that confirmed that individual fixed bonuses would be paid from the fund as well as the discretionary bonuses. It would admittedly leave some imprecision, for example, on the question of how much could be withheld for contingencies. But I have no doubt that the parties would recognise that it would be a reasonable figure of the kind typically withheld for this purpose in the past. The fundamental principles of the scheme were entirely clear and the fact that there were some loose ends does not in my view begin to constitute a degree of uncertainty necessary to defeat the parties’ intention that the agreement should be capable of enforcement. The court will be slow to hold that otherwise contractually enforceable obligations cannot be enforced because they are too uncertain: see Hillas v Arcos (1932) LT 503, 512 & 514 per Lord Tomlin and Lord Wright.

I will first set out the analysis of the judge on this point. He started from the following three premises. The first was that the question whether there is an intention to create legal relations must be considered objectively. He referred to the following passage from the judgment of Lord Clarke in RTS Flexible Systems Ltd v Molkerei Alois Mueller GmbH and Co KG (UK) Productions [2010] UKSC 14; [2010] 1 WLR 753, para 45:

“Where there is a binding contract between the parties and, if so, upon what terms depends upon what they have agreed. It depends not upon their subjective state of mind, but upon a consideration of what was communicated between them by words or conduct, and whether that leads objectively to a conclusion that they intended to create legal relations and had agreed upon all the terms which they regarded or the law requires as essential for the formulation of legally binding relations.”

To similar effect is the observation of Lord Bingham CJ as he then was in Edmonds v Lawson [2000] All ER 31, at para 21 when he said:

“Whether the parties intended to enter into legally binding relations is an issue to be determined objectively and not by enquiring into their respective states of mind. The context is all important.”

Second, the judge held that the onus of proving that there was a lack of intention to create legal relations would be on the Bank since they were asserting that no legal effects were intended. He relied for this proposition on certain observations of Megaw J in Edwards v Skyways [1964] 1 WLR 349 at 355, and Aikens J in Mamidoil Jetoil Greek Petroleum: SA v Okta Crude OilRefinery AD [2003] 1 Lloyd Rep 554.

Third, he emphasised what Lord Bingham CJ had said in Edmonds v Lawson, namely that “the context is all important”.

In considering context the judge noted the following features: that the guaranteed bonus pool was created to stabilise the work force; that it arose in part in response to FSA requests that a retention strategy should be put in place; that both the decision to establish the pool and the method of communicating it had been approved by the Compensation Committees of both DBAG and Allianz; that the reference to a “guaranteed” minimum bonus pool was strongly indicative of an intention to undertake a binding legal obligation; and that subsequent statements which on numerous occasions reasserted the existence of the guaranteed bonus pool reinforced the conclusion that there was an intention to create legal relations.

The judge then referred to a number of arguments advanced by the defendants as to why the inference should be drawn that there was no intention to create legal relations. Two of these were also advanced in their own right, namely that the term was not appropriate for incorporation into the individual contract, and that it was too uncertain to be given legal effect. I have already dealt with these issues and have explained why I do not accept them. Nor do I consider that they have any weight when determining this question.

A third ground was to the effect that there was a procedure in clause 1.4 for varying the contract and that it was highly relevant that this was not used. The Bank contends that the judge failed properly to address this issue at all. The claimants disagree and say that the judge identified it in his judgment at para 146:

“Secondly, the defendants pointed to the fact that the announcement was not communicated in writing, arguing that if there was an intention to create a binding obligation, then it could have been communicated to the entire workforce via the Company Intranet, as Mr Hindle did in his email of 20^th October.”

A fourth and related argument was that the Town Hall was an informal forum for the exchange of information. It was submitted that the parties would not have anticipated that a formal variation of a contract would be effected in this way. Moreover, it was alleged that many of the staff would not have seen the announcement for a variety of reasons and there was no recording or other official record of it.

A fifth ground was that the announcement had the potential for wholly unreasonable consequences because, if unsuccessful, it would mean that many of the more efficient workers who could readily obtain alternative employment would leave and the bonus pool would be available for distribution to a dwindling number of less efficient staff members. Taken to its logical conclusion, the last man standing would scoop the whole pool. Mr Linden submitted that the defendants could never have intended to make a binding commitment to distribute the pool in those circumstances. Equally, it was unreasonable to assume that there would ever have been an intention to create a binding legal obligation to distribute the bonus pool irrespective of losses made by DBAG.

The judge took on board these points but he nonetheless reached the clear conclusion that:

“.. the inference plainly to be drawn from both the terms which the announcement was made and from the context in which it was made is that, viewed objectively, it was made with the intention of creating a legally binding obligation for those employed in DKIB.”

