ING Bank N.V. v Ros Roca S.A.

The defendant (“Ros Roca”) is incorporated under the laws of Spain, and is involved in the waste and environmental services sector. In June 2006 it was introduced to what came to be called “Project Bird”. The aim of Project Bird was to negotiate the purchase by Ros Roca of a British company, Dennis Eagle Group Ltd (“DEG”) which was involved in the same sector, particularly in the manufacture of refuse collection vehicles. As I shall explain in the next paragraph, the claimant (“ING”) is a bank incorporated under the laws of the Netherlands. It acted as Ros Roca’s financial adviser for the purposes of Project Bird. As Project Bird progressed an arrangement was made under which ING would, in addition to its advisory role, itself provide financial assistance by underwriting a bridge facility of up to €63,500,000 (the “Bridging Facility”) to cover part of the financing of Project Bird. It was a term of the Bridging Facility that the funds provided be repaid through a capital increase (the “Transaction”). The additional capital was to be issued by Ros Roca, and was to be the subject of an investment by a third party.

A separate agreement was made between Ros Roca and ING under which Ros Roca engaged ING as its exclusive financial adviser in connection with the search for an investor, loosely described as a “partner”, for the Transaction. A letter addressed to Ros Roca set out terms on which the “Corporate Finance Division of ING Bank N.V., London Branch”, described in the letter as “ING Corporate Finance”, would act for Ros Roca in this regard. The letter was on notepaper which stated that ING Bank N.V. was registered in England and regulated by the Financial Services Authority for the conduct of investment business in the UK. These circumstances may explain why the claim form identified the claimant as “ING Bank N.V., London Branch.” However, the notepaper also made it clear that ING Bank N.V. was incorporated with limited liability in the Netherlands. It is common ground that the claim is not brought by a distinct entity in the form of ING Bank N.V.’s London Branch but by ING Bank N.V. itself.

Although the letter was dated 31 October 2006 it appears to have been countersigned by Ros Roca on 30 October 2006. The letter was headed “Project Bird II”, which was the name initially given to the project to bring about the Transaction. However this was later replaced by the name “Project Hawk”. Accordingly I shall refer to the agreement set out in the letter as “the Hawk Retainer.”

The Hawk Retainer stated that, in part, ING’s fees would depend on a formula, which I shall call the “Entry Ratio”, expressed in this way:

the Enterprise Value/EBITDA 2006 (“EV/EBITDA 06”) entry multiple implicit in the Transaction.

Project Hawk was successful. The eventual investor was Deyà Capital SCR, S.A. (“Deyà”). Pursuant to an investment agreement dated 7 December 2007 (“the Investment Agreement”) it made an equity investment of €63,500,000. In the event, the equity investment was made in an entity called Ros Roca Environment, S.L. Nothing turns on this for present purposes. The Transaction was thus achieved, and Ros Roca was enabled to repay the Bridging Facility. However a dispute arises as to the fee payable to ING under the Hawk Retainer. The dispute has two elements.

The first element involves a construction issue. It concerns the true meaning of the words used in expressing the Entry Ratio. There is no dispute as to the meaning of the numerator: “Enterprise Value” is defined in the Hawk Retainer as “the pre-money valuation of the partner’s economic offer for its equity investment, plus any debt outstanding in Ros Roca before completion.” It is common ground that “the pre-money value” refers to the valuation of the whole of Ros Roca prior to this investment, that “plus any debt outstanding” means that such debt is to be included in what is valued, and that in the result the Enterprise Value amounts to €441 million. The higher the Enterprise Value express or implicit in a bid by a potential investor, the more beneficial the Transaction would be to the existing shareholders of Ros Roca. They were prepared to dilute their shareholding in order to secure an investment of €63.5 million, but would want the proposed investor to ascribe a high value to the business so that the number of new shares issued in exchange for the investor’s €63.5 million, and the consequent dilution of their existing shareholding, would be as low as possible.

As to the denominator, it is common ground that “EBITDA” refers to earnings before interest, tax, depreciation and amortisation and reflects the underlying profitability of the business. It is also common ground that a similar ratio, broadly described as “the EV/EBITDA multiple” is, as was stated by Mr Muro-Lara, Managing Director of ING’s Spanish subsidiary, a measure widely used in financial markets to value companies and as a means of comparing whether the price at which a company’s securities are valued is cheap compared with their peers. Although Mr Muro-Lara did not say so expressly, there can be no doubt that this is because the higher the EV/EBITDA multiple, the higher the capital value which the purchaser is prepared to ascribe to the Company in comparison with what is known about its underlying profits.

