England and Wales High Court (Commercial Court) Decisions

IN THE HIGH COURT OF JUSTICE

		Royal Courts of Justice Strand, London, WC2A 2LL
		06/08/2014

Section	Para
A. INTRODUCTION	1
The bonus clause	2
The dispute	4
The arguments in brief	6
Witnesses	9
B. FACTUAL BACKGROUND	12
The claimants join Investec	13
The Structured Equity Derivatives business	15
Investec retail structured products	17
EVA	24
Bonuses for 2008/2009	26
Bonuses for 2009/2010	28
Bonuses for 2010/2011	31
C. THE ISSUES	33
Accord and satisfaction	33
The rate issue	35
Other issues	38
D. WAS THERE AN ORAL AGREEMENT?
The alleged oral agreement	39
The claimants' evidence	42
The Bank's evidence	47
The claimants' credibility	49
No documentary evidence	52
Inconsistent conduct	54
Possible discussions	60
The claimants' own evidence does not support their case	66
Conclusion	68
Other matters allegedly agreed	69
E. INTERPRETATION OF THE BONUS CLAUSE	73
The applicable legal principles.	74
The meaning of EVA	75
Good faith and rationality	91
Discretion and its limits	95
The allegations of bad faith and irrationality	102
F. THE RATE ISSUE	104
The claimants' case	105
Discussion	107
Reasonable expectations	115
G. OTHER ISSUES	118
Reserving for kick out products	119
Adjustments to the Funding Gap Reserve	123
The 'profit payaways' issue	133
The 'early bird' issue	142
H. CONCLUSION	149

Mr Justice Leggatt :

A. INTRODUCTION

1. On 10 April 2007 Mr Andrew Brogden and Mr Robert Reid, the claimants in this action, entered into contracts of employment with the defendant, Investec Bank (UK) Ltd ("Investec" or "the Bank"). They joined Investec as, respectively, the Head and Deputy Head of Equity Derivatives. The claimants' contracts of employment provided in each case for a basic salary of £120,000 per annum but the main part of their remuneration was expected to consist of bonus. An important consideration for the claimants in deciding to join Investec was that their bonuses were not discretionary but a matter of contractual right.

The bonus clause

2. Mr Brogden's contract provided for a guaranteed bonus of £6,200,000 payable in his first year of employment. This amount was payable on the condition that Mr Brogden's employment with the Bank was not terminated by reason of his resignation or gross misconduct at any time before 30 June 2011 – in which event the full amount would be repayable. The bonus clause in the contract further provided as follows (with underlining added):

"In your second financial year of operation starting on 1 April 2008, the bonus calculation will be based on an EVA formula calculated as 40% of the EVA generated by the Equity Derivative business. The bonus pool available for distribution to members of the team will be calculated after deduction of NI.

In the third financial year starting on 1 April 2009 and thereafter the bonus calculation will be normalised based on a formula calculated as 30% of EVA generated by the Equity Derivative business. The bonus pool available for distribution to the members of the team will be calculated after deduction of NI, an amount for Head Office and an amount for TSF support.

The proportion of the available bonus pool to be distributed to each member of the team in any year will be determined by you following consultation with David Van Der Walt.

The bonus pool will be paid to members of the team on or before 23 June in each year and no part of the bonus pool in each financial year will be deferred."

3. The bonus clause in Mr Reid's contract of employment was identical to the clause in Mr Brogden's contract, save that Mr Reid's guaranteed bonus payable in his first year of employment was £3,800,000.

The dispute

4. Any hope that agreeing a contractual formula for bonus would result in certainty has not been fulfilled. In the second and each of the two subsequent years of the claimants' employment there was disagreement about the amount of their bonus entitlement. For the financial year 2008/2009, and again for the financial year 2009/2010, the disagreement was amicably resolved and Investec agreed to pay bonus amounts which the claimants accepted. However, at the end of the financial year 2010/2011, no such agreement was reached. According to the Bank's calculation, the performance of the structured equity derivatives business was such that the claimants were not entitled to any bonus for that year. The claimants on the other hand contended that they were entitled to a bonus pool of over £8 million. In June 2011 the Bank made discretionary bonus payments of £150,000 to Mr Brogden and £100,000 to Mr Reid, while maintaining that no bonus was contractually due. The claimants had reached the time when under the terms of their bonus clause they were free to leave without penalty, and they did so. By a letter dated 20 July 2011, Mr Brogden resigned from his employment with Investec. Mr Reid followed him on 3 August 2011.
5. In this action begun on 2 September 2012 the claimants seek damages from Investec for breach of contract in failing to pay bonuses which they say should have been paid to them for the financial year 2010/2011. According to the particulars of claim, the sums which Mr Brogden and Mr Reid should have been paid are, respectively, £3.6 million and £2.7 million.

The arguments in brief

6. I have underlined the key contractual language in the bonus clause quoted in paragraph 2 above. It can be seen that the description of the bonus formula is on any view elliptical. It is the claimants' case that, in addition to their written contracts of employment, there was an oral agreement made before they joined the Bank that certain ground rules would be applied in calculating the amount of their bonus pool. Further and alternatively, the claimants contend that these ground rules can be identified by interpreting the bonus clause in their contracts in accordance with the ordinary and well established legal principles applicable to the interpretation of contracts in English law. They rely in particular on the point that "EVA" is agreed to be an acronym for "Economic Value Added" and argue that the contractual language, properly interpreted, requires the parties – and, if they cannot agree, the court – to work out the true added economic value generated by the equity derivative business. The claimants also argue that, if and to the extent that the Bank had any discretion in relation to the calculation of EVA, the Bank had a duty to exercise that discretion rationally and was in breach of that duty.
7. Investec adamantly denies that any oral agreement was made as alleged by the claimants. Its case, in brief, is that the Bank has an established method for calculating bonuses which was explained to the claimants before they joined and which is part of the background against which their contracts of employment are to be construed. The Bank contends that, construed against this background, the words "EVA generated by the Equity Derivative business" mean the amount calculated as such "EVA" by the Bank. For 2010/2011 no bonuses were payable because Investec calculated that such EVA was nil.
8. The litigation has been conducted acrimoniously and each side has levelled accusations of dishonesty or bad faith against the other. The claimants' pleaded case includes allegations that the Bank acted in bad faith and deliberately manipulated or adjusted the inputs to the EVA calculation in order to reduce the claimants' bonuses. For its part, the Bank in its written opening submissions for the trial described the oral agreement alleged by the claimants as a dishonest fiction and accused them of making a "contrived and dishonest claim".

Witnesses

9. At the trial Mr Brogden and Mr Reid gave evidence on their own behalf. The Bank called three witnesses of fact: Mr Van Der Walt, who was responsible for recruiting the claimants and at the time was Head of the Bank's Capital Markets Division (he is now CEO of Investec); Mr John Barbour, who at the time was the Head of Central Treasury; and Mr Kevin McKenna, who was the Chief Operating Officer of the Capital Markets Division and is now the Bank's Chief Operating Officer.
10. Each side also adduced expert evidence from an expert in accountancy. The claimants' expert was Mr Steven Brice and the Bank's expert was Mr Hugo Watson-Brown.
11. The issues relating to the bonus calculation have been referred to cryptically as the "rate" issue, the "kick out" issue and the "early bird" issue. To render these issues intelligible I must first describe more of the relevant factual background.

B. FACTUAL BACKGROUND

12. Before they joined Investec, Mr Brogden and Mr Reid were employed by Abbey National Plc and then, following its takeover, by Santander. In 2006 they were running a successful equity derivatives business at Santander.

The claimants join Investec

13. In the later part of 2006 Investec decided to recruit a new equity derivatives team. A previous attempt to set up such a business within the Bank had not been successful. With the assistance of head hunters, Investec approached a number of equity derivatives teams working for other investment banks. Mr Van Der Walt was particularly interested in the team run by Mr Brogden. After an initial meeting on 30 October 2006 they began serious discussions. These discussions continued into the early part of 2007 and involved several further meetings. Most of the meetings and discussions were between Mr Van Der Walt and Mr Brogden but Mr Reid was also present at two meetings in February 2007. The negotiations culminated in the claimants agreeing to leave Santander and to join the Bank. I will consider later the claimants' evidence about what they allege was orally agreed in the course of the negotiations.
14. As already mentioned, contracts of employment were signed on 10 April 2007. The claimants began working at the Bank in July 2007 and set about building an equity derivatives business from scratch.

The Structured Equity Derivatives business

15. The various business teams within Investec's Capital Markets Division were physically arranged in "desks" and the business operated by the claimants was known as the Structured Equity Derivatives (SED) desk. Each desk was operated and accounted for as a separate or stand alone business unit within the Bank.
16. Two main types of business were conducted by the SED desk. One type of business was trading financial instruments. The other type involved structuring and selling custom made equity derivative products. A large part of Mr Brogden's work, and most of Mr Reid's work, consisted of structuring (i.e. designing and pricing) such products. As Head of the desk, Mr Brogden was also responsible for managing what became a team of approximately 20 traders, structurers and sales people, with the assistance of Mr Reid. The claimants were also responsible for designing and implementing all the necessary systems, controls and processes for the SED business.

Investec retail structured products

17. In late 2007 the claimants saw an opportunity for the Bank to enter the UK retail market for equity-linked structured products by designing such products and selling them to members of the public through Independent Financial Advisors. Mr Brogden produced a business plan in January 2008 and obtained the necessary internal approvals. The first Investec retail structured product was launched in May 2008.
18. This Investec structured products business was extremely successful and grew rapidly. As it turned out, the success of this business came at an opportune time for the Bank. In the financial crisis which followed the collapse of Lehman Brothers in October 2008 the willingness of banks to lend to each other was severely impaired and the cost of borrowing in such 'institutional', or 'wholesale', markets rose steeply. The Bank looked for other sources of funds. The sale of Investec's retail structured products became a very valuable and important source of term funds for the Bank.
19. The basic concept of these products was that the customer would invest a sum of money for a term of three or five years. At maturity the customer was guaranteed to receive at least the return of their initial deposit. The customer would receive a larger payout if the FTSE 100 index rose over the relevant period.
20. Many of the products incorporated a feature referred to as "kick out". Where this feature was present, the customer's investment would be repaid early with an uplift, if on an anniversary date the FTSE 100 index was higher than on the date of the original investment.
21. Another feature of some products involved paying investors who sent in their application forms and deposits more than, say, six weeks before the start date of the investment a favourable rate of interest on their deposits during this period. This feature was known as "early bird".
22. The cost to the Bank of providing a retail structured product had two main elements. One element was interest paid on the customer's deposit. The other was the cost of buying an equity derivative which was built into the product. When structuring a new product (or a new issue of an existing product) the desk would need to determine the rate of interest that would be payable to the customer on the funds invested. In principle, the higher the rate, the more attractive the product would be. In order to fix this rate, the SED desk would need to know the interest rate which the Bank's Central Treasury was prepared to pay for the use of the funds. Interest at this rate was credited internally within the Bank by the Central Treasury to the SED desk. In order to make a profit for itself, the SED desk needed to price the product at a rate which was sufficiently below the rate credited by the Central Treasury after the costs of issuing the product were taken into account.
23. To begin with, the rate of interest to be paid internally by Central Treasury to the SED desk on the funds raised by each new product launched was discussed and agreed between Mr Brogden and Mr Van Der Walt. However, in around July 2009 the Bank introduced a new committee structure. Thereafter, the rate which the Bank was prepared to pay for funds was determined by a committee known as the Pricing and Profit Forum. Mr Reid was a member of this forum on behalf of the SED desk.