In reaching that conclusion he specifically answered each of the issues raised by the defendants. As to the submission that the Town Hall announcement was not the kind of procedure one would expect to adopt to create a binding legal commitment, the judge accepted that there was force in the submission. But he observed that it was necessary to weigh against that the fact that the Board of DBAG had approved this method of formally communicating the terms of the resolution to establish the guaranteed minimum bonus pool. Moreover, in his witness statement Dr Jentzsch had explained that it was the “quickest and most effective way in which to announce the minimum bonus pool to all staff” and added that it was the primary means by which he communicated important information to staff.

Nor was the judge persuaded by the alleged unreasonable consequences. He said that DBAG had taken a calculated risk that the bonus pool would encourage staff to remain and ensure that the investment banking division operated as a going concern until the point of sale, a strategy which had indeed proved successful. It was a calculated gamble but did not point against any intention to create legal relations. Similarly, the notion that the term might lead to bonuses being paid notwithstanding the fact that the Bank may make losses was not a matter of weight, not least because the Bank was already bracing itself for very substantial loss, exceeding by far the amount in the bonus pool, even when the promise was made. The purpose was to prevent those losses becoming even more catastrophic.

The judge added that he reached this conclusion quite independently of the subjective understanding of the parties, but he was confirmed in his conclusion by the evidence of Dr Jentzsch and Mr Hindle. They both understood the Bank to have made a binding commitment. Dr Jentzsch said in evidence that “commitment for me always implies legally binding otherwise it’s not a commitment.”

The grounds of appeal largely reiterate the arguments unsuccessfully advanced before the judge but they also assert that the judge erred in various ways. First, it is submitted that the judge was wrong to conclude that the burden of proof was on the Bank to show that there was no intention to create legal relations. Mr Linden accepted that that is the position where there is a bilateral arrangement but submitted that it is different in the case of a unilateral contract.

His second argument is premised on the assumption that the variation was not effected in accordance with clause 1.4. Mr Linden submits that whilst it was still open to the judge to find that there was a variation effected outside the formal terms of the contract, nonetheless there would need to be cogent and compelling evidence before it should be inferred that this is what the parties intended. The judge failed to give any real weight to that factor.

Third, Mr Linden submitted that when considering whether, objectively viewed, the parties intended to create legal relations, it was wrong for the judge to have regard to matters which were known to the employer but not to the workforce. So the judge ought not to have had regard to the concerns of the FSA or the board of the Bank when considering this matter since these were unknown to the employees at the material time. Similarly, with respect to the fact that the Board had itself chosen to disseminate the information via the Town Hall meeting, Mr Linden submits that the test should simply be how a reasonable employee would have interpreted the announcement and matters unknown to them could not enter into the equation. In support of this submission he relies upon two authorities in particular. They are the case well known to all first year law students, Carlill v Carbolic Smoke Ball Company [1893] 1 QB 256 and the decision of the Court of Appeal in Bowerman v Association of British Travel Agents Ltd [1996] CLC 451. Each of those cases involved a public statement to consumers of products and the issue was whether they were intended to create legal relations; were they promises which, if accepted, would constitute a contract? In each case the relevant question was held to be how the recipient of the promise would understand it.

Finally, it is submitted that once the analysis is limited to relevant factors which can properly be taken into account, the only proper inference is that there was no intention to create legal relations.

I will consider these grounds in turn.

The judge held that he who asserts that there is no intention to create legal relations should bear the burden of establishing it. Mr Linden correctly pointed out that the two authorities on which the judge relied were both cases of bilateral contracts. He submitted that there is no reason to assume that the same principle applies where there is a unilateral statement rather than bilateral negotiations.

I do not see why that should make any real difference, at least in a case where the parties are already in a contractual relationship when the unilateral promise is made. Indeed I would start from the premise urged upon the court by the claimants that where a term is being introduced into a pre-existing contractual relationship, there will be a very strong presumption that it is intended to be legally binding. In Thomas Judge v Crown Leisure Limited[2005] IRLR 823Lady Justice Smithexpressed the view, in the context of an employment relationship, that if sufficiently certain words are used, then no issue of intention to create legal relations would arise. She said this (para 23):

“In my view, with respect, [the claimant’s counsel] has misunderstood the ET's decision. The Tribunal did not hold that there was no intention to create legal relations. Indeed, in my view, that question never arose. These two men were employer and employee; in effect, legal relations already existed between them. If words had been uttered that were capable of amounting to a contractual promise, it could not sensibly have been suggested that there was no intention to create legal relations.”

Even if that may be said to be putting the position too high – and I refrain from expressing a final view on that - it at the very least supports the proposition that in a case where a change is being introduced against the background of an existing contractual relationship, the onus will be on the party asserting that there is no intention to create legal relations to establish that fact.