If the Transaction had been concluded at a time when the “current” EBITDA for Ros Roca was an EBITDA for 2006, the words used in expressing the Entry Ratio would have caused no problem. It is common ground that in those circumstances they would require a calculation of the denominator by reference to the combined EBITDA for 2006 of both Ros Roca and DEG. This would result in a denominator of €33.1 million, and an Entry Ratio of 13.3. The problem arises because the Transaction occurred at a time when the EBITDA for 2006 was no longer current. Ros Roca says that the current EBITDA at the time of the Transaction was a forecast figure of €42.6 million for 2007 which had been circulated during the second half of 2007. On this basis the Entry Ratio would be 10.35. ING’s case is that the words “implicit in the Transaction” refer only to the numerator of the Entry Ratio, while the denominator must in all cases be an EBITDA for 2006, even if an EBITDA for some other period were thought to be “implicit in the Transaction”. The resultant remuneration for ING is very much greater than that arising if the denominator were the forecast EBITDA for 2007 of €42.6 million circulated during the second half of 2007 and said by Ros Roca to have been “current” at the time of the Transaction. Nevertheless, submits ING, the deliberate use of “EBITDA 2006” shows that the parties intended a formula which worked by reference to the EBITDA for the specific year 2006 only. This means that, when calculating the Entry Ratio, the words “implicit in the Transaction” merely require that the Enterprise Value implicit in the Transaction be identified, and that this then be divided by a known figure for EBITDA 2006. By contrast, Ros Roca asserts that the true agreement between the parties was that an entry multiple should be calculated in the normal way, namely by dividing the Enterprise Value implicit in the Transaction by the EBITDA current at the time of completion of the Transaction.

The dispute as to ING’s fee emerged after ING sent Ros Roca an invoice dated 6 May 2008 in the sum of €7,350,749.05. The invoice comprised three items. The first and third were a fixed fee of €635,000 and expenses of €15,749.05. The second item was an additional fee of €6,700,000, based on an entry ratio of 13.3. On 28 November 2008, after this action was commenced, Ros Roca paid €1,594,671 on the basis that the fixed fee and expenses were not in dispute, but that the correct figure for the additional fee was €943,922.44, based on an entry ratio of 10.35.

ING now seeks a declaration that its construction of the Hawk Retainer is correct, and a judgment in its favour for the difference between the amount paid by Ros Roca and the amount which would be payable under that construction. ING’s claim will fail if Ros Roca’s construction is found to be correct. If the court concludes that ING’s construction is correct, however, that would not be an end of the matter. Ros Roca advances a fallback argument. This is the second element in the dispute. It involves an assertion that the doctrine of estoppel assists Ros Roca in the present case and bars ING from relying upon the construction it advances. This second element had the result that I heard substantial witness evidence on the estoppel defence. Much of that evidence is inadmissible on the question of construction.

When bringing these proceedings ING adopted the procedure under Part 8 of the Civil Procedure Rules. This is an alternative to the normal procedure which requires precise statements of case from each side and focuses attention on issues of fact as well as law. The Part 8 procedure will often be appropriate where the only issue is as to construction. In this case, however, it soon became apparent that significant issues of fact arose. In response to the claim Ros Roca did not confine itself to countering ING’s suggested construction, but also advanced two fallback arguments to cater for the possibility that ING’s construction might find favour with the court. The first fallback argument was rectification. This was abandoned shortly before trial. The second fallback argument asserted that ING was estopped from relying on its suggested construction, which is now the second element in the dispute as described above. Both these fallback arguments were supported by lengthy witness statements and would plainly require resolution of a substantial number of issues of fact. Nevertheless the claim continued to proceed under the Part 8 procedure. This was less than satisfactory, for that procedure is not designed for cases where substantial factual issues arise.

The witnesses on both sides were Spanish, and for the most part gave evidence through an interpreter. I am sure that each of the witnesses was doing his best to assist me. In that regard I particularly mention Mr Vilagrasa, Executive Vice President of the Ros Roca Group. His evidence under cross-examination was the subject of an attack by ING as having been, in relation to the expected timeline, “evasive and incredible.” For reasons given below I reject that attack. In the context of the estoppel defence Mr Vilagrasa during cross-examination gave evidence of fluctuating estimates of transaction costs at one stage of €4 million, then €3 million, and then returning to €4 million. He recalled that he did not remember exactly, but he had spoken, he thought to Mr Muro-Lara, “carefully applying the development of the costs” and agreed it would be good to maintain the €4 million figure. ING’s written closing in this regard accused Mr Vilagrasa of giving untruthful evidence to the court. I am quite sure that Mr Vilagrasa was not seeking to deceive. He made it plain that he did not have an exact recollection. His evidence, in so far as it concerned assumptions shared by the parties, did not involve any assertion going significantly beyond what is common ground as to such assumptions. The answers apparently relied upon in order to accuse Mr Vilagrasa of untruthfulness in my view reflect, at worst, a muddle during cross-examination.