EVA

24. "EVA", short for "Economic Value Added", is a measure used within the Capital Markets Division of Investec to assess the performance of business units and their contribution to overall profitability. It is also used as a basis for awarding bonuses, although it was the evidence of Mr Van Der Walt that none of the other 14 business units within the Capital Markets Division had a contractual right to be paid a bonus calculated in accordance with their applicable EVA formula. More specifically, EVA is a measure of return on capital deployed. The basic formula used by Investec to calculate the EVA for each business unit is: EVA = revenue minus costs, minus the cost of the capital employed by the unit (after giving credit for the return on capital which the unit would have received if it had raised its own capital).
25. Under the Bank's internal accounting policies and procedures, calculations of EVA are generated automatically from inputs to the Bank's automated trading systems and adjusted at the financial year end. Reports showing the financial performance of each business unit, including figures for profit and loss and EVA, are generated on a daily and monthly basis. Mr Brogden was required to and did approve such reports on behalf of the SED desk.

Bonuses for 2008/2009

26. As mentioned earlier, in their first year of employment the claimants received guaranteed bonuses and so no bonus calculation was required. The first time that a bonus calculation had to be made was at the end of the financial year which started on 1 April 2008 and ended on 31 March 2009.
27. The initial EVA calculation for the SED business prepared by Mr McKenna after the 2008/2009 year end gave three alternative figures depending on whether a particular adjustment was made to P&L to establish a reserve for kick out products and whether the acquisition costs of the SED team should be brought into account. Mr Brogden's response in an email sent to Mr Van Der Walt on 14 May 2009 was to state that the figures were "some distance away from my view of what is right, fair and/or realistic". A revised calculation of EVA was provided by Mr Van Der Walt to Mr Brogden on 29 May 2009 which calculated a total bonus pool for the SED desk of £2.426m. Following further discussions between Mr Van Der Walt and Mr Brogden, the Bank ultimately agreed to pay Mr Brogden and Mr Reid bonuses of £1.45m and £1.4m respectively (with much smaller bonuses also paid to members of their team). The amounts paid were therefore more than the Bank considered to be due under the bonus formula contained in the claimants' contracts of employment, though still less than Mr Brogden had contended for.

Bonuses for 2009/2010

28. For the year 2009/2010, Mr Brogden was paid a bonus of £3m and Mr Reid was paid a bonus of £2.25m. There was a considerable amount of discussion at the year end before these sums were agreed.
29. By this time Mr Van Der Walt had been succeeded as the Head of Capital Markets by Mr Andrew Clapham, who was not initially aware of the bonus formula contained in the claimants' contracts until Mr Brogden mentioned its existence to him. For this year the claimants' contracts required the bonus pool to be calculated as 30% of EVA generated by the SED business (reduced from 40% the previous year) and to take into account for the first time deductions for Head Office costs and TSF support. Mr Brogden produced his own proposal for the bonus pool for his desk which differed significantly from Mr McKenna's calculations.
30. Mr Van Der Walt became involved in the discussion. On 14 May 2010 he sent Mr Brogden an email which began:

"We seem to be missing each other a lot on the issue. Where I feel the bank is trying to be generous you seem to feel that we are not coming to the party. That does not mean that I am not equally concerned as I thought that there was a reasonable understanding of how we operate and what would be expected. That being said this does not preclude us from having a good discussion around the issue."

Mr Van Der Walt went on to set out in the email a calculation of EVA which resulted in a bonus pool available to the desk of £4.2m. He continued:

"Against this background it was felt by the group and myself that £6 million would be a reasonably decent gesture as extra recognition for some of what the desk has achieved and trying to marry expectations to the reality that faces us."

The offer to make available a bonus pool of £6m was accepted by the claimants and the distribution of this pool among the SED team was also then agreed with Mr Van Der Walt.

Bonuses for 2010/2011

31. On 14 April 2011 Mr Brogden sent to Mr Clapham his proposals for the bonus numbers that he would like to pay everyone in his team for the 2010/2011 year. The total amount proposed, which did not include his own bonus but included a bonus of £2.7m for Mr Reid, was around £5m. In his email Mr Brogden observed:

"Obviously the reported P&L is not too pretty this year, but we would make the case that for a variety of reasons (some good, some bad) this is not a fair reflection of the achievements of the business."

32. The response of Mr Clapham after discussion with Mr Van Der Walt was that, as the P&L and EVA figures for the SED desk for the year were negative, there was no money in the bonus pool. The Bank subsequently refused to budge from this position, despite a number of attempts by Mr Brogden to argue that the EVA had been inaccurately and unfairly calculated and that the claimants were entitled to a substantial bonus pool. As mentioned earlier, the disagreement led to the claimants resigning from their employment and bringing this action against the Bank.

C. THE ISSUES

Accord and satisfaction

33. One issue raised by the Bank in its defence can be disposed of shortly. The Bank sent letters to the claimants dated 2 June 2011, not reflecting any prior discussion, which informed them each that the Bank was "pleased to award you a discretionary bonus". The amount of the bonus was £150,000 in the case of Mr Brogden and £100,000 in the case of Mr Reid. These sums were subsequently paid into their bank accounts. The Bank contended that by accepting these sums without any reservation the claimants settled their claims for bonuses by accord and satisfaction.
34. This argument was plainly misconceived. The payments were not tendered in settlement of the claims of Mr Brogden and Mr Reid to be paid a contractual bonus. Indeed, the sums paid were said to be discretionary bonuses and therefore did not relate to any contractual entitlement. The defence of accord and satisfaction was bound to fail and was rightly, if belatedly, abandoned by the Bank in closing submissions.

The rate issue

35. By far the most important disputed issue in financial terms is the "rate issue". This issue is whether, in calculating EVA for the purposes of determining the claimants' contractual entitlement to bonuses, the rate of interest credited to the SED business on funds raised from sales of Investec structured products was to be:

i) The 'institutional market rate' of interest that the Bank would have been required to pay had it had to borrow in the wholesale lending market similar term funds to those deposited – which is the claimants' case; or

ii) The interest rates actually credited to the SED desk by the Bank's Central Treasury on the funds – as the Bank contends.

36. If the claimants are right on the rate issue, then they calculate that they were entitled to a bonus pool for the 2010/2011 year of at least £4.27m (and more if they succeed on other issues). On the other hand, if the claimants are wrong on the rate issue, they accept that the result of the EVA calculation is negative so that there was no bonus pool, even if they succeed on all the other issues.
37. The claimants' case on the rate issue is founded primarily on an oral agreement which they say was made in the course of their discussions with Mr Van Der Walt before they joined the Bank. Alternatively, they rely on what they contend is the proper construction of the bonus clause in their contracts. I will consider these issues first.

Other issues

38. The remaining issues, which I shall consider afterwards, relate to the treatment in the EVA calculation of (i) kick out products and (ii) early bird deposits.

D. WAS THERE AN ORAL AGREEMENT?

The alleged oral agreement

39. The claimants allege that in their discussions with Mr Van Der Walt before they joined the Bank it was orally agreed that the EVA for the SED desk would be calculated in accordance with the following ground rules:

i) Profits from trades would be recognised immediately rather than accruing over the life of the trade; the SED business would operate on a mark to market basis and therefore unrealised profits and losses would be taken into account;

ii) Fair accounting principles would be applied in line with standard practice in the equity derivatives market;

iii) In relation to structured products, the desk would be credited for funds raised at the prevailing institutional market rate; and

iv) There would be no cross-subsidy of or from other businesses.

40. Of cardinal importance to the claimants' case is the third of these points. It is their contention that it was expressly agreed with Mr Van Der Walt that, for bonus purposes, the value of funds raised by the SED business would be assessed by comparing the actual cost of funds raised with the cost that the Bank would have incurred by raising the same funds, for the same periods, in the institutional bond market.
41. There are two aspects to this issue. The first is whether the discussions alleged by the claimants took place. This is a pure question of fact. The second aspect is whether, if such discussions did take place, the parties intended to make a legally binding agreement. It is common ground that this question is to be judged objectively. In other words, it is not material whether the parties privately thought that they were making an agreement which would be legally binding; the test is whether their communications would reasonably have been understood as intended to create legally enforceable rights and obligations: see e.g. Chitty on Contracts (31^st Edn), vol 1, para 2-164.^[1]

The claimants' evidence

42. There is no contemporaneous record of the discussions which the claimants say took place. Nevertheless, they have given in their witness statements made for the purposes of these proceedings, and affirmed in their oral evidence, detailed accounts of the alleged discussions. According to the claimants' evidence, the discussions arose out of a concern they had about Investec's credit rating. At the time the rating was Baa2 (with Moody's) but Mr Brogden says that he was told by Mr Van Der Walt that the Bank was expecting its credit rating to improve in the near future to at least A3. The claimants' concern was that a poor credit rating would be a hindrance to doing business particularly with other financial institutions.
43. The claimants say that they discussed this question between themselves in early January 2007, when Mr Brogden first told Mr Reid about the potential opportunity to start up a new equity derivatives business at Investec. According to Mr Reid, following his initial conversation with Mr Brogden he looked into the credit rating issue and reported back a few days later with his provisional conclusion that the Bank's credit rating should not be a problem "providing we would receive the full market rate from Investec, so that we could offer competitively priced products".
44. Mr Brogden apparently had a meeting with Mr Van Der Walt at the Bank's offices on 23 January 2007. There is no note or written record of anything said at this meeting and the documents in the trial bundles contained no reference to it. Mr Brogden did not explain how he had identified the date or the fact of this meeting and I can only surmise that it was from a diary entry. According to Mr Brogden, he explained to Mr Van Der Walt on this occasion that the claimants could work with the Bank's credit rating provided that the Bank paid a fair market rate for funds raised and that the rating was going to improve to A3. Mr Van Der Walt reiterated his confidence that the credit rating would shortly improve and said that the Bank's Treasury would pay the "institutional market rate" for all funds generated by the SED desk. Mr Brogden says that from his previous experience he had thought it more likely that the Bank's Treasury would pay the market rate minus a small 'bid/offer' spread, rather than the full market rate, and he was pleasantly surprised by Mr Van Der Walt's statement that they would receive the full market rate. He says that Mr Van Der Walt explained the rationale for this to be that the Bank would much prefer the claimants to raise money at the same time as building a client franchise rather than the Treasury simply raising money on the open market and, further, that any profit made from the desk raising money below the market rate was a profit for the Bank which would not be made if the Treasury sourced the funds on the open market.
45. There was a further meeting on 19 February 2007 which Mr Reid also attended and at which he met Mr Van Der Walt for the first time. There is a brief reference to this meeting in an internal email sent on 21 February 2007 by Mr Van Der Walt, in which he mentioned only his favourable impression of Mr Reid and that he was now working out what offer he could make to the claimants regarding bonuses. However, according to the claimants there was at this meeting another discussion of the Bank's credit rating and Mr Van Der Walt was keen to hear Mr Reid's view on how the rating could affect customer business. Mr Reid says that he took this opportunity to mention a small structured product hedging transaction which he had executed at Santander with Investec Private Bank in which Investec had paid a rate in excess of LIBOR. Mr Reid says that he mentioned this to gauge the Bank's appetite for such business. The claimants both say that Mr Van Der Walt's response was that the Bank would encourage this type of business and would credit the claimants' desk with the full institutional market rate. In his oral evidence Mr Reid elaborated further and said that Mr Van Der Walt mentioned that the Bank was currently raising money, possibly two year money, in the institutional market and was paying a rate which Mr Reid recalls being about LIBOR + 3.5%.
46. It is submitted on behalf of the claimants that these discussions gave rise to a legally binding agreement between the claimants and the Bank, albeit conditional on the claimants accepting offers of employment, that in calculating their bonus pool the claimants would be credited with the 'institutional market rate' of interest on funds raised by their desk.