The submission is that the principal issue being relied on by the Bank was that the fact that the clause 1.4 procedures were not used in order to effect the variation points strongly against inferring an intention to create legal relations. It is said that the judge simply failed to deal with the point. The short answer to this is that clause 1.4 was in fact used, for the reasons I have given. But quite apart from that, I think that on a fair reading of paragraph 146, reproduced in paragraph 67 above, the judge is reflecting this argument, albeit not as transparently as he might have done. It was the Bank’s case that any change had to be made in writing pursuant to clause 1.4 in order to be legally effective, and the judge is then focusing upon the potential significance of the fact that the Bank chose to implement the change in some other way.

The contention here is that when looking at how a unilateral promise of this kind is interpreted, the authorities show that it is necessary to focus simply on how the reasonable recipient of the promise would have understood the offer. Since the employees would not share all the facts known to the employer, those matters unknown to them could not be taken into account.

On the facts, I do not think that this submission materially helps the Bank, even if it is correct, because the substance, if not the detail, of the facts which Mr Linden says were not known to the employees were in fact known to them. The two particular matters which Mr Linden relied upon include the FSA involvement and the fact that the Board had approved of Dr Jentzsch announcing the bonus pool in the way he did. There was an abundance of evidence that the workforce were aware that the FSA was concerned about the stability of the business; we were referred to over a dozen witness statements from claimants where the point was expressly mentioned. In any event the key point was that the promise was being made to sustain stability and this would have been obvious to staff in circumstances where the Bank was suffering heavy losses and where there were rumours that it was proposing to sell the investment banking division; the FSA involvement merely reinforced the importance of securing a stable workforce. Moreover, although the workforce may not have known for a fact that Dr Jentzsch was making the announcement with the approval of the Board, they would certainly assume that as Chief Executive of DKIB he had the requisite Board authority both to promise the bonus and to communicate that promise in the way he did. They would also know that important announcements were frequently made by Town Hall meetings.

Mr Hochhauser contended that in any event the legal premise was wrong and that background factors, even if not known to the employees, could be taken into account as part of the wider context when considering objectively whether there was an intention to create legal relations. It would be wholly artificial, he submits, to leave them out of account and the judge was right not to do so. He relied upon Carmichael v National Power [1999] ICR 1226, a decision of the House of Lords, to support his position that all relevant documents could be considered provided they have a bearing on the question in issue. That case was, however, concerned with issues of construction rather than intention to create legal relations and is not directly in point, although Mr Hochhauser relies on it by way of analogy.

In my judgment, this evidence is admissible even on the assumption that it was not known to the employees when the offer was made. The purpose behind the rule is that if the recipient of the unilateral promise would in the light of all the circumstances known to him reasonably understand the promise to be intended to be legally binding, the other party should not be allowed to escape liability by relying on evidence unknown to the recipient to establish that there was no such intention. But a rule of this nature must not be allowed to work an injustice. So a party who in fact knows that the other party does not intend to create legal relations cannot seek to contend otherwise by asserting that the evidence, objectively analysed, supports his case. He knows the truth and should not be allowed to deny it: Pateman v Pay (1974) 232 E.G.467. (A similar rule applies to stop an offeree from seeking to snap up an offer on terms consistent with the objective evidence when in fact he knows that the offeror did not intend to agree those terms: Hartog v Colin & Shields [1939] 3 All ER 566.)

Similarly, in Lark v Outhwaite [1991] 2 Lloyd’s Rep. 132 the plaintiff was asserting an intention to create legal relations but there was evidence from his agent which unambiguously showed that subjectively he did not have any such intention. Hirst J held that whilst the test whether a promise was intended to have legal consequences was primarily objective, the court should not be obliged to ignore entirely evidence of subjective intention. In my view the justification for admitting objective evidence relevant to understanding the likely intention of the offeror, such as the objective evidence in this case about the involvement and concern of the FSA and the need to keep staff is, if anything, stronger. An offeror should not be allowed to assert that there was no intention to create legal relations and at the same time seek to take advantage of a rule designed to benefit the offeree in order to conceal from the courts evidence which is inconsistent with his assertion.

As to the submission that that the judge reached a conclusion which was not sustainable on the evidence, Mr Linden accepted that he would have to satisfy the court that this was not a conclusion which a reasonable judge could have reached on the evidence before him. That was a realistic submission and reflects the fact that this court can only interfere if the judge below reached a conclusion that was wrong. In a context like this where inferences are drawn from primary facts, the court can only interfere if the judge steps beyond the range of reasonable decisions, adopting the same approach as it would in reviewing the exercise of a discretion, as Lord Justice Clarke pointed out in Assicurazioni Generali SpA v Arab Insurance Group [2002] EWCA Civ 1642; [2003] 1 WLR 577, para 16.