In this judgment I first examine the construction issue. For that purpose I deal with relevant legal principles before turning to additional background facts, the arguments of the parties, and my analysis. In the second main part of this judgment I deal with the assertions made by Ros Roca in support of its estoppel argument.

Until recently the leading case on construction of contracts was Investors Compensation Scheme v West Bromwich Building Society (No. 1) [1998] 1 WLR 896. At p 913C Lord Hoffmann said:

the meaning of the document is what the parties using those words against the relevant background would reasonably have been understood to mean. The background may not merely enable the reasonable man to choose between the possible meanings of words which are ambiguous but even (as occasionally happens in ordinary life) to conclude that the parties must, for whatever reason, have used the wrong words or syntax

That case and others have now been reviewed by the House of Lords in Chartbrook Ltd v Persimmon Homes Ltd [2009] UKHL 38. The principal speeches were delivered by Lord Hoffmann and by Lord Walker of Gestingthorpe, who agreed with Lord Hoffmann. Lord Hope, Lord Rodger and Baroness Hale also agreed with Lord Hoffmann. At paragraphs [21] and [25] Lord Hoffmann said this:

[21] When the language used in an instrument gives rise to difficulties of construction, the process of interpretation does not require one to formulate some alternative form of words which approximates as closely as possible to that of the parties. It is to decide what a reasonable person would have understood the parties to have meant by using the language which they did. The fact that the court might have to express that meaning in language quite different from that used by the parties (“12th January” instead of “13th January” in Mannai Investment Co Ltd v Eagle Star Life Assurance Co Ltd [1997] AC 749 ; “any claim sounding in rescission (whether for undue influence or otherwise)” instead of “any claim (whether sounding in rescission for undue influence or otherwise)” in Investors Compensation Scheme Ltd v West Bromwich Building Society [1998] 1 WLR 896) is no reason for not giving effect to what they appear to have meant.

…

[25] What is clear from these cases is that there is not, so to speak, a limit to the amount of red ink or verbal rearrangement or correction which the court is allowed. All that is required is that it should be clear that something has gone wrong with the language and that it should be clear what a reasonable person would have understood the parties to have meant.

At paragraph [42] Lord Hoffmann affirmed the rule that the court may not have regard to evidence of what was said or done during the course of negotiating a contract for the purpose of drawing inferences about what the contract meant. He added that the rule does not exclude the use of such evidence for other purposes, for example to establish that a fact which may be relevant as background was known to both parties. Earlier, in paragraph [39], Lord Hoffmann stressed that English law does not ask what the parties’ intentions actually were but rather what a reasonable observer would have taken them to be.

Mr Stanley for ING added that under English law the conduct of the parties subsequent to a written agreement is not admissible as evidence to assist its interpretation: James Miller & Partners Ltd v Whitworth Street Estates (Manchester) Ltd [1970] AC 572 at 603C–E (Lord Reid). Mr Colton for Ros Roca suggested that this rule, like the rule about negotiations, does not exclude the use of such evidence to establish that a fact which may be relevant as background was known to both parties. In principle I think Mr Colton’s submission is likely to be correct. As I have not needed to examine the evidence about subsequent conduct in this regard I express no concluded view.

For the purposes of this section I confine my examination to an identification of additional matters which existed prior to or at the time that the parties entered into the Hawk Retainer.

Section 1 of the Hawk Retainer envisaged that ING would, if appropriate and required, provide services to Ros Roca (also referred to as “the Company”) as follows:

a)

using information provided by the Company for the purposes of the Engagement, provide the Company with a valuation of the Ros Roca’s assets that form part of the transaction;

b)

advising on the best long term financial structure for Ros Roca;

c)

preparation, with the assistance of the management of the Company, of the information which will be made available to the potential buyers, including the elaboration of a descriptive sale memorandum, with detailed information of the businesses and its economic financial situation.

d)

in consultation with the Company, developing, updating and reviewing a list of potential purchasers (the “List”) and contacting those in the List which have been approved by the Company;

e)

advising the Company on the conduct of the Transaction, including advising on obtaining confidential undertakings from potential purchasers in respect of confidential information, dealing with enquiries from potential purchasers, accompanying potential purchasers as required on due diligence meetings with the Company and management and site visits, and distributing further information;

f)

advising and assisting in the negotiations the Company may hold with potential purchasers or any other party in the Transaction and its advisers and/or investors and, if appropriate, the advice on tactics which the Company may wish to adopt in relation to such negotiations;

g)

assisting the Company on the final terms of the Transaction;

h)

collaboration and co-ordination of the Company’s other advisers, which will prepare the economic, financial, administrative, technical, tax and legal information (Vendor Due Diligence and Data room) to be delivered to the potential interested parties; and

i)

co-ordination of, and assistance with the preparation of any documentation required to execute the Transaction.