The Bank's evidence

47. Mr Van Der Walt did not claim to have any specific recollection of his discussions with the claimants, but denied that he would have made any agreement of the kind alleged. According to Mr Van Der Walt, the claimants' focus at the time when they joined the Bank was on making commissions by selling derivative products to other institutions and at that stage they had no plan to take deposits by selling structured products directly to retail customers. Mr Van Der Walt said that such a plan was first suggested only in January 2008, after the claimants had been at Investec for several months. In these circumstances there would have been no reason to discuss at his pre-contractual meetings with Mr Brogden and Mr Reid the rate of interest which the Bank's Treasury would pay on deposits.
48. Mr Van Der Walt also said that, until this dispute started, he had never used the expression "institutional market rate" nor heard it used in the industry. He had heard, and used himself, the term "institutional rate" to refer generally to prices in the bond market. However, this is not a fixed or defined concept such as LIBOR. He denied that in these circumstances he could conceivably have agreed that the Bank's Treasury would pay the "institutional market rate" on funds raised by the claimants' desk.

The claimants' credibility

49. The claimants' unsupported testimony as to what was said at meetings which took place some seven years ago is not a promising basis on which to found a claim that they made a contract with the Bank, separate from and additional to their contracts of employment, which controlled how the Bank was required to calculate their bonuses. The fact that the claimants have a vested interest in the existence of such an oral agreement since, if their evidence is accepted, they each stand personally to win a payout of several million pounds gives all the more reason to doubt the reliability of their evidence. Nonetheless, Mr Cavanagh QC on their behalf urged the Court to reject the attacks on their honesty made by the Bank and to find that both Mr Brogden and Mr Reid gave honest and truthful testimony.
50. I do not find the claimants to be dishonest. I accept that they genuinely believe that the discussions which they have described in detail took place exactly as they now recall them. However, I do not accept that their evidence is true. I regard their claim that an oral agreement was made to use the "institutional market rate" in calculating their bonuses as wholly incredible. This case seems to me to illustrate the truth of the observation of Lord Justice Browne that "the human capacity for honestly believing something which bears no relation to what really happened is unlimited": quoted in Tom Bingham, 'The Business of Judging' (OUP, 2000) at p.15, and by Mostyn J in A County Council v M and F [2011] EWHC 1804 (Fam) at para 30.
51. I have formed that view for three main reasons.

No documentary evidence

52. The first is that there is no documentary record of, nor documented reference to, any such discussion or agreement. It is not merely that there is a want of proof – though there is. This is a situation in which absence of evidence is itself evidence that no agreement was in fact made. The size of the bonuses to which they would be contractually entitled, and hence the way in which such bonuses would be calculated, was clearly a matter of great significance to the claimants. On their own evidence, it featured prominently in their discussions with Mr Van Der Walt and in their decision to leave secure jobs and join Investec. If in those discussions an agreement had been reached about how the claimants' bonuses were to be calculated which was intended to bind the Bank, I am sure that they would have thought it of sufficient importance to make some record of that fact. I would expect them, indeed, to have wanted the points agreed to be incorporated in their contracts of employment or, at the very least, in a side letter. I am equally sure that Mr Van Der Walt would have thought it necessary to make a record of any such agreement. He would also have needed to obtain internal approval for the obligations he was undertaking or proposing to undertake on behalf of the Bank.
53. Mr Van Der Walt did set out in emails and a memorandum sent to the Bank's two senior executives in March 2007 the terms as to remuneration and bonus that he was proposing to offer to the claimants. The very fact that he had not yet made an offer regarding remuneration makes it inherently unlikely that he had by then already made an agreement intended to have binding effect about how bonuses paid to the claimants would be calculated. In an internal email dated 20 March 2007 Mr Van Der Walt outlined the final proposal. In this email he described the proposed bonus arrangements for the first two years and stated:

"In year 3 and thereafter a normal EVA model would apply at a rate proposed of 30% which is how the other trading businesses are remunerated."

It is unlikely that Mr Van Der Walt would have written in these terms if he was agreeing with the claimants any special terms which would apply in calculating their bonuses and which did not apply to the other trading businesses.

Inconsistent conduct

54. The second reason why I regard the claimants' case as incredible is that throughout the period of four years during which they were employed by Investec, not once did they suggest that the agreement now alleged had been made or allude to its existence. Moreover, that is so despite the fact that, on their evidence, the Bank's Central Treasury stopped paying the institutional market rate on funds raised by the SED desk in about February 2009. No suggestion was made, then or later, that the failure to pay that rate was a breach of any promise nor that the claimants nevertheless had a contractual right to have their bonus pool calculated on the basis of the "institutional market rate".
55. The point in fact goes further. It is not just that there was silence. Mr Brogden did make arguments, particularly at the end of the financial year 2010//2011, that the bonus calculation for his desk should be adjusted because it undervalued the funds which the desk had raised for the Bank. However, those arguments were couched in terms of what was fair. Not once – not even when he decided to resign over the issue of bonuses and accused the Bank of acting in breach of contract – did he allege that he had an agreement with the Bank of which the Bank was in breach, which obliged Investec to calculate the claimants' bonuses by reference to the "institutional market rate".
56. I will mention some of the occasions on which, if Mr Brogden had believed at the time that he had an agreement that the Bank would calculate bonuses using the "institutional market rate", I am sure that Mr Brogden would have mentioned that fact:

i) I referred earlier to the fact that, on being sent EVA figures for 2008/2009 by Mr McKenna on 14 May 2009, Mr Brogden emailed Mr Van Der Walt to say that "this is some distance away from my view of what is right, fair and/or realistic". It was Mr Brogden's evidence that by "right" he meant that the figures did not reflect the method of calculation that he and Mr Van Der Walt had originally agreed and that this "was my way of saying that this was a breach of contract". I consider this fanciful. If Mr Brogden had meant or wanted to say that the figures did not reflect a method of calculation that he and Mr Van Der Walt had agreed, then I am sure that that is what he would have said.

ii) At around this time Mr Brogden prepared a note for himself of differences between the Bank's figures for P&L and what Mr Brogden described as the "market standard". In this note Mr Brogden wrote that it is "market standard for the structured product desk to pay a 'friendly' rate to the bank's internal treasury desk". (It is clear from the context and in particular the fact that Mr Brogden's note treated the 'friendly' rate as leading to understatement of the SED desk P&L, that Mr Brogden was, despite the wording of his note, referring to the rate paid to the SED desk by the Bank's Central Treasury. Mr Brogden confirmed in his oral evidence that this was so.) He went on to calculate what he described as "Theoretical P&L", said to be "based on more realistic funding rate assumptions". I do not think it conceivable that Mr Brogden would have written the note in the terms that he did if at the time he had believed that the Bank had agreed to calculate P&L for the SED desk on the basis of funding rates which were not "market standard".

iii) In the bonus discussions for the following year (2009/2010) Mr Brogden was now dealing in the first instance with Mr Clapham who had not been a party to the discussions which led to the claimants joining the Bank. If there was an agreement governing the way in which the SED bonus pool was to be calculated, it would have been necessary to ensure that Mr Clapham was aware of this. However, Mr Brogden did not.

iv) In an email sent on 16 April 2010 to Mr Clapham, Mr Van Der Walt described a conversation with Mr Brogden about the SED bonus pool numbers. It is clear from this email and from a further email which Mr Van Der Walt sent to Mr Brogden himself on 5 May 2010 that in their discussion Mr Brogden had put forward an argument that the rates of interest paid to the SED desk by the Bank's Central Treasury did not reflect the value to the Bank of the deposits raised and that this should be taken into account when considering bonus. Mr Brogden's argument was based on the proposition that "if we [i.e. the Bank] had gone to the markets to raise term cash then we would have had to pay more". In his email to Mr Brogden Mr Van Der Walt set out reasons why he considered the argument to be flawed. I cannot believe that, if Mr Brogden had made an agreement with Mr Van Der Walt that the SED desk would be credited with the rate of interest that Investec would have had to pay in the institutional market, Mr Brogden would have neglected to remind Mr Van Der Walt of that fact. The fact that he did not and put forward only an argument of a theoretical nature is in my view in and of itself a compelling demonstration that there was no such agreement.

v) One of the points made by Mr Van Der Walt in his email rebutting Mr Brogden's argument was that "we generally do not reward on an EVA basis for deposit gathering". If, despite what may generally have been the case, such an arrangement had been specially agreed, I am sure that Mr Brogden would have pointed that out. Likewise, when in a later email sent on 14 May 2010 (quoted in paragraph 30 above) Mr Van Der Walt expressed concern that he and Mr Brogden did not seem to have a mutual understanding of "how we operate and what would be expected" and invited further discussion, Mr Brogden made no attempt to dispel the misunderstanding by reminding Mr Van Der Walt of what he now says they had previously agreed.

vi) It is most telling of all that Mr Brogden made no reference to the alleged oral agreement at the end of the 2010/2011 year, when differences about bonus led to his resignation from the Bank. In these discussions Mr Brogden put the fact that the interest rates paid to his desk by the Central Treasury were significantly below market rates and the effect which this had on the calculation of bonus at the centre of his arguments. Yet he conspicuously did not at any point suggest that the claimants were entitled to be credited in their bonus calculation with market rates because of an alleged agreement.

vii) This omission continued even when Mr Brogden resigned from his employment with the Bank. In his resignation letter dated 20 July 2011 he alleged a breach of contract in relation to the payment of bonus but it was a different breach from those now asserted (being that he had not been allowed to decide on the allocation of the bonus pool among members of his team). It was only in a letter before claim sent by his solicitors on 21 October 2011 that it was asserted for the first time that during their pre-contractual negotiations Mr Van Der Walt had told Mr Brogden that the Bank would pay the "institutional market rate" on sums raised. At this stage the agreement was said to have been made "in around November 2006".