In my judgment, the conclusion of the judge was manifestly justified, essentially for the reasons he gave. For reasons I have already given, in my view there was a very strong presumption that a promise of this nature would be intended to be legally enforceable, and there were a whole series of matters supporting that conclusion. Perhaps the critical one is that this was a promise made in the context of a pre-existing legal relationship. In my judgment, viewed objectively, the natural inference would be that any promises made to staff relating to the terms of their employment would take effect in the same way as other contractual terms. Other factors reinforcing that conclusion include the fact that the source of the promise was the Chief Executive Officer with the obvious implication that the creation of the bonus pool must have been approved by the highest levels in the Bank; that it was part of a vitally important strategy to retain staff and prevent the potential collapse of the investment banking division; that the nature of the promise assured staff that the fund was guaranteed come what may; and that it was related to pay, the most fundamental obligation under the employment contract. There was, in truth, overwhelming evidence justifying the conclusion that this promise was intended to be legally binding. There were some factors lending a modicum of support to the Bank’s submissions, but the judge fairly considered them and gave cogent reasons why they were not ultimately persuasive.

I should add that I have taken on board three further points addressed by Mr Linden which did not perhaps figure as significantly before the judge. First, he emphasised that his submission did not mean that the promise, even if non-contractual, would have no legal significance. Any failure to respect the promise could be challenged under the term not to undermine trust and confidence, which was indeed precisely what the claimants have done as an alternative to their main case. It was therefore wrong of the judge to say that “if [the announcement] was not intended to have legal effect, it is difficult to see how it could have the intended effect”.

Second, it was submitted that to make the obligation legally binding would mean that the court would ultimately have to become embroiled in fixing the bonuses of staff, something which Lord Justice Mummery had said in Keen v Commerzbank [2007] ICR 623 the court would not do. His Lordship observed in that case that it was not for the courts to act as wage fixing bodies for the banking industry.

Third, it was said that the staff gave nothing in exchange for the promise.

None of these points, in my view, carries any significant weight, and certainly not enough to cast doubt on the judge’s conclusion. As to the first, the fact that the promise may be subject to some legal controls even if not contractually enforceable does not justify a court starting from the premise that this tells against an inference that there was an intention to create a binding contractual term. An objective analysis of intention would not start from the assumption that because there are general legal controls over the exercise of discretions by the employer, this is a factor pointing away from an intention to create contractual relations. Moreover, the judge’s observations to the effect that the purpose of introducing the promise would not have been achieved without the promise being legally binding were justified. It is fanciful to believe that the stability of staff would have been secured anything like as effectively had staff been told that the promise made to them was not contractually binding but that it might be indirectly secured by virtue of a general implied duty of trust and confidence.

As to the Keen point, I do not envisage that the court would have to fix the bonuses. It would simply require that when the Bank did so it distributed the full amount of the bonus pool. No doubt in the unlikely event that the Bank refused to do so the court may have to interfere faute de mieux, but as Maurice Kay LJ pointed out in the course of argument, this is something which courts exceptionally have to do from time to time when dealing with discretionary trusts. In any event, the Bank could not set up its own unwillingness to co-operate as a reason for inferring that there was no intention to create legal relations.

Finally, as regards the lack of anything in exchange, that is wrong as a matter of fact. The judge found in terms that each of these claimants had given good consideration in that the bonus pool was at least a factor, to a greater or lesser extent, in their decision to remain with the Bank. That was so notwithstanding that there may have been some reduction in job opportunities after the financial crash in September.

It follows that in my view there was an intention to enter into legal relations when this offer was made.

Mr Linden submits that even if there was an intention to create legal relations, this merely establishes that there was an offer to staff to set up the guaranteed bonus. Well established contract theory requires acceptance, however, and there was none here.

In my view, there are at least two related answers to this submission. First, as Lord Justice Bowen observed in the Carbolic Smoke Ball case (supra, p.269), “.. as notification of acceptance is required for the benefit of the person who makes the offer, the person who makes the offer may dispense with notice to himself.” In that case the mode of acceptance was held by implication to be actually performing the conditions attached to the offer. Here, in my view, it is plain that the employer has dispensed with the need for any response to the offer at all. This was a promise without any disadvantage, actual or potential, of any kind to the employees. Nobody hearing the promise made in this announcement would for one moment expect the employee to be able to benefit from it only if he or she positively accepted the offer. It would be a wholly formal and unnecessary exercise; the only sensible implication is that all employees who might potentially benefit from the promise would be deemed to have accepted it.