Section 2 of the Hawk Retainer dealt with fees and expenses. It provided for payment of fees “upon the successful completion of the Transaction”, and for payment in addition of “all costs and expenses which ING… may properly incur…”. The fees payable were of 2 types. First, section 2(a) provided for a fixed fee. This was (save as regards a contingency which did not in fact occur):

… 1% over [i.e. of] the higher of the equity bridge facility or the equity investment in Ros Roca, …

Section 2(b) concerned an additional fee payable on top of the fixed fee. It was in these terms:

b)

an additional Fee based on the Enterprise Value/EBITDA 2006 (“EV/EBITDA 06”) entry multiple implicit in the Transaction.

For an implicit EV/EBITDA 06 multiple in the following range	An additional fee per 0.1 multiple of
Below 8.9	EUR 0
Larger than 8.9 and below/equal to 9.2	EUR 25,000
Larger than 9.2 and below/equal to 9.5	EUR 50,000
Larger than 9.5 and below/equal to 10.0	EUR 75,000
Larger than 10.0 and below/equal to 10.5	EUR 100,000
In excess of 10.5	EUR 200,000

In this letter of agreement, the term “Enterprise Value” means the pre-money valuation of the partner’s economic offer for its equity investment, plus any debt outstanding in Ros Roca before completion.

Illustrative example: in case of total equity raising from a financial partner of EUR 60m at an entry EV/EBITDA 06 multiple of 9.5, proceeds for ING would amount to a total of EUR 825,000 (fixed fee of EUR 600,000 plus additional fee of EUR 75,000 + EUR 150,000).

The Hawk Retainer included an appendix headed “Terms and Conditions”. These included an entire agreement clause. However neither side attached any substantial significance to the appendix.

In addition to matters identified earlier it is possible to identify certain additional features of the surrounding circumstances which existed at the time of the Hawk Retainer and were known to both parties. In particular:

ING sought to rely on other matters which Ros Roca disputed. I shall take them in turn.

ING advanced a contention that it was not reasonably in contemplation that any subscription in Ros Roca would be concluded in 2006 or early 2007. As to this:

ING asserted that its contention was in part obvious common sense given what it was known had to be done. It also asserted that Mr Vilagrasa’s evidence under cross-examination was “evasive and incredible.” I do not agree with either of these assertions. It was not disputed that at the time that the Hawk Retainer was drafted it was contemplated that the Bridging Facility would become operational in early November. Absent the fulfilment of stringent conditions, the Bridging Facility would therefore become due and payable in early May 2007. This, along with the imminent achievement of Project Bird (the acquisition of DEG) within a timescale of 6 months, supports Mr Vilagrasa’s evidence of a timeline envisaging completion of the Transaction by May 2007. ING’s witnesses did not give evidence to the contrary. ING relied on the Bridging Facility, but the relevant passages are no more than contingency provisions identifying ways in which there might be extensions through to June 2008. It also relied on a sentence in Mr Goma’s witness statement, where having described relevant parts of the Bridging Facility he acknowledged that it was envisaged that the Transaction would take place during 2007 and in any event no later than June 2008. Neither of those matters makes good a contention that at the time that the Hawk Retainer was drafted it was not “reasonably in contemplation” that any subscription in Ros Roca would be concluded by May 2007. In the event matters arose which delayed the subscription until the end of 2007. At the time that the Hawk Retainer was drafted, however, I conclude that it was at least a realistic possibility that the Transaction would be concluded by May 2007.

ING asserted that it was not abnormal, or even unusual, to make use of an EV/EBITDA multiple in which the EV date and the EBITDA period were different. In my view the witness evidence did not go this far. Witnesses accepted that it could be done, and agreed that words used in contracts should be clear. That evidence is hardly surprising. As indicated at paragraph 23(7) above, the witness evidence also was that when valuing a company, or comparing valuations, the ‘current’ figures for Enterprise Value and EBITDA would most commonly be used – and in the case of EBITDA the ‘current’ figure might well be a forecast. ING did not contend otherwise.

ING contrasted what the parties could assume at the date of the Hawk Retainer about the numerator and denominator of the Entry Ratio. The feature identified at paragraph 23(3) above (a bid to invest might explicitly identify an Enterprise Value in the sense used in the Hawk Retainer; if it did not then it would always be possible to extrapolate that value) meant, said ING, that it was certain that any offer would state or imply an ‘Enterprise Value’ as defined. By contrast, said ING, it was not necessary that any offer should state or imply any assumption on the part of the offeror about EBITDA or any EV/EBITDA multiple. ING claimed this was in any event obviously true, and that it was consistent with ING’s case that Enterprise Value had been defined in the Hawk Retainer while EBITDA had not. In response Ros Roca accepted that an offer might not state any connection between the Enterprise Value and EBITDA. However, Ros Roca added the comment that whether the EV/EBITDA entry multiple was used to calculate the Enterprise Value offered, or was used merely as a cross-check following calculation by some other means (eg, using discounted cash flow), the parties were both aware that EV/EBITDA comparables were commonly used to check whether a proposed offer is appropriate.