57. In his evidence Mr Brogden gave two explanations for why he never mentioned the alleged oral agreement to credit his desk with the "institutional market rate" on funds throughout the period of his employment with the Bank. One explanation was that he did not need to mention it in his conversations with Mr Van Der Walt as they both knew what they had agreed. This does not bear scrutiny. There came a point, very early on, at which if Mr Brogden's evidence is correct it must have been obvious to him that Mr Van Der Walt had either forgotten or was choosing to ignore what they had agreed. At that point a reminder was plainly needed and would, I am sure, have been given if Mr Brogden had at that stage recalled any such agreement. It would in any event have been necessary to ensure that other relevant individuals were aware of the agreement, such as Mr Clapham. Yet even at a meeting with Mr Clapham on 9 May 2011, by the end of which Mr Brogden said that he was so shocked and angry at Mr Clapham's assertion that there was no entitlement to bonus that he was physically shaking, he did not enlighten Mr Clapham as to the basis of his entitlement.
58. Mr Brogden's second explanation was that he deliberately refrained from mentioning his contractual rights because he did not want to 'rock the boat' and believed that, as soon as he did so, "it would severely damage or even end our working relationship". For example, he said at one point in his oral evidence:

"I am trying to find a sensible solution for everybody here. I am not trying to hold people to contracts. At the moment I say we have [a] contract, we have a clear dispute and I think at that point the working relationship is basically ended."

59. I do not doubt that Mr Brogden, believing as I accept that he now genuinely does that he made an oral agreement with Mr Van Der Walt about paying the "institutional market rate", has rationalised his behaviour to himself in this way. However, again the explanation does not stand a moment's serious scrutiny. There was, in the first place, plainly no need for Mr Brogden to have reminded Mr Van Der Walt of their alleged agreement in a way that was confrontational. Doing so – if there had been such an agreement – would, on the contrary, have been the obvious starting point for trying to solve their differences. Secondly, it is in any event plain that Mr Brogden was not shy about asserting a contractual entitlement when he believed that he had one, and in holding people to contracts. In June 2009 Mr Brogden and Mr Reid received letters stating that their bonuses were being split into a cash portion and a non-cash portion to be paid in the form of shares. Mr Brogden immediately telephoned Mr Van Der Walt and told him that there was no contractual basis to defer any of their bonuses, as the claimants' contracts stipulated that their bonuses should be paid in cash. After this telephone conversation with Mr Van Der Walt on 8 June 2009 Mr Brogden sent himself an email entitled "note to self" to make a record of it. A third reason why this explanation is patently unsustainable, is that there came a time when Mr Brogden did have a clear dispute with the Bank and when he concluded that the working relationship had ended. At that point any possible reason for inhibition about referring to contractual rights had on any view disappeared. Yet, as I have indicated, nothing was said about any agreement.

Possible discussions

60. Despite all this, I do not think it likely that the evidence given by Mr Brogden and Mr Reid about their discussions with Mr Van Der Walt is a pure invention on their part, let alone a dishonest one. Not only did they seem to me to be sincere and straightforward individuals, but the account they gave does not have the hallmarks of deliberate concoction; it much more likely has its origin in something actually said, however great the distortion in the claimants' recollections. The effect of Investec's credit rating and the rates of interest which the Bank would charge or pay on funds borrowed or raised by the claimants' desk were relevant considerations for the Bank and the claimants when discussing the potential to build a profitable equity derivatives business. There is, moreover, some documentary evidence that these topics were indeed discussed. An internal email sent by Mr Van Der Walt after meeting Mr Brogden on 14 November 2006 included the following points:

"5. Credit rating is not an issue, they can structure around this. I asked this 3 or 4 times.

6. They can be number 1 in UK retail products without Abbey reliance and do not need our credit to play. The UK retail market is £4 - £5 bil per annum and sold to building societies and life offices.

7. They do not need a lot of cash to run the business and can structure to be short or long cash. Currently stay short due to internal prices. Told him we charge LIBOR +1/8th and pay LIBOR from central treasury.

8. Abbey retail network makes the[m] £10mill per annum – this is a nice to have but the business will stand alone and it is not reliant on Abbey."

61. A two page outline business plan which Mr Brogden had sent to Mr Van Der Walt on 1 November 2006 referred to retail structured products in the context of "flow business", which I understand to mean trading on behalf of clients, and stated:

"Although not the primary source of profit for the desk, it is important to have access to flow business in order to either offset other (more profitable) trades, or as an efficient way for a trader to take a proprietary view. Retail structured products would be a good source of flow for the business and we would aim to cover virtually all of the UK retail structured product market as well as some parts of mainland Europe."

This description and the statement in Mr Van Der Walt's email that the claimants could "structure to be short or long cash" supports Mr Van Der Walt's evidence that involvement in the retail structured product market was at that stage expected to take the form of selling derivatives to institutions which issued such products and that the claimants were not expecting to be taking deposits. Furthermore the reference to LIBOR as the rate paid by Central Treasury could be what Mr Brogden has remembered as a statement that the Bank would pay the institutional market rate on funds deposited with the Central Treasury by the SED desk.

62. It is common ground that at the time of the meetings with Mr Van Der Walt (i) inter-bank credit was cheap and readily available and (ii) the SED business was not expected to be a substantial deposit taker. As Mr Brogden said in his second witness statement: "the reality is that neither Mr Van Der Walt nor I foresaw that Investec's credit spread would widen so dramatically as it did in the period 2008-09." Nor did the claimants and Mr Van Der Walt foresee that the sale of retail structured products would become an important source of term funding for the Bank.
63. I accept, however, that one kind of transaction which the claimants may have contemplated as a possibility and which might have been mentioned at the meeting which Mr Reid attended would have involved the Bank providing a retail structured product for another bank or financial institution to sell on terms whereby Investec received the customer's deposit after deduction of a fee charged by the issuer. A term sheet prepared by Mr Reid in May 2007 before the claimants had actually started work at the Bank illustrates an arrangement of this kind. It is possible that at the meeting attended by Mr Reid a question was asked of Mr Van Der Walt about what rate the Bank would be prepared to pay on funds raised in this way and that Mr Van Der Walt indicated that the Bank would pay the desk the full market rate. This would have been relevant information in order to gauge whether such business would be viable.
64. I think it most unlikely that, if such a discussion took place (and I make no positive finding that it did), Mr Van Der Walt would have used the expression "institutional market rate", which he says – and I accept – was not part of his vocabulary. The likely gist of any such discussion is that Mr Van Der Walt indicated that the Bank would pay to the SED desk the same rate as it would expect to pay to borrow the funds from another bank.
65. If any such statement was made, however, I am sure that it was not understood and could not rationally have been understood as a contractual promise, binding on the Bank, to pay an institutional rate of interest or any particular rate of interest if the claimants were to join the Bank and if in the course of their employment they were to raise funds from selling retail structured products. To be fair to the claimants, it is not their case that a binding promise of that nature was given. Their case, as Mr Cavanagh QC expressly reiterated in his closing submissions, is not that Mr Van Der Walt bound the Bank's Treasury to pay the "institutional market rate" of interest on any term funds raised by the claimants at any time during their employment if they joined the Bank; it is that he made a binding agreement that any bonuses which the Bank agreed to pay to the claimants would be calculated using that rate.

The claimants' own evidence does not support their case

66. This leads me, however, to the third main reason why I consider the oral agreement alleged by the claimants to be incapable of belief. This is that the claimants' own evidence of what was discussed with Mr Van Der Walt, even if accepted, does not support the existence of such an agreement. As counsel for the Bank pointed out, the discussions which Mr Brogden and Mr Reid described in their witness statements were discussions about the rate of interest which the Bank's Central Treasury would pay on funds raised by the SED desk. They were not discussions about how the claimants' bonuses would be calculated.
67. Mr Brogden said in his oral evidence that it was all one discussion, the reason being that the rate of interest paid by the Bank's Treasury would be one element in the performance of the desk and would therefore feed through into the calculation of the bonus. It is plainly true that the interest rate paid would affect the bonus calculation. It is one thing, however, to have had a discussion about the rate of interest which the Central Treasury would pay; it would be quite another to discuss and agree that, if for some reason the Central Treasury did not pay an institutional market rate on funds raised by the desk, the claimants' bonuses would nevertheless be calculated as if it had. I think it inconceivable that any discussion to that effect took place at any of the meetings with Mr Van Der Walt. As the possibility is not one that was envisaged, there would simply have been no reason to have discussed how the claimants' bonuses would be calculated if it occurred. Nor did I understand the claimants to suggest that there was such a discussion. Had they done so, I could not have accepted that they were being honest.

Conclusion

68. For these reasons, I am sure that no oral agreement was made during the pre-contractual discussions between the claimants and Mr Van Der Walt about the rate of interest on funds raised by the SED desk to be used for the purpose of calculating their bonuses. I am all the more sure that no agreement was made that would require any different rate to be used than the rate that was actually paid on such funds by the Bank's Central Treasury.

Other matters allegedly agreed

69. In addition to the point about the institutional market rate, the claimants allege that a number of other matters, which I have set out in paragraph 39 above, were agreed in their pre-contractual discussions with Mr Van Der Walt.
70. I see no reason to suppose, and I do not accept, that there was any express negotiation of those matters, let alone one which was intended to create a binding contract between the claimants and the Bank. I do, however, accept that the discussions were predicated on the assumption that, if the claimants joined the Bank, their desk would operate and be accounted for internally as a separate business unit within the Bank. I take that to be standard industry practice and in any event the way in which Investec operates, as would have been explained by Mr Van Der Walt. It is also implicit in the notion that bonuses would be determined by the performance of the equity derivatives business. To measure such performance, it is necessary to identify the revenue and costs attributable to that business, treating it as a separate business from other operations of the Bank.
71. I also accept that it would have been understood that the SED business would be accounted for on a mark to market basis so that unrealised profits and losses would be taken into account. Again, my understanding from the evidence is that this is standard practice in the banking industry. The claimants would also have expected that in preparing separate management accounts for the SED business fair accounting principles would be applied.
72. I find that none of these matters, however, was the subject of an oral agreement.

E. INTERPRETATION OF THE BONUS CLAUSE

73. I turn to the second way in which the claimants put their case on the rate issue, based on the proper interpretation of the bonus clause in their contracts of employment.