Second, the nature of the promise is inconsistent with the notion of individual acceptance. It would mean that if, say, a minority of the relevant staff accepted the offer, the employers would be bound to pay the whole of the bonus pool, if only to the minority. It is wholly unrealistic to think that the Bank would be willing to pay its bonuses from the fund by reference to whether the particular employee had formally accepted the offer or not. Indeed, if acceptance is needed, the minority who accepted the offer would at least arguably have a contractual claim against the Bank if it sought to diminish the fund by applying it for the benefit of those staff who had not accepted the offer and therefore had no claim on the fund. That too would be a bizarre result.

Accordingly, I see no merit in this argument. It follows that even if there was no unilateral change pursuant to clause 1.4 of the contract, there was in any event a binding contractual obligation to pay bonus payments in the usual way at least to the limits of the guaranteed fund.

The issue here is whether the introduction of the MAC clause amounted to a breach of the duty of trust and confidence. In dealing with the relevant law, the judge cited passages from the judgments of Lords Nicholls and Steyn in the leading case of Malik v Bank of Commerce and Credit International SA [1998] AC 20 and summarised the relevant principles as follows:

“…in considering whether the introduction of the MAC clause amounted to a breach of the duty of trust and confidence, it is necessary to consider whether, on an objective view,

(a)

the introduction of the clause was calculated or likely to destroy or seriously damage the relationship of mutual trust and confidence between the claimants and DKL, and if so

(b)

whether it was introduced without a reasonable and proper cause.”

It is not suggested that there is any error in this approach.

The second issue involves asking two questions: first, what was the reason the employer acted as he did? Second, did that reason constitute reasonable and proper cause?

Mr Linden properly reminded us that when analysing this part of the appeal, we must keep in mind two important points. The first is that it must be assumed that there is no contractual obligation to retain the minimum bonus pool; the clause was therefore introduced in a context where the maintenance of the bonus pool was discretionary. The second is that the issue is whether the introduction of the MAC clause was itself a breach of the trust and confidence term. Whilst with hindsight it is clear that the clause was implemented in a manner which dramatically reduced the bonuses otherwise payable, that was not clear at the time. The focus with respect to this argument must beon the introduction and not the implementation of the clause. He submitted that the judge lost sight of this fundamental point.

There is one further matter which I address at this stage. In argument I suggested that when considering the effect of the introduction of the clause on the trust and confidence of the employees, the focus should be not only on the clause itself but also on the assurance given at the time by Dr Jentzsch to the effect that the clause was unlikely to be exercised at all, and certainly in no significantly deleterious manner. Some employees said that they were reassured by his comments, as one would expect.

I would have thought that there is a powerful case for saying that the impact on trust and confidence can only properly be assessed by looking at the context in which the clause was introduced. However, if this submission was advanced below, it seems to have been very much in a low key and that I think explains why it was not directly addressed by the judge.

Moreover, it would seem likely that had he actively advanced the case in that way, contending that, objectively viewed, the employees would not at the time have envisaged any significant reduction in the bonus following Dr Jentzsch’s contemporaneous comments, it would simply have shifted the focus of the argument to the triggering of the clause. The bonuses paid were so out of line with the reassurance given by Dr Jentzsch that even if the introduction of the clause had not, objectively viewed, involved a breach of the duty of trust and confidence, there would be a powerful case for saying that the implementation of the clause would have done so.

Accordingly, I will examine the issue, as the judge did, by focusing almost exclusively on the terms of the MAC clause itself, not when read together with the soothing and reassuring balm of Dr Jentzsch’s observations.

The judge set out the context against which the MAC clause had been introduced: a promise made with the intention of retaining and incentivising staff, which the staff were intended to rely upon when making decisions as to their future, and which involved payments to be made from the guaranteed pool “no matter what”.

The judge then analysed the reasons for introducing the MAC clause. He noted that there were no members of the Board of DBAG or of Allianz who had given evidence about this. The only member of the DBAG management board who did so was Dr Merkel, who was general counsel for DBAG and attended meetings as a minute taker rather than as an active participant. The only other evidence dealing with the justification for introducing the clause was given by Mr Blessing, CEO of Commerzbank, but who did not become Chairman of DBAG until January 2009, after the clause was introduced

The claimants contended that the pressure to introduce the clause emanated from Mr Blessing and the Board of Commerzbank. This was in substance what the judge found.

The judge traced the events relating to the involvement of Mr Blessing and Commerzbank in the introduction of the MAC clause. I will not rehearse the evidence in any detail. It demonstrates that even before the acquisition of the Bank by Commerzbank, senior management in the Bank were concerned that Commerzbank might renege on the bonus pool promise. These concerns were well founded as subsequent events confirm.

On 3 November 2008 Mr Blessing had written to Mr Diekmann, CEO of Allianz, indicating that Commerzbank wished to be involved in the process of fixing and distributing bonuses and suggested that before they were finalised representatives of the joint compensation Committee should be convened from both Commerzbank and DBAG to decide on distribution.