In my view Ros Roca’s comment is sound. While I do not have the benefit of expert evidence, the matters set out at paragraph 23(6) and 23(7) above demonstrate the use of EV/EBITDA comparables in the market. They reflect the acceptance by ING’s witnesses that valuations for actual transactions would use the figures for enterprise value and EBITDA current at the time of the transaction – which in the case of EBITDA might be a forecast.

ING’s skeleton argument prior to trial, in asserting that the words used to express the Entry Ratio had the meaning advanced by ING, submitted that the words were clear, and that there was neither vagueness nor ambiguity. By the end of the hearing ING put the positive basis for its asserted meaning somewhat differently, submitting that “the natural way to read the [words used] is … as the Enterprise Value (as defined) divided by the EBITDA 2006 …” (emphasis added). Ros Roca’s construction was said, by contrast, to be improbable, unnecessary and commercially problematic.

The main argument on behalf of Ros Roca was that the Hawk Retainer should be construed to refer to the “EV/EBITDA entry multiple implicit in the Transaction”, as this gave effect to the commercial intention of the clause. It would reward ING for its performance in securing a high (current) EV by reference to a (current) EBITDA, comparing like with like, whereas ING’s construction would reward ING not for its negotiating skills in achieving a beneficial EV/EBITDA multiple, but rather for achieving a ratio between two numbers which may have no connection at all. Indeed it might reward ING merely for delaying, as using an outdated EBITDA would reward ING for increases in enterprise value derived from improved company performance, even though ING may have had no involvement in such improvement.

Ros Roca added that ING’s construction in effect deleted the phrase “entry multiple implicit in the Transaction”. No entry multiple involving EBITDA for 2006 was implicit in the Transaction.

ING had criticised Ros Roca’s construction first on the basis that it was improbable that the parties would have used “EBITDA 2006” when they simply meant “current EBITDA”. In the context of commercial transactions this was not particularly improbable. In any event, resort to probabilities was inappropriate.

The second objection on the part of ING was that Ros Roca’s construction was “unnecessary” because “there is nothing inherently contradictory in an entry multiple being calculated by taking the ratio of the enterprise value and a historical EBITDA figure”. As to that, Ros Roca answered that it was simply wrong. A ratio of enterprise value to a historical EBITDA figure could be calculated mathematically, but that was not an “entry multiple”. Common sense and industry practice both required that what was implicit in a transaction was a comparison of current EV with current EBITDA. The parties could have simply said that the additional fee would be based on the ratio of EV to EBITDA 2006 but the words they actually used were that it was based on the “EV/EBITDA 2006 entry multiple implicit in the Transaction.”

It was said by ING that reference to current EBITDA would be commercially problematic, because an investor might not in fact calculate EV by applying a multiple to EBITDA. Ros Roca’s answer was that it was for this reason that the Hawk Retainer expressly referred to a multiple “implicit” in the Transaction. As to a suggestion that the current EBITDA at any time was “relatively ill defined”, there was no difficulty in practice, answered Ros Roca. The Transaction could be expected to make reference to EBITDA current at the time.

Finally Ros Roca relied upon the statement by Lord Reid in E.L. Schuler AG v Wickman Machine Tool Sales Limited [1974] AC 235:

The more unreasonable the result the more unlikely it is that the parties can have intended it and if they do intend it the more necessary it is that they shall make that intention abundantly clear.

At a relatively early stage during oral evidence I made comments to the effect that ordinarily when making a comparison one compared like with like. On this basis I questioned whether, if the denominator of the Entry Ratio was indeed an EBITDA for operations in 2006, then one might expect that the numerator would be an Enterprise Value of the same operations in 2006. I made this comment in case it might affect the questions put to witnesses. As I shall explain below, neither party considered that this was a tenable construction of the words used in the Hawk Retainer.

The words used to express the Entry Ratio were not clear, and ING was right to depart from the stance initially taken in this regard in its skeleton argument. Simply as a matter of language there were numerous potential ambiguities. Some of them overlap. Potential questions include the following:

As mentioned earlier, during oral evidence I made comments to the effect that ordinarily when making a comparison one compared like with like, and I questioned whether, if the denominator of the Entry Ratio was indeed an EBITDA for operations in 2006, then one might expect that the numerator would be an Enterprise Value of the same operations in 2006. Any construction of the Hawk Retainer to that effect was rejected by ING on the grounds that it conflicted with the express definition of Enterprise Value, and that there was no expert evidence that one could work back from an Enterprise Value for 2007 to compute an equivalent value for 2006. Ros Roca also drew attention to difficulties in such an exercise.