The applicable legal principles

74. In what has become the standard formulation of the legal test, the basic aim of contractual interpretation is said to be to ascertain what a reasonable person having all the background knowledge which would reasonably have been available to the parties in the situation they were in at the time of the contract would have understood the parties to have meant by the words they used: see Investors Compensation Scheme v West Bromwich Building Society [1998] 1 WLR 896, 912H–913D, per Lord Hoffmann; Chartbrook Ltd v Persimmon Homes Ltd [2009] UKHL 38; [2009] 1 AC 1190, paras 14 and 21; Rainy Sky SA v Kookmin Bank [2011] 1 WLR 2900, paras 14 and 21. Counsel for the claimants in their submissions referred to this test as the 'reasonable bystander test'.
75. I am not persuaded that the 'reasonable bystander test', to adopt that label, adequately captures the extent to which the interpretation of contracts in English law is an objective task, and is not concerned with the subjective intentions of the parties. To refer to what a reasonable person would have understood the parties to have meant by using the language which they did might still be taken to suggest that the task of the court is to identify, so far as it can, an actual shared intention of the parties – albeit that the court has no window into the minds of the parties and must necessarily draw inferences from the language of their contract understood in its factual context about what that intention is. The truth is, however, that "actual intentions are happily irrelevant, since, were it otherwise, many, and perhaps most, disputes upon points of construction would be resolved by holding that the parties were not ad idem": Summit Investment Inc v British Steel Corp (The "Sounion") [1987] 1 Lloyd's Rep 230, 233, per Sir John Donaldson MR. The court identifies the meaning of contractual language not simply by adopting the point of view of a reasonable bystander but by assuming that the parties themselves were reasonable people using the language of the contract to express a common intention. As Lord Wilberforce said in Reardon Smith Line Ltd v Hansen-Tangen (The "Diana Prosperity") [1976] 1 WLR 989, 996:

"When one speaks of the intention of the parties to the contract, one is speaking objectively ... and what must be ascertained is what is to be taken as the intention which reasonable people would have had if placed in the situation of the parties."

76. One important principle relevant in ascertaining the intention which reasonable people would have had if placed in the situation of the parties is that, if contractual language is capable of more than one interpretation, the court generally should prefer the interpretation which is most consistent with business common sense: see Rainy Sky SA v Kookmin Bank [2011] 1 WLR 2900, paras 20-30.
77. A further well established principle, reaffirmed by the House of Lords in Chartbrook Limited v Persimmon Homes Ltd [2009] 1 AC 1103, is that the law excludes from admissible background the previous negotiations of the parties. However, the law does not exclude the use of such evidence for the purpose of establishing that a fact that may be relevant as background was known to the parties. This includes facts communicated by one party to the other in the course of the negotiations: see Chartbrook Limited v Persimmon Homes Ltd [2009] 1 AC 1103 at para 42; Oceanbulk Shipping SA v TMT Asia Ltd [2011] 1 AC 662 at paras 40 and 48.

The meaning of "EVA"

78. I repeat for convenience the language of the claimants' contracts of employment which, for the 2010/2011 year, determined their entitlement to bonus:

"In the third financial year starting on 1 April 2009 and thereafter the bonus calculation will be normalised based on a formula calculated as 30% of EVA generated by the Equity Derivative business. The bonus pool available for distribution to the members of the team will be calculated after deduction of NI, an amount for Head Office and an amount for TSF support."

79. Although the clause must of course be read as a whole, the key term used is "EVA". No case has been advanced by either party that this term has a standard market meaning. Still less is it a term used in ordinary speech. Although it was not one of the issues on which the accountancy experts were asked to opine, the claimants' expert, Mr Brice, stated his understanding that "EVA" is a trademarked concept created by the US consulting firm Stern Stewart & Co as a measure of financial performance. Mr Brice also quoted from and exhibited to his report various US academic publications discussing the concept. It has not been suggested, however, that the parties would reasonably have consulted any of this literature or had it in mind when agreeing the claimants' contracts of employment.
80. What the parties would reasonably, and indeed clearly did, have in mind was how a measure of performance referred to as "EVA" was used within Investec. The evidence showed that within the Bank the term "EVA" had a particular, conventional meaning. It meant revenue minus costs, minus the cost of capital, all calculated before tax.
81. This meaning may or may not correspond to how the term "EVA" is used by Stern Stewart, or anyone else.^[2] It was, however, explained to the claimants in their discussions with Mr Van Der Walt before they joined the Bank as being how Investec uses the term. Mr Brogden said in his witness statement that Mr Van Der Walt told him at their first meeting on 14 November 2006 that:

"Investec preferred to use EVA, rather than revenue, to measure financial performance, and that EVA was defined as revenue minus costs minus cost of capital, all calculated before tax."

Similarly, it was Mr Reid's evidence that the concept of "EVA" was explained to him in this way by Mr Brogden in early January, when Mr Brogden told him about the opportunity to join Investec and described the Investec model for measuring performance and paying bonuses.

82. Moreover, the word "normalised" in the bonus clause itself indicates that the method for calculating bonus provided for in the claimants' contracts was not a one-off, but was the normal method used within the relevant part of the Bank to calculate bonuses. This was explicitly stated in an email dated 19 March 2007 from Mr Van Der Walt to Mr Brogden which described the Bank's proposal to offer high upfront payments to induce the claimants to join Investec and then "to pay a more normalised EVA down the line to help with integration and alignment. … In year 3 and thereafter a normal EVA model would apply at a rate proposed of 30% which is how the other businesses are remunerated …" Mr Brogden sought to argue that the words "normalised" and "normal" in this description referred only to the percentage of EVA which the claimants would receive. Clearly this was part of what aligning the claimants with other businesses would involve. But it is abundantly clear that receiving 30% of EVA could only be normal within the Bank because the norm was to calculate bonuses as a proportion of EVA, and that EVA was an established measure within the Bank calculated using a model which the Bank used as the basis for remunerating all its trading businesses.
83. It is clear beyond doubt that Mr Brogden and Mr Reid both understood this when they entered into their contracts of employment. Indeed, on their own evidence it was a major attraction of Investec and reason for deciding to leave their previous jobs that their bonus pool would not be a matter of discretion but would be calculated in accordance with a pre-determined formula, used to measure the financial performance of the business conducted by the claimants.
84. Against the background of Investec's established method for measuring the financial performance of the businesses within its Capital Markets Division and for calculating bonuses using the measure known within the Bank as "EVA", all of which was known to the claimants, reasonable people in the position of the parties would have understood and intended the term "EVA" in that sense. In other words, the phrase "EVA generated by the Equity Derivative business" used in the claimants' contracts of employment on its proper interpretation means "the amount calculated as the 'EVA' of the Equity Derivative business using the method normally used by the Bank to calculate a measure of performance known as 'EVA' for each business unit".
85. This interpretation is reinforced by the next sentence of the bonus clause which states:

"The bonus pool available for distribution to the members of the team will be calculated after deduction of NI, an amount for Head Office and an amount for TSF support."

It could not sensibly be supposed the amounts for Head Office and TSF support should be decided other than by Investec. This confirms that it is Investec who is to calculate the bonus pool.

86. On behalf of the claimants, Mr Cavanagh QC submitted that the key to the proper interpretation of the clause is that "EVA" is an abbreviation for "Economic Value Added" and that reasonable people in the situation of the parties would have intended that the calculation of "EVA" should be carried out using those accounting principles and methods which best measure objectively the profitability of the Equity Derivative business considered as a stand alone entity. If the parties cannot agree on what those principles and methods are or on how they should be applied, then the court must adjudicate. It cannot reasonably have been intended, he submitted, that the "EVA" of the Equity Derivative business should mean whatever Investec calculates it to be. Such an interpretation would entail that there would be no objectivity at all in determining the profitability of the SED desk; it would defeat the object of providing for bonus to be calculated according to a formula rather than being a matter of the Bank's discretion and would deprive the claimants of any protection against Investec deciding unilaterally to ascribe a value to activities of the SED desk which was wholly inconsistent with the true, objective value of those activities. All of this, Mr Cavanagh submitted, would be completely contrary to business common sense.
87. A fundamental problem with this argument, as it seems to me, is that even if in some ultimate sense, viewed sub specie aeternitatis, there is a 'true', 'objective' or even 'best' method for measuring the economic performance of a business, there is no consensus – and endless scope for disagreement – about what that method is. It cannot reasonably have been supposed that, for the purpose of calculating the claimants' bonuses, the parties should have to construct a measure of their economic performance from first principles and resort to independent adjudication if there was any aspect of the method to be used about which they could not agree. I agree with the submission made by counsel for Investec that this would not be a formula for a bonus system but a recipe for litigation, and would be commercially absurd.
88. A further reason why this interpretation does not make business sense is that it would potentially divorce the calculation of the claimants' bonuses from the methods and numbers used by the Bank in preparing the accounts which form the basis on which management decisions are made, the financial position of the Bank is reported under the Companies Act and dividends are paid to shareholders. It would mean that potentially, for each team within the Bank which had a bonus formula based on "EVA", the calculation of EVA would be a matter for separate debate and might have to be done differently. All this would defeat the purpose of using EVA as a common metric.
89. In any event, the system operated by Investec and explained to the claimants before they joined the Bank did not involve at the end of each financial year attempting to work out the economic value added by each business unit starting from the ground up. It involved using the figures contained in the Bank's books and records. Those figures were generated on a running basis by the Bank's automated systems and adjusted by the Bank's accountants at the year end. I accept the evidence of the Bank's witnesses, which is borne out by many documents, that the concept of EVA was, in Mr Van Der Walt's words, "embedded in Investec's process and policy" and I am sure that Mr Van Der Walt conveyed this to the claimants in the course of their pre-contractual discussions. Against that background, I consider that interpreting "EVA" to mean "the figures calculated as 'EVA' by the Bank" is the only interpretation which makes sense.
90. I accept the claimants' submissions, however, to the extent that it also cannot reasonably have been supposed that Investec would have a completely free rein to calculate the EVA of the SED business however it liked and use that figure to determine the claimants' bonuses. There were, in my view, implied limits to how the calculation could be performed consistently with the claimants' contracts of employment. These limits are shown by two lines of authority.