Mr Blessing was equivocal about the purpose of this letter when cross questioned about it, and the judge interpreted it as the beginning of a campaign to undo the bonus promise which had been made.

Mr Blessing then sent an email to Dr Walter, CEO of DBAG, on 8 November 2008 where he stated that in his view the amount in the pool was completely excessive and that he was surprised by it in view of the high losses. He added that nobody would be able to understand why higher bonuses were being paid in DBAG than Commerzbank.

On 10November he, together with Eric Strutz, CEO of Commerzbank, wrote a letter to Mr Diekmann in which they said they were irritated and also a little upset at the way bonuses were being dealt with. They noted that “the variable compensation within our profession is being discussed intensely and controversially in public”. That letter was directed more towards bonuses for Back Office staff, but it referred to an article in the Frankfurter Allgemeiner Zeitung in which the reference had been made to the €400 million pool as “a taste of things to come”. When Mr Blessing addressed staff of DKIB at the Town Hall meeting on 12November however, he did not make clear that he was calling the size of the guaranteed bonus pot into question.

On 17November there was a meeting of a DBAG Steering Committee dealing with acquisition issues and Mr Diekmann reported that Mr Blessing agreed with the approach taken by DBAG in establishing the bonus pool, but notes of the meeting further record that “…. the promised DKIB bonus pool would be submitted to another inspection regarding its appropriateness and design”.

On 19 November Mr Zinssius, a Human Resources manager, sent an email to various members of the DKIB and Allianz Boards, but excluding Dr Jentzsch, in which he questioned “the appropriateness and structure of the DKIB bonus pool” and asked for information as to precisely how it had been set up.

In my view, it is highly relevant that Mr Blessing accepted in his evidence that by this stage “enquiries are being made left, right and centre to find any excuse to try to avoid having to pay out the whole of the bonus pool”.

As indicated above, the actual decision to introduce the MAC clause was then taken at a DBAG management meeting on 4 December. It considered a number of options before agreeing to the creation of the MAC clause. The minutes record that this was the recommendation of the Allianz Board. The judge further inferred that:

“In the absence of any evidence or explanation from the Allianz Board members (or any disclosure of the Allianz documents) it is a legitimate inference that what drove Allianz to push for the introduction of a MAC clause was pressure from Commerzbank.”

The judge then stated his conclusions about what lay behind this pressure:

“It is also clear the pressure successfully brought to bear by Mr Blessing was borne of an understandable sensitivity to the public perception of the payment of bonuses on such a scale in the context of the massive support for Commerzbank by the German Government. There was no financial motive in that the price Commerzbank had agreed to pay for the Bank reflected the accrual for the guaranteed minimum bonus pool so that the payment out of the whole of the guaranteed bonus pool would not have had an adverse financial effect for Commerzbank.

It follows that, in my judgment, the introduction of the MAC clause by DBAG was driven by Commerzbank for reasons unrelated to the problems of DKIB.”

The judge had earlier in his judgment observed that Mr Blessing had stated that the Bank could not stand by its promise because of a radical change in circumstances. These included the financial collapse; the takeover by Commerzbank; the fact that Commerzbank had to accept Government bailout; and the fact that the financial situation of the Bank in November and December was eroding further. The judge noted that it was difficult to reconcile these reasons with the fact that the Bank was emphasising that the bonus pool would be available even after the financial collapse and the proposed takeover by Commerzbank, and indeed even as late as 5 December, by which time it was already known that the financial situation in the Bank was eroding.

The judge then turned to whether the introduction of the clause was regarded as likely seriously to damage trust and confidence. He referred to various observations of management to that effect. Mr Hindle had expressed the view that there was a moral argument against the reduction of the bonus pool and Dr Jentzsch had regularly stated that he thought it would be wrong to reduce it. More specifically, on 28^th November he commented that:

“A retraction of the commitments will destroy the trust of the employees and the executive management fundamentally, finally and irrevocably … I consider it neither comprehensible nor responsible at this time to step back from the commitments made.”

He maintained that position thereafter.

The judge noted that the claimants too had expressed surprise and disappointment when they received the letters, although the comments of Dr Jentzsch at the Town Hall meeting gave them considerable reassurance. The judge added: “But for such assurances, the reaction to the introduction of the MAC clause could have been devastating” and he noted that this was acknowledged by Mr Aiken, a member of management who had been called as a witness for the Bank.

Finally, the judge had regard to an observation by Dr Jentzsch at the Town Hall meeting on the 19 December when, in response to a question, he said that in the current climate the use of the MAC clause was “a prudent thing to do” and should be considered in that spirit. The judge did not put any weight on that observation in context. In my judgment he was entitled to do that; this was an attempt to calm a potentially highly charged workforce. It was plain that Dr Jentzsch’s true opinion was that this clause was not appropriate.