In these circumstances I apply the principles in Lord Hoffmann’s speech in Chartbrook, especially at paragraphs [21] and [25]. I have no hesitation in concluding that the language used gives rise to difficulties of construction. Is it clear that something has gone wrong with the language, and if so is it clear what a reasonable person would have understood the parties to have meant?

ING asserts that nothing has gone wrong with the language. Its contentions identified above are all, it says, linguistically consistent with the words used – “2006” was a specification of EBITDA which made the meaning of “entry multiple” clear. I do not agree, for I doubt that it is possible linguistically to attribute the words “implicit in the Transaction” only to the numerator described earlier in that sentence and not to the denominator which appears in between those words and the numerator. Assuming, however, that it is linguistically possible to arrive at ING’s construction, that does not of itself establish that nothing has gone wrong with the language.

Mr Stanley in his closing oral submissions accepted that, at least on one view of the matter, on ING’s case the parties had used the term “entry multiple” idiosyncratically. I do not consider that ING’s construction of the words used to express the Entry Ratio is merely idiosyncratic. In my view it is so unreasonable in its result that the parties cannot have intended it – or if they had intended it they would have taken steps to make that intention abundantly clear.

The key point here is not only that valuations generally use current values for both numerator and denominator when computing an entry multiple, but also that this is done for good reason. Only by using the EBITDA which is “current” for the actual or proposed purchase in question can one measure the extent to which the purchase price constitutes a high or low assessment of the company’s intrinsic worth by reference to its earnings. The obvious purpose of an additional fee is to give ING a success fee over and above its fixed fee to reflect the extent to which the eventual purchaser has made a high rather than a low assessment of Ros Roca’s intrinsic worth. On the assumption that EBITDA 2006 would be the current EBITDA at the time of the Transaction this was exactly what the words used to express the Entry Ratio achieved.

The Hawk Retainer was drafted at a time when, to put it no higher, there was a realistic possibility that EBITDA 2006 would be the current EBITDA at the time of the Transaction. It is understandable that at such a time the parties used the words “Enterprise Value/ EBITDA 2006 entry multiple implicit in the Transaction”, and did not pick up on the need to allow for the possibility that some other EBITDA would be current at the time of the Transaction. While one may say that in general it is improbable that parties would inappropriately specify a particular year, in the present case I do not find it particularly improbable that ING drafted a letter referring to 2006 and neither party spotted that this overlooked the possibility that 2006 might no longer be apposite at the time of the Transaction.

Moreover I find it impossible to detect any rational basis upon which the parties could have concluded that upon the premise that EBITDA 2006 might no longer be current at the time of the Transaction it should nevertheless be used for the purpose of calculating the additional fee. Ros Roca went so far as to say that it would give ING a reward for delay. It seems to me that that would not necessarily result – it may be said to occur on the facts in this case, but it might not occur on other facts. Accordingly I do not found my decision on the undoubted fact that the result of applying ING’s construction is to arrive at an overall fee which is high. The crucial points here are that on this premise one will not be comparing like with like, and there is no obvious relationship between a denominator of EBITDA 2006 and a numerator achieved by calculating the Enterprise Value for a Transaction at a time when EBITDA 2006 is no longer current.

ING at one stage suggested that the parties – or at least ING - might wish to stick with EBITDA 2006 so as to avoid the risk that things would deteriorate in 2007. In my view this is fanciful. If the parties had any expectation about 2007 it was that the merger would affect growth in earnings positively rather than adversely. More fundamentally, the suggestion attributes to the parties a desire to depart from the normal principle that a success fee will reflect the worth of what has been achieved - with no countervailing logic to explain why either of the parties would consider it desirable to have a complete about-turn at the stage when EBITDA 2006 ceased to be current.

ING then suggested that the expected growth in earnings could have been allowed for in the Hawk Retainer. This, too, appears to me to be fanciful. It ignores the fundamental point noted in the last paragraph. Some reliance was placed on the stages at which ING’s additional fee would be ratcheted up under the Hawk Retainer as demonstrating that such an allowance was made, but they are entirely consistent with the background facts as to entry multiples on current values for other transactions in the sector.