Good faith and rationality

91. Counsel for the claimants relied on the now well established principle that in the absence of very clear language to the contrary, a contractual discretion must be exercised in good faith for the purpose for which it was conferred, and must not be exercised arbitrarily, capriciously or unreasonably (in the sense of irrationally): see e.g. Abu Dhabi National Tanker Co v Product Star Shipping Ltd, (The 'Product Star')(No 2) [1993] 1 Lloyd's Rep 397, 404; Paragon Finance Plc v Nash [2002] 1 WLR 685, paras 39-41; Socimer International Bank Ltd v Standard Bank London Ltd [2008] 1 Lloyd's Rep 558, 575–577; British Telecommunications Plc v Telefónica O2 UK Ltd [2014] UKSC 42, para 37. That principle applies to discretions in relation to remuneration like any other contractual discretion. The leading authorities in which the principle has been applied to payments of discretionary bonuses are Clark v Nomura International plc [2000] IRLR 766, Horkulak v Cantor Fitzgerald International [2004] EWCA Civ 1287, [2005] ICR 402, and Keen v Commerzbank AG [2006] EWCA Civ 1636, [2007] ICR 623. As expressed by Burton J in Nomura at para 40:

"My conclusion is that the right test is one of irrationality or perversity (of which caprice or capriciousness would be a good example) ie that no reasonable employer would have exercised his discretion in this way."

92. This principle is buttressed in the employment context by the overarching obligation implied by law as an incident of the employment contract that an employer will not without reasonable and proper cause conduct itself in a manner calculated or likely to destroy or seriously damage the relationship of confidence and trust between employer and employee: see Malik v Bank of Credit and Commerce International SA [1998] AC 20. The essence of this obligation is one of fair dealing: see Johnson v Unisys Ltd [2001] UKHL 13; [2003] 1 AC 518, para 24, per Lord Steyn. As explained by Lord Nicholls (with whom Lords Hoffmann, Rodger and Brown agreed) in Eastwood v Magnox Electric plc [2004] UKHL 35; [2005] 1 AC 503, para 11, it means "in short, that an employer must treat his employees fairly in the conduct of his business" and that "in his treatment of his employees, an employer must act responsibly and in good faith". It allows "a balance to be struck between an employer's interest in managing his business as he sees fit and the employee's interest in not being unfairly and improperly exploited": see Lord Steyn in Malik at 46D.
93. The obligation of trust and confidence has been described as "undoubtedly the most powerful engine of movement in the modern law of employment contracts": Freedland, The Personal Employment Contract (OUP, 2003), p.166. It will inform and regulate the contours of any contractual discretion. By way of recent example, in Attrill and Annar v Dresdner Kleinworth [2013] EWCA Civ 394, [2013] 3 All ER 607, the Court of Appeal held that on the facts of that case the obligation of trust and confidence precluded the exercise of a unilateral contractual power to vary the terms of contracts of employment to allow the employer to withhold contractual bonuses in the event of a "material adverse change".
94. Counsel for Investec accepted, as they were bound to do, that a power conferred by a contract on one party to make decisions that affect them both must be exercised honestly and in good faith and not arbitrarily, capriciously or irrationally. They submitted, however, that it is a necessary precondition to the application of this principle that there is a contractual discretion to regulate. On behalf of Investec, Mr Nash QC was prepared to concede that it would probably be an implied term of the contract that the Bank would not manipulate the P&L in order to reduce the claimants' bonus entitlement. He did not accept, however, that any further constraint of rationality applied. He argued, in particular, that although the preparation of accounts such as were necessary for the purposes of calculating the claimants' bonus pool involves questions of judgement about which reasonable people may differ, it does not involve the exercise of a discretion.

Discretion and its limits

95. As the late Professor Ronald Dworkin observed, discretion, like the hole in a doughnut, does not exist except as an area left open by a surrounding belt of restriction.^[3] It is therefore a relative concept. Like all terms, its meaning is sensitive to context. In legal argument the term 'discretion' is used in several different senses.
96. The term is sometimes used in a weak sense, simply to mean that the standards that a decision-maker must apply cannot be applied mechanically and demand the use of judgement. In Professor Dworkin's example, a sergeant has a discretion in this sense if he is told to select his five most experienced men to take on patrol. Another sense in which the term is used is to indicate that a person has final authority to make a decision which cannot be reviewed or reversed by someone else: for example, when we say that the umpire's decision is final. The term 'discretion' is also used in a stronger sense to indicate that the authority which imposes duties on someone does not control their decision of a particular issue. Dworkin's sergeant would have a discretion in this sense if he was simply told to select any five men of his choice to take on patrol.
97. In a contractual context references to a 'discretion' are often to a discretion in the strong sense, meaning that the contract confers a duty or a power on one party to take a particular decision but does not expressly specify any standard which controls or constrains the decision. An example is Paragon Finance Plc v Nash [2002] 1 WLR 685, where the contract gave a mortgage lender an apparently unfettered discretion to vary the rate of interest charged to the borrower. However, by no means all the cases where a duty to act rationally and in good faith has been implied have involved an exercise of discretion in this strong sense. For example, in The 'Product Star' the relevant discretion was that of a shipowner to decline to obey the charterer's order to proceed to a port if the owner considered it dangerous. In that case the contract expressly specified a standard – whether the port was dangerous – which had to be applied. However, there was clearly considerable scope for differences of opinion about whether the standard was met, and there was therefore a discretion in the first, weaker sense. The Court of Appeal held that the shipowners were in breach of contract because they had not honestly believed that the port in question was dangerous and such a conclusion would in any event have been unreasonable and capricious.
98. Similarly, in Socimer International Bank Ltd v Standard Bank London Ltd [2008] 1 Lloyd's Rep 558, the contract gave one party (the seller) the right to liquidate or retain a portfolio of assets and provided that the value of the assets was to be determined on the relevant date by the seller. Determining the value of an asset is not simply a matter of choice but a matter of judgement, albeit one where there can be a variety of different techniques used and differences of view about how those techniques should be applied. The Court of Appeal rejected the judge's interpretation that the seller was bound to identify the true market value of the assets and interpreted the clause as one which gave the seller a discretion limited by constraints of good faith and rationality.
99. In Compass Group UK and Ireland Ltd (t/a Medirest) v Mid Essex Hospital Services NHS Trust [2013] EWCA Civ 200, a case cited by counsel for Investec, Jackson LJ reviewed some of the authorities, including The 'Product Star' and Socimer, and said (at para 83):

"An important feature of the above line of authorities is that in each case the discretion did not involve a simple decision whether or not to exercise an absolute contractual right. The discretion involved making an assessment or choosing from a range of options, taking into account the interests of both parties. In any contract under which one party is permitted to exercise such a discretion, there is an implied term."

The reference to "taking into account the interests of both parties" in this passage must mean that the decision is one which affects the interests of both parties, since in none of the cases was there any express requirement to take the interests of the other party into account.

100. Both on the authorities and as a matter of principle, it seems to me that where a contract gives responsibility to one party for making an assessment or exercising a judgement on a matter which materially affects the other party's interests and about which there is ample scope for reasonable differences of view, the decision is properly regarded as a discretion which is subject to the implied constraints that it must be taken in good faith, for proper purposes and not in an arbitrary, capricious or irrational manner. Those limits apply in circumstances where the decision is final and binding on the other party in the sense that a court will not substitute its own judgment for that of the party who makes the decision. There is therefore also a discretion in the second sense distinguished earlier. The concern, as Rix LJ observed in Socimer at para 66, is that the decision-maker's power should not be abused. The implication is justified as a matter of construction to give effect to the presumed intention of the parties. In the employment context it is further reinforced by the obligations of good faith and fair dealing which, as discussed above, are a necessary incident of the employment relationship.
101. As I have construed the bonus clause in the claimants' contracts of employment, the clause conferred on Investec a discretion in the relevant sense to determine the EVA generated by the equity derivative business – a determination which involved numerous and substantial exercises of judgement and which fixed the amount of bonus that the claimants were entitled to receive. I conclude that the assessment of EVA was subject to implied requirements of good faith and rationality, as contended for by the claimants as their alternative case.

The allegations of bad faith and irrationality

102. Although a case was pleaded in the particulars of claim that the Bank acted in bad faith and deliberately manipulated or adjusted the inputs to the EVA calculation in order to reduce the claimants' bonuses, that case was not put in cross-examination to any of Investec's witnesses. In those circumstances, although it was not formally abandoned in the claimants' closing submissions, it was not open to them to maintain it. There was in any event not the slightest evidence to support the allegation of bad faith. The relevant standard by which to test the calculation of EVA for bonus purposes is therefore that of rationality.
103. Against that benchmark I turn to consider the claimants' particular complaints about the calculation of EVA for the 2010/2011 year, starting with the "rate issue".

F. THE RATE ISSUE

104. As already discussed, in calculating the profit and loss and EVA generated by the SED business from sales of retail structured products, the SED desk was credited with interest by the Bank's Central Treasury on the funds raised. The rate of interest credited was determined before each product launch and was taken into account by the claimants and their team in structuring the product offered to investors. In order to make a profit for the desk, there had to be a margin between (i) the implied rate of interest paid to investors plus the costs of issuing the product and (ii) the rate of interest paid to the desk by the Central Treasury. As also mentioned earlier, the rate of interest paid by the Central Treasury was, to begin with, agreed between Mr Van Der Walt and Mr Brogden for each product launch; but from around July 2009 the rate was set by a body called the Product and Pricing Forum. The rate set was a function of the Bank's appetite for funds of the relevant kind, the other sources of funding available to the Bank and the price which the SED desk thought it necessary to offer to make the relevant product attractive to investors.

The claimants' case

105. It is the claimants' case that using the rate of interest set in this way in the calculation of EVA was not a rational means of assessing the performance of the SED desk as a stand alone business. The claimants argue that for that purpose it was necessary to treat the SED desk as if it was a separate entity dealing with the Bank on an arm's length basis. The rate of interest credited to the desk by the Central Treasury, they argue, was not an arm's length rate. Rather, it was a rate fixed according to the Bank's own particular needs and purposes, which therefore did not necessarily reflect the fair market value of the funds raised. Much emphasis was placed by counsel for the claimants on a statement in a draft document about "funds transfer pricing policy" dated May 2011 that "profit/loss on liability raising desks is essentially fabricated and is simply determined and manipulated by whatever internal rate is placed on the funds." An earlier statement of the policy to similar effect used the term "fictitious" instead of "essentially fabricated", while the version included in Investec's Individual Liquidity Adequacy Assessment dated June 2011 used the phrase "essentially 'left pocket, right pocket'". All these formulations, Mr Cavanagh submitted, amounted to admissions by Investec that the rate of interest credited internally to the SED desk on funds raised by the desk was arbitrary and not a rate which could rationally be used in calculating the EVA generated by the desk.
106. The claimants maintain that, in order properly to calculate the profit generated by the SED desk on a stand alone basis, it is necessary to measure the value of funds raised by comparison with the rate of interest which Investec would have had to pay to borrow an equivalent volume of money for an equivalent term from an alternative source by transacting in the market with other institutions on an arm's length basis. The claimants contend that for the purpose of this comparison the relevant market is the institutional bond market and the applicable rate of interest is therefore the 'institutional market rate'.