In the light of all the material, the judge was satisfied that the introduction of the clause constituted a breach of the implied term of trust and confidence:

“it is clear that the clause was introduced simply as a means of enabling DBAG to go back on the promise made by Dr Jentzsch, the purpose for which it was eventually used, rather than for the much more limited purpose for which on its proper construction it could be used…”

In short, the judge thought that Commerzbank did not believe that the promise should have been made and the purpose of the MAC clause was to enable DBAG, at the behest of Commerzbank, to row back from it. The judge was not persuaded that it was genuinely a response to changing adverse circumstances because this was inconsistent with the fact that the stability of the bonus pool had frequently been confirmed notwithstanding those changes.

Mr Linden advances three principal grounds of appeal with respect to this aspect of the judge’s findings. First, he submits that the judge failed to focus on why DBAG had introduced the clause. Instead, he concentrated on the reasons why Commerzbank wanted to adopt the clause and ignored material evidence from both Dr Merkel and Dr Jentzsch which strongly supported the contention that DBAG was genuinely concerned about the deteriorating financial situation. Second - and this is a related argument – had the judge properly focused on the Bank’s reason for implementing the clause, he could only have concluded that it was a proper and reasonable one.

Moreover, to the extent that there was concern about the hostile public perception of the bonuses, it was artificial to treat this as distinct from the financial performance; the two were inextricably interlinked. The perception was that it was improper to pay such large bonuses precisely because the Bank’s losses were so huge.

Finally, when the judge considered whether trust and confidence had actually been undermined, he put too much weight on the subjective evidence of the witnesses and failed properly to assess the question objectively.

The evidence on which Mr Linden relies in relation to the first ground, and which he says the judge did not adequately consider, is this. Dr Merkel gave evidence that although he was not privy to any discussions between the Bank and Allianz or Commerzbank, nevertheless he could say from his own knowledge that the Board of DGAB was concerned about the deteriorating financial situation of DKIB. Moreover, he had advised the Board, as their legal counsel, that it was necessary to reconsider the bonus issue because it was possible that the payment of the bonus pool would place the Bank or its directors in breach of German law if, for example, it was not in the best interests of the Bank. The clause was specifically drafted in order to allow bonuses to relate to the actual financial performance which was still uncertain at the time the clause was introduced. Dr Jentzsch had similarly accepted that there was a real concern about the November and December figures. He had told the staff that in the circumstances it was prudent to adopt the clause. These were the only two witnesses who actually attended board meetings of the Bank when the decision was made.

I do not accept that this conclusion was not open to the judge. He explained why he took the view that the guiding hand was Commerzbank. There was an abundance of evidence which supported that inference, and the judge was entitled to have regard to the fact that no-one from Allianz or DBAG had come to explain why the decision had been taken. This was a striking absence of evidence given that the onus of establishing a reasonable and proper cause lay on the employer, once the objective breach of trust and confidence had been established. Once the judge had concluded that Commerzbank’s concerns were the real reason why the clause was introduced, he was entitled to find that the ostensible reason expressed in the minutes did not reflect the true underlying reason. Moreover, the fact that the promise of the bonus pool was reiterated so frequently even after the dramatic financial collapse and even when it was already clear that the November results in particular were seriously poor suggests that these were not the principal reasons why DBAG thought it necessary to introduce the clause. It is also pertinent to note that the Bank was never pleading impecuniosity; they were not saying that they could not afford the payment. Indeed, the existence of a €400 pool had been factored into the price paid by Commerzbank for DBAG.

In particular, in my view the judge was entitled not to place too much emphasis on Dr Merkel’s evidence. First, he was not a director of DBAG and only attended the meetings to take minutes and give legal advice. Second, whilst I do not doubt that there was some concern about the legality of the bonus pool under German law, this is not mentioned at all in the minutes, a striking omission if this were an important concern.

In any event, Dr Merkel says only that these concerns explain why the question of the bonus pool was taken back to the Board for further consideration; it was not suggested that they dictated the decision to introduce the MAC clause or indeed that they even played any role in that context. Finally, and importantly, Dr Merkel accepted in evidence that he was not privy to discussions between the Bank and Allianz or Commerzbank. So on the central question whether the Board were acting because of pressure elsewhere, he was simply in no position to say. Likewise, Dr Jentzsch was not a party to the discussions with Allianz or Commerzbank. Mr Blessing was the only witness who could speak of those relations, and the judge relied on his evidence.

In view of all this evidence, the judge could properly conclude that the reason for the pressure was the particular and understandable sensitivity of Commerzbank to the public perception that such bonuses should not be paid in view of the massive government support for Commerzbank. Mr Linden suggests that even so, that was capable of constituting a reasonable and proper cause. There are two problems with that submission. First, the susceptibilities of Commerzbank would not justify DBAG reneging on its promises to its staff. Second, the case was never advanced before Mr Justice Owen on that basis.