Accordingly I conclude that ING’s construction is a commercial nonsense, and I accept Ros Roca’s contention that an alternative construction is necessary. That leaves ING’s contention that Ros Roca’s proposed construction is commercially problematic. Here ING contrasts EBITDA 2006 which it says would be a known quantity by the time any calculation came to be carried out, with ‘current’ EBITDA which ING describes as ‘relatively ill-defined’. I am prepared to assume that EBITDA 2006 would be likely to be a known quantity by the time any calculation came to be carried out, and that this known quantity would be likely to be what the parties had in mind for a transaction at a time when the current EBITDA was an EBITDA for 2006. Even so it does not seem to me that using a later EBITDA for a transaction at a time when the current EBITDA was no longer that for 2006 involves uncertainties so significant as to create an insurmountable obstacle. ING commented that the investor may not have used EBITDA to compute the amount of its offer – and indeed was not required to use any particular methodology, let alone disclose it. That, however, does not prevent the parties from identifying the EBITDA current at the time of the Transaction. It is common ground that when valuing or comparing values it is general practice to identify such an EBITDA, which may be a forecast rather than an actual figure. If this can be done as a matter of general practice then there is no reason to conclude that it will involve insuperable commercial problems in the context of the Hawk Retainer.

For all these reasons I consider it clear that a reasonable person would have understood the parties, when using the words they did in expressing the Entry Ratio, to have included references to 2006 by oversight, and to have intended that the denominator should be EBITDA without linking this to a specified year.

My conclusion on construction makes it unnecessary to turn to Ros Roca’s case on estoppel, for this is an alternative defence which would only arise if Ros Roca’s case on construction failed. I shall nevertheless deal with it briefly lest the matter be taken further.

Ros Roca’s skeleton argument for the trial described its case on estoppel as being that “ING is estopped from claiming any Additional Fee in excess of that calculated by reference to the EV / EBITDA 2007 multiple (10.35x).” The skeleton argument identified four elements required for an estoppel by convention, and described them in this way:

(1)

a shared assumption of a matter of fact or law;

(2)

which is communicated between the parties by some mutually manifest conduct;

(3)

which is relied or acted upon by the party raising the estoppel;

(4)

as a result of which it would be unfair, unjust or unconscionable to permit the estopped party from resiling from such shared assumption.

Ros Roca’s skeleton argument went on to specify four “shared assumptions between the parties”, which I shall call “the opening assumptions”, as follows:

(1)

the total costs for Project Hawk would be in the region of €4 million; and

(2)

the total fees due to ING (including the fixed fee of €635,000) would therefore be in the region of €2 million; and

(3)

the EV/EBITDA entry multiple implicit in the Transaction was EV / EBITDA07; and

(4)

accordingly,

(a)

EV / EBITDA06 would not be used in order to calculate the Additional Fee (for this would have given an Additional Fee alone of €6.7 million); and

(b)

the Additional Fee would be calculated by reference to EV / EBITDA07.

Ros Roca’s written closing set out a case on estoppel similarly relying only on estoppel by convention, and accepting that for this purpose Ros Roca must found its case upon a shared assumption objectively manifested by the parties. However in place of the four opening assumptions it identified two shared assumptions only, which I shall refer to as closing assumption 1 and closing assumption 2. They are said to have been:

1.

Ros Roca’s total transaction costs for Project Hawk, for ING and for its other advisers, would be in the region of €4 million; and

2.

this €4 million transaction costs figure was calculated on the basis of an Additional Fee calculated using an “EV/EBITDA... entry multiple implicit in the Transaction” of 10.35x.

In support of closing assumption 1 Ros Roca relied on paragraph 25(b) of ING’s written closing. Paragraph 25(a) needs to be read as a prelude to paragraph 25(b):

25.

It is respectfully submitted that the material facts are as follows:

(a)

In the period from late September 2007 up to and including December 2007, the parties shared an assumption (sufficiently evidenced by mutual communications between them) that a reasonable estimate of total transaction costs for the purposes of preparing the estimated net debt required by an Annex to the agreement with Deyá was in the region of €4 million: see the correspondence concerning the transaction costs.

(b)

That estimate was originally prepared (in September) by Mr Gomà for the purposes of a management presentation by Ros Roca to potential investors; it was thereafter adopted, without further express discussion, as a reasonable figure to use in the net debt estimate included in Annex 6.8.1. It appears (alongside other figures) as one item in a number of complex spreadsheets dealing with the net debt—but never, so far as can be seen, as the focus of any specific attention.

Paragraph 25(a) and (b) concern an assumption of fact, namely that a reasonable estimate of total transaction costs for the purposes of preparing the estimated net debt required by an Annex to the agreement with Deyá was in the region of €4 million. By contrast closing assumption 1 was an assumption as to the future: that Ros Roca’s total transaction costs for Project Hawk, for ING and for its other advisers, would be in the region of €4 million. Such an assumption is not an assumption of fact or law and thus does not fall within either limb of Ros Roca’s own formulation of the first of the four elements required for an estoppel by convention.

When this was pointed out Mr Colton in his oral closing suggested that ING’s case could be reformulated as a promissory estoppel. I doubt whether it would be just to allow this to be done. Moreover there is simply no basis for identifying any express or implicit promise in what passed between the parties.