Discussion

107. I am not persuaded by this argument. For what it is worth, I would not have considered it right to calculate the profit generated by the SED desk by reference to the 'institutional market rate' even if I had accepted the claimants' contention that on the proper interpretation of their contract it is for the court to determine the appropriate methodology. Still less do I consider that the approach used by Investec was irrational in the sense of being one that no reasonable investment bank could have used.
108. In the first place, the 'institutional market rate' could only be an appropriate rate of interest to use if the Bank would have paid that rate for equivalent funds had funding from the sale of retail structured products not been available. However, on the evidence that proposition was not made out. Mr Barbour's evidence, which I accept, was that the retail structured products issued by the SED desk were a relatively cheap source of term funds but were not the only source of term funds available to the Bank. Other sources of similar funds which the Bank would have used or used to a greater extent if the funds raised by the SED desk had not been available included the High 5 account – a form of savings account which was extremely successful – and retail bonds. Investec might also have raised a smaller volume of funds. Given the extremely high cost of institutional borrowing in the aftermath of the financial crisis, it is very unlikely that Investec would have sought to raise money in the wholesale bond market.
109. Mr Brogden in his oral evidence accepted that at the relevant time it would not have been commercial for Investec to raise money in the bond market and that, if the Bank had had to pay the 'institutional market rate' for funds it would probably not have been able to make a profit from lending those funds on the asset creation side of its business. Mr Brogden later withdrew that evidence saying that he had overlooked the fact that the funds raised by the SED desk represented only a comparatively small part of the Bank's overall balance sheet. He said that, since the cost of funds charged by the Central Treasury to the asset creators was a blended rate which reflected the overall cost of borrowing from all sources, paying the institutional market rate on the funds raised by the SED desk would only have had a small impact on the rate charged and would not have rendered lending uneconomic.
110. It does not seem to me that this latter point made by Mr Brogden affects the substance of the argument. The fact that the cost of funds derived from all sources is blended does not make it economic to raise additional funds at a marginal rate which is not itself economic. Unless there was some regulatory or other requirement which compelled Investec to borrow money at the institutional market rate, the point therefore still remains that it would not have made commercial sense for it to do so. There was no evidence that it would have been necessary for regulatory reasons for Investec to go into the bond market to seek to raise term funds equivalent to those raised from retail structured products if the latter source of funds had not been available. I note in this regard that – according to an analysis of all retail structured product launches carried out by one of Investec's accountants, Mr Verma – of the total volume of £571m raised from Investec retail structured products launched in the 2010/2011 year, £373m (or approximately two-thirds) related to kick out products. Such products could not have satisfied a regulatory requirement for long term funding since, according to Mr Barbour, for the purpose of its liquidity adequacy assessment Investec had to make the assumption that the funds would be repaid on the first possible kick out date.
111. More generally, I can see nothing wrong let alone irrational in the Bank paying the SED desk the lowest rate of interest on funds raised which the desk was prepared to accept. There was no requirement on the claimants to issue retail structured products. They developed that business because it was business on which they hoped to and did make a profit. If the rate of interest offered to the SED desk for any proposed product launch did not enable the desk to price the product competitively and make a profit, the claimants could have chosen not to go ahead with the launch. They were also free, and indeed encouraged by Mr Van Der Walt, to develop other areas of business and not to become too reliant on a single type of product. As it was, Mr Brogden confirmed that the claimants' normal expected margin on selling structured products to other institutions, as they had done at Santander, was 2% or maybe even less; by contrast, when selling such products directly into the retail market, as they did at Investec after building such a distribution channel, the profit margin was upwards of 5%.
112. Mr Brogden agreed that the Bank could decide on the volume of funds which it wanted to have from this source, had the ability indirectly to control the volume by the rate which it was willing to offer and would have been free to decide that the desired volume was nil. The SED desk was equally free to decide whether to transact with the Bank or not. I am satisfied that in this sense the rate of interest paid by the Central Treasury to the desk was an arm's length rate.
113. I accept that the rate was not an arm's length rate in the stronger sense of being equivalent to the rate at which independent market participants would transact with each other on an arm's length basis. However, I do not see any reason why Investec was obliged to treat the SED desk as if it were an independent market participant in this way. Indeed, it seems to me that it would be artificial and lacking in commercial sense to do so. Although permitted and encouraged to operate independently in many ways, the fact remained that, when raising funds, the SED desk was doing so for the Bank and was competing, not with other enterprises in an institutional market, but with other channels of funding within the Bank. I cannot see that it would have made sense for Investec to have assessed the SED desk's contribution to its profits other than on the basis of this reality.
114. At the heart of the claimants' argument that they should be credited with the institutional market rate is the proposition that they brought economic value to Investec by raising funds more cheaply than the Bank could have raised them elsewhere. That seems almost axiomatic, since if Investec could have raised the money more cheaply from another source, it would presumably have done so. I fail to see, however, why this should entitle the claimants to have their contribution to the Bank's profit and loss calculated as if the Bank had raised the same funds more expensively elsewhere. The claimants' argument seems to me to treat a cost which the Bank has avoided as if it were a profit made by the Bank in which they were entitled to share. I am unable to see merit in that argument. In any event, and on any view, I am satisfied that the method used by the Bank for calculating the economic value added by the SED business based on the profit and loss recorded in the Banks' books cannot be regarded as irrational.

Reasonable expectations

115. The claimants also sought to rely on the principle that, if an employer has acted in such a way as to engender particular expectations in an employee, those expectations are a relevant consideration in assessing whether an employer has acted rationally. In support of this principle Mr Cavanagh cited the case of IBM United Kingdom Holdings Limited v Dalgleish [2014] EWHC 980 (Ch), where Warren J applied the principle in the pensions context and said (at para 441):

"…it seems to me that breach of expectations is, at root, an aspect of irrationality or perversity. In other words, if expectations have been engendered by an employer, that may have been done in such a way that to disappoint those expectations would, absent some special change in circumstances, involve the employer acting in a way that no reasonable employer would act; in which case, irrationality or perversity, as those concepts are to be understood in this context, is established. …"

116. Warren J drew a distinction in this regard between "reasonable expectations", which are relevant in judging whether an employer has acted irrationally or perversely, and "mere expectations", which are not. He defined a "reasonable expectation" as an expectation as to what will happen in the future engendered by the employer's own actions which gives employees a positive reason to believe that things will take a certain course. A "mere expectation", on the other hand, is one which an employee may have in fact as to the future, in the sense that they anticipate, assume or expect that something will happen in the ordinary course of events: see paras 386(iv) and 454.
117. According to the claimants, from when they joined Investec until around February 2009, Investec's Central Treasury paid the institutional market rate (or close to it) on funds raised by sales of Investec retail structured products and by the SED desk generally. Neither that fact, however, nor any other conduct of Investec to which the claimants have been able to point could have led to anything more than a "mere expectation" that Investec would continue to pay a rate of interest equivalent to the institutional market rate on such funds. It could not have engendered a "reasonable expectation" to that effect.

G. OTHER ISSUES

118. The remaining issues relate to the accounting treatment of kick out products and early bird deposits.

Reserving for kick out products

119. Many of the retail structured products devised and sold by the SED desk were kick out products which, as described earlier, had the potential to terminate early if the stock market performed well. To begin with, the Bank was unsure how to account for these products and this gave rise to some internal discussion. The difficulty was to decide when booking profit from the sale of a kick out product what, if any, reserve to make for the risk that the investor's money would have to be repaid early. The cost of such early repayment could be measured as the cost of obtaining alternative funds from other sources for the remainder of the full product term.
120. Three possible methods for reserving were considered. The first, referred to as "maximum life", was to book profit from the sale of kick out products on the assumption that they would run for the full term. This was the approach advocated by Mr Brogden. The objection to it was that it involved recognising profit which might not ultimately be realised.
121. The second possible method, referred to as "minimum life", was to reserve on the assumption that products would "kick out" on the first possible date. This was the most conservative approach, which was favoured by some of the Bank's senior management. It was used at one stage, during the 2009/2010 financial year, although not in the accounts prepared at the year end.
122. The third method, known as "expected life", was to estimate at the time of sale the expected life of the investment and establish a reserve based on the assumption that the Bank would have the funds for this period and not for the full term. This was the approach favoured by the Bank's auditors and was the approach adopted at all times except for the period during the 2009/2010 year when the minimum life method was used. The reserve established was known as the "Funding Gap Reserve". It was inherent in the "expected life" method that the reserve established on Day 1 would almost inevitably turn out to be too large or too small, depending on whether the product in fact matured later or sooner than its originally expected life. The way in which this was taken into account was by adjusting the Funding Gap Reserve continuously to reflect the expected life of the product at any given time. Such adjustments continued to be made until the product actually matured.

Adjustments to the Funding Gap Reserve

123. In the 2010/2011 year, adjustments were made to the Funding Gap Reserve for kick out products as a result of market movements and consequent changes to the expected life of the products. Those adjustments had the effect of reducing the profits of the SED business for the year.
124. The claimants have argued that the Bank was wrong to make these adjustments in the calculation of EVA used to determine their bonuses. Their argument is that the Bank's Treasury could have entered into transactions to hedge the risk that market movements would result in kick out products terminating earlier (or being expected to terminate earlier) than had been expected when profits from the sale of such products were booked.
125. In response, the Bank has argued that it was the claimants' responsibility to decide whether or not to hedge the risk of earlier kick out. Counsel for Investec referred, in particular, to an email from Mr Brogden dated 15 April 2009 in which Mr Brogden described the methods used by the SED desk for hedging risks arising from the possibility of kick outs and said:

"We have chosen not to execute this 'hedge' because we believe that commercially we are already long the market and we do not want to increase this position."

The question of hedging was discussed further at a strategy meeting held in Johannesburg on 9 July 2009. Mr Brogden gave a presentation at this meeting in which he discussed the "liquidity risk [that] arises from term hedging kick out products".