Finally, Mr Linden submits that the judge failed to analyse objectively the question whether trust and confidence had been undermined. I can deal with this issue briefly. First, on any fair reading of the decision, he did. It is true that he did not state in terms in his conclusion that trust and confidence had been objectively undermined, but when he set out the principles of law he said that this was an objective test. There is simply no basis at all for asserting that he failed to give effect to his own self-direction.

Moreover, in my judgment, there was ample evidence justifying the conclusion that the introduction of the clause undermined trust and confidence. Mr Tozzi rightly placed considerable emphasis on the importance of the original promise. It had been reiterated time after time from its inception not only by Dr Jentzsch but more generally at management meetings and in the FAQs. This continued after the collapse of Lehman Brothers and the onset of the financial crisis and after the loans to Commerzbank by the German Government. Its significance to the relevant staff was obvious; the judge found as a fact that in respect of all of them it has played some part, to a greater or lesser extent, in their choosing to stay with the Bank. Even if it was not legally binding, it was a promise to pay the bonus from a guaranteed pool, “no matter what”, made in good faith by the Bank and intended to be honoured.

It is not difficult to imagine how in those circumstances the introduction of a clause which involves a volte face by the Bank and threatens the guaranteed fund would cause considerable alarm. Furthermore, management were acutely aware of the considerable potential damage to trust and confidence, as the judgment made clear. Both Dr Jentzsch and Mr Hindle considered it to be an employee relations disaster and morally unjustified. Perhaps the most powerful evidence was from Mr Aiken, who gave evidence on behalf of the Bank, and accepted that but for the assurance given by Dr Jentzsch at the time that the clause would probably not be implemented, the Bank “would have had a riot on its hands.” This is also consistent with the view that whilst these calming words might have headed off a riot, there was still significant damage to trust and confidence at the prospect of anticipated bonus payments, an important element in the pay of any investment banker, being removed.

Mr Linden submitted that whilst the staff would be disappointed by the threatened reduction of their expected bonus, they would recognise that there were understandable reasons for this. Moreover, many of them would not have been able to obtain alternative employment in any event, given the lack of jobs so they would not in practice be disadvantaged in any way by a failure to honour the promise. The latter observation is not borne out by the evidence which showed that there were still opportunities for good staff elsewhere. As to the remaining submissions, they amount in essence to rearguing the case pursued and lost before the judge. They do not identify any error of law.

Mr Linden contended that the judge had wrongly relied upon evidence post-dating the introduction of the clause which was inadmissible in this context. In fact he only did so with respect to two observations of Mr Blessing. One was reported in the minutes of the meeting of the supervisory board of DBAG in March 2009 when he pointed out that internally the board had frustrated expectations and thereby eroded trust. The other was a statement to similar effect made in February. It is true that these comments were made when the full extent of the reduction had been established, but the evidence was nonetheless relevant to the issue whether the introduction of the MAC clause, which was always likely to lead to some reduction in bonus, would undermine trust and confidence. In any event, it simply reinforces other powerful evidence to the same effect.

Mr Linden also submitted that the judge was wrong to place emphasis on the subjective views of the parties as to whether trust and confidence had been undermined. It seems to me that this is an unreal complaint. Of course, it is possible that they may be unduly sensitive and treat something as seriously damaging trust when objectively it does not do so. But the reaction of a large number of employees, genuinely expressing their shocked and disappointed reaction to the introduction of the MAC clause, whether or not expressed formally in terms of trust and confidence, is surely some evidence of how a reasonable employee would have reacted to it.

For these reasons, I would reject this aspect of the appeal also.

The third issue raised a number of difficult questions of construction of the MAC clause. The context is that the claimants submit that if they have no contractual claim, and if the MAC clause was properly introduced without any breach of the implied term of trust and confidence, nonetheless the Bank cannot rely on it because they did not properly comply with its terms. The issue is relevant only to this particular clause, now of historical interest only. Since I have found for the claimants on the first two grounds, it is unnecessary to delve further into the interesting submissions advanced by counsel on this point and would only add unduly to an already long judgment. Accordingly, I decline to engage with this issue.

In my judgment, the appeal fails on both the grounds I have analysed in this judgment. There was an effective unilateral contractual amendment pursuant to clause 1.4 of the contract. But even if I am wrong about that, there was in any event a binding contractual promise resulting from the terms of the promise and the circumstances in which it was made.

In the circumstances it is strictly irrelevant whether the introduction of the MAC clause constituted a breach of the duty of trust and confidence or not. However, in my view it did. I would pay tribute to the careful and detailed analysis of the judge below.

I agree.

I also agree.