126. Mr Brogden accepted in evidence that at that time the responsibility for deciding whether or not to adopt a hedging strategy for kick out products lay with his desk. However, he said that a decision was taken by the Bank's Group Finance Director, Mr Burger, that kick out products should be used as a means of hedging the performance of the rest of the Bank's balance sheet such that it was thereafter no longer for the SED desk to decide whether or not to hedge the risks associated with kick outs. In his second witness statement made on 2 February 2014 Mr Brogden said that he and Mr McKenna had a series of discussions with Mr Burger which led to this decision in around March 2009. However, at the start of his oral evidence Mr Brogden corrected this to say that the discussions took place later in the year – no doubt because he recognised that the March 2009 date was inconsistent with his email dated 15 April 2009.
127. There is no documentary evidence of the discussions or decision said to have been made by Mr Burger, and Mr McKenna was not asked about the alleged discussions. If a decision had been made which removed the responsibility for deciding whether or not to hedge the risk of being short of funds as a result of the early termination of kick out products from the desk which issued the products, I would expect it to have been documented. I am not prepared to accept as reliable Mr Brogden's evidence that such a decision was made. I think it most likely that the discussions which Mr Brogden had in mind were in any event the discussions in March and April 2009 in which he explained why he did not think it necessary to hedge the liquidity risk arising from the kick outs and the Bank's senior management were content for him to adopt that approach. That provides no basis for saying that the risk did not lie with the SED desk.
128. A further point made by Mr Brogden was that risks associated with early kick out were not included in the daily market risk reports for the SED desk, which showed that managing the risk was not regarded as the desk's responsibility. However, it was Mr McKenna's evidence, which I accept, that the risk of having to replace liquidity was not a market risk of a kind which would ever have appeared in those reports.
129. There seems to me in any event to be a deeper flaw in the claimants' argument. I accept that from the point of view of the Bank's Central Treasury the kick out feature of structured products created a risk that funds would have to be repaid before the full term of the product. The Central Treasury might well have taken the view that this liquidity risk was not one they needed to hedge having regard to the Bank's overall balance sheet. I do not accept, however, that this has any relevance to the proper accounting treatment of kick out products in the management accounts and EVA calculations prepared for the SED desk.
130. In accounting for the SED desk as a stand alone business, the relevant risk was that the funds deposited by the SED desk with the Central Treasury would fall to be repaid early, with the result that the desk would not earn interest at the rate paid to the desk by the Central Treasury for the full term of the investment. Whether that risk, if it eventuated, would result in a loss to the desk or merely in the desk not receiving a profit that would otherwise have accrued depended on the accounting treatment adopted. On either the 'full life' or the 'expected life' approach the possibility existed that a loss would be recognised if products kicked out sooner than originally expected. If on the other hand the 'minimum life' approach was used, no loss could arise, but earlier than expected kick outs would result in the desk making less profit than it would otherwise have done.
131. No doubt if the claimants had wished to execute hedging transactions which would offset any loss or create a profit that would not otherwise have arisen in the event of an earlier than expected termination of the investment, they could have done so. However, I can see no reasonable basis for saying that the decision whether or not to hedge should make any difference to the proper accounting treatment of kick out products in the SED desk accounts. If pursuant to the accounting treatment adopted an adjustment to the Funding Gap Reserve was required because of market movements during a particular year, then that adjustment was properly made.
132. Mr Brogden was plainly aware of the Funding Gap Reserve and its purpose and of the system for adjusting it. Indeed, according to Mr McKenna, Mr Brogden helped to design the model used for this purpose. He therefore understood that, if products kicked out or became expected to kick out earlier than had previously been expected, this would result in a loss to the SED desk. This risk was one which self-evidently lay with the desk and which only the desk could have hedged by entering into transactions which would, upon the relevant movements in the market, have generated revenue for the desk to offset such a loss. In these circumstances the claimants' argument that the adjustments to the Funding Gap Reserve made in the 2010/2011 year were somehow inappropriate or wrong appears to me to be wholly unmeritorious.

The 'profit payaways' issue

133. There is a separate issue, which I gave the claimants permission to raise by an amendment to their particulars of claim made at the start of the trial,^[4] regarding the effect of 'profit payaways' made in 2010/2011 to the Funding Gap Reserve. As part of the accounting process followed at the end of each financial year, profits or losses made by a trading desk during the year are 'paid away' to the central management of the Bank so that the desk starts the next financial year with zero P&L. The Funding Gap Reserve was introduced on 31 March 2010. The profit payaway carried out in May 2010 reduced the cash balances of the SED desk. This resulted in an increase to the Funding Gap Reserve (reflecting the projected increased cost of funding flowing from the reduction in cash balances) of £4.1m.
134. In July 2010 some refinements were made to the calculation of the Funding Gap Reserve. One of these was to take into account on a monthly basis the projected removal of profits at the end of the financial year and its effect on the desk's cash balances, rather than waiting until after the year end to take this into account. A consequence of this change was that during the remainder of the 2010/2011 year the effect on the Funding Gap Reserve of the profit payaway at the end of the year was taken into account, which had a negative impact on the profits of the desk for the year of £545,453.
135. The claimants' argument is that the change made to the methodology for calculating the Funding Gap Reserve ought to have been applied retrospectively. This would mean that, when the change was introduced in July 2010, the SED desk P&L for the 2010/2011 year should have been restated so as to reduce the profit for that year by £4.1m. By that time the claimants' bonuses for the 2009/2010 year had been paid, and they do not suggest that the reduction in profit for that year should result in any retrospective adjustment to their bonuses. However, the effect of removing the effect of the profit payaway relating to 2009/2010 from the 2010/2011 year would be to increase the profits of the desk for that year by £4.1m. This would potentially affect the bonus calculation for 2010/2011.
136. The view of Investec's accountancy expert, Mr Watson-Brown – with which the claimants' expert, Mr Brice, agreed – is that the question comes down to whether the change which led to two profit payaways being taken into account in the same year was a change in an accounting policy or a change in an accounting estimate.
137. As explained by Mr Watson-Brown, the relevant accounting standard is IAS 8, "Accounting Policies, Changes in Accounting Estimates and Errors". This standard requires an entity which changes an accounting policy to apply that change retrospectively by adjusting all the "comparative amounts disclosed for each prior period presented as if the new accounting policy had always been applied" (para 22). On the other hand, an entity which adjusts an accounting estimate should recognise it prospectively "by including it in the profit or loss in: (a) the period of the change, if the change affects that period only…" (para 36).
138. The relevant definitions given in IAS 8, para 5, are as follows:

"Accounting policies are the specific principles, bases, conventions, rules, and practices applied by an entity in preparing and presenting financial statements."

"A change in accounting estimate is an adjustment of the carrying amount of an asset or a liability … that results from the assessment of the present status of, and expected future benefits and obligations associated with, assets and liabilities. Changing in accounting estimates result from new information or new developments and, accordingly, are not corrections of errors."

Both experts agreed that where the dividing line falls can be a grey area.

139. It was Mr Watson-Brown's opinion that the change made in July 2010 was a change in accounting estimate. In support of this view, he referred to the fact that a change in valuation technique is regarded as a change in an accounting estimate: see IFRS 13, "Fair Value Measurement", para 66. Mr Watson-Brown considered the change made by Investec to be a change in its technique of valuing the liability represented by the Funding Gap Reserve. By contrast, a change in "measurement basis" is a change in accounting policy: see IAS 8, para 35. Examples of a "measurement basis" are "historical costs, current costs, net realisable value, fair value or recoverable amount": see IAS 1, para 118. In Mr Watson-Brown's view, Investec was at all relevant times trying to establish the fair value of the Funding Gap Reserve and there was no change in measurement basis. He also took into account the guidance in paragraph 35 of IAS 8 that:

"When it is difficult to distinguish a change in accounting policy from a change in accounting estimate, the change is treated as a change in accounting estimate."

Mr Watson-Brown made the point that if past accounts had to be restated every time there was a change in valuation technique, companies would have to be restating their accounts all the time, which would be impractical.

140. Mr Brice disagreed with Mr Watson-Brown's opinion. In Mr Brice's view, the effect of the change introduced in July 2010 was to recognise an item twice in a year, once on a cash basis and once on an accruals basis. The change thus amounted to a change in the basis of recognising a liability, which falls within the definition of a change in accounting policy. Mr Brice observed, however, that "there is an element of grey in this area" and that Mr Watson-Brown had grounded his opinion in accounting standards. He agreed that a reasonable accountant could form the view that the change was a change in accounting estimate.
141. On this evidence I find it impossible to say that the accounting treatment of profit payaways adopted by Investec was irrational in the sense of being one that no reasonable accountant could adopt. It follows that Investec was not in breach of contract in adopting that treatment in the bonus calculation for 2010/2011.

The 'early bird' issue

142. The final issue concerns the proper accounting treatment of 'early-bird' deposits.
143. The claimants' case is that they ought to have been, but were not, credited in the accounts for their desk with interest from the Central Treasury on such deposits over the period from when the money was received until the term of the product began. At the same time they say, and it is common ground, that the interest paid to the customer during this period was recognised as a cost to the SED desk.
144. It seems to me that it would be difficult rationally to justify charging the cost of such funds to the SED desk while at the same time ascribing no benefit to the receipt of the funds. The argument that it was appropriate to ascribe no benefit to the funds because the Bank would have returned the money to the customer if the customer had decided not to proceed with the investment seems to me to be a bad one, not least because a 'cooling off' period applied to all retail structured products investments but was never otherwise, as I understand it, taken into account.
145. It was Mr McKenna's evidence, however, that as a matter of fact 'early bird' funds were deposited into the SED desk's current account, thus reducing the desk's 'overdraft' with the Central Treasury and avoiding interest that would otherwise have been charged to the desk. Although this effect would not exactly have matched interest credited at the rate quoted for the product by the Central Treasury, it is not suggested the difference would have been material.
146. Mr Brogden's evidence was that he did not believe that there was such a set-off although he did not know for sure what the position was. I am unable to attach any weight to a theoretical calculation included in the claimants' written closing submissions which sought to draw from various figures an inference that early bird deposits were not used to reduce the desk's overdraft. The calculation was not put in evidence, the Bank had no opportunity to comment on it and I am not in a position to judge whether the inference has any reasonable basis or not.
147. The claimants point out that Investec has adduced no documents which confirm the position one way or the other. However, it is also the case that the claimants have not sought disclosure of such documents which might have enabled them to prove their case.
148. The evidence on this point is unsatisfactory. However, it is the claimants who bear the burden of proof, and in the result I find that they have failed to prove that they were not given credit for early bird deposits.

H. CONCLUSION

149. Despite the conclusions I have reached in this judgment about the merits of their claim, I was impressed by the claimants when they gave evidence. They both struck me as decent and highly talented individuals. It is clear that they have in the past been very successful in their line of business and I do not doubt that they will succeed again. It is also clear that they have a strong sense of grievance about how they were treated by Investec and sincerely believe that the Bank's refusal to recognise any entitlement to bonus for the 2010/2011 year was unfair. What is fair in this context is dependent on perspective. Mr Brogden and Mr Reid believe that they developed a retail structured product business for Investec which, at least in a broad sense, generated economic value for the Bank of which they should be given a share. If the lens is broadened further, I doubt there are many outside the world in which the claimants operate who would think that they were under-rewarded by Investec. The task of the court is not to adopt either of those perspectives but to judge what is fair simply in terms of adherence to contract. Judged by that standard, I conclude that the claimants had no right to be paid any bonus for the 2010/2011 year and the claim therefore fails.

Note 1    It is unnecessary in this case to consider the qualifications to the objective test discussed in that paragraph.     [Back]

Note 2    In fact, it does not appear to correspond to how the term “EVA” is used by Stern Stewart, who define EVA as net operating profit after taxes minus the cost of capital.     [Back]

Note 3    See R Dworkin, ‘Taking Rights Seriously’ (Harvard, 1977), p.31.    [Back]

Note 4    I refused permission to make other amendments raising a number of other additional issues.    [Back]