Mr Colton then advanced a contention that closing assumption 1 contained within it an assumption, which I shall call “closing assumption 1A”, to the effect that ING could not claim fees from Ros Roca which would cause Ros Roca’s total transaction costs materially to exceed the 4 million Euro figure. The contention is without foundation. There was no evidence that the parties made any such assumption. As to the suggestion that closing assumption 1A could be found “within” closing assumption 1, there is a vast difference between an assumption that a particular estimate was reasonable and an assumption that ING could not claim a fee which would result in the estimate being breached. The latter goes far beyond the former.

Moreover closing assumption 1, based as it is upon ING’s paragraph 25(b), would not – even if it could form a legal basis for an estoppel by convention - get Ros Roca anywhere. ING’s claim in these proceedings would not be barred by that assumption. It is not necessary for ING to assert that the estimate in the region of €4 million was unreasonable, and ING makes no such assertion.

That leaves closing assumption 2. The only support for this assumption in Ros Roca’s written closing was an assertion that it flowed from closing assumption 1 by a process of reasoning involving two stages, which I shall call “stage 2(1)” and “stage 2(2)”:

2(1) An independent and objective observer, considering the communications passing between the parties in November and December 2007, would note:

(i)

The Enterprise Value implicit in the Transaction was known to be €441 million.

(ii)

ING had itself decided to insert the figure of €4 million as transaction costs … .

(iii)

ING had, separately, described a multiple of 10.35x as the ‘entry multiple’ for the Transaction, calculated by dividing €441 million by an EBITDA 07 of €42.6 million, and treated this as the relevant multiple for assessing the Transaction. This was consistent with ING’s and Ros Roca’s approach to EV/EBITDA multiples prior to the EV being fixed, too: … .

2(2) From these facts, an independent and objective observer, knowing of the Additional Fee clause in the Engagement Letter, could only realistically have concluded that the Additional Fee was to be calculated by reference to the EV/EBITDA entry multiple implicit in the Transaction, and not by reference to EV/EBITDA 2006.

Ros Roca is wholly unable to point to any part of the history which involved an express assertion by either party that the Entry Ratio in the Hawk Retainer was 10.35 or was to be computed by reference to anything other than EBITDA 2006. It is no doubt for this reason that Ros Roca’s written closing sought to rely on what it said an independent and objective observer, knowing of the Additional Fee clause in the Hawk Retainer, “could only realistically have concluded”.

The insuperable problem for Ros Roca in this regard is that the conclusion does not flow from the premise. At stage 2(1) Ros Roca identifies three things for an observer to note. The first, that the Enterprise Value implicit in the Transaction was known to be €441 million, merely means that an individual wanting to compute the Entry Ratio would know the numerator for that ratio. The second is that ING had itself decided to insert the figure of €4 million as transaction costs. Assuming this to be correct, it takes us no further than closing assumption (1). The third is that ING had, separately, described a multiple of 10.35x as the ‘entry multiple’ for the Transaction, calculated by dividing €441 million by an EBITDA 07 of €42.6 million, and, consistently with ING’s and Ros Roca’s approach to EV/EBITDA multiples prior to the EV being fixed, treated this as the relevant multiple for assessing the Transaction. All of the matters referred to in this regard concerned the approach to comparative valuation of the offer and resultant Transaction. None of it was concerned with the calculation of ING’s fee.

It is illogical to say that from stage 2(1) the observer can only conclude what is set out in stage 2(2). At best from Ros Roca’s point of view all that can be said is that a figure of €4 million as transaction costs could have been derived by using 10.35 as the Entry Ratio applicable under the Hawk Retainer. Consistently with all these matters, however, ING – and for that matter Ros Roca - could have identified €4 million as transaction costs on the basis that it appeared to be a reasonable “ball park” figure without going to the trouble of working out what the Entry Ratio would be.

In oral submissions Mr Colton sought to buttress closing assumption 2 by saying that ING either encouraged Ros Roca to believe that the correct Entry Ratio was 10.35 or came under a duty to speak because it realised that its approach to the Entry Ratio would result in transaction costs far greater than a figure of €4 million. As to the former of these, however, for the reasons given earlier nothing in stage 2(1) was directed to encouraging any view about the correct Entry Ratio. The latter was in effect a claim for breach of duty in the performance of the Hawk Retainer. Of itself it gives no ground for asserting that there was any common assumption. I would add that if it were to be relied on, Ros Roca’s case in that regard was so different from anything asserted prior to trial that it would have needed to be carefully and distinctly formulated and to be the subject of an application for permission.

For these reasons I conclude that none of the various ways in which Ros Roca has formulated the claimed estoppel establishes any shared assumption of the kind which is needed. In those circumstances it is unnecessary to consider factual and legal issues which might arise in relation to other matters which Ros Roca would have to establish in order to succeed in its estoppel defence. Those matters are likely to depend upon the precise terms of the assumption in question. I do not consider it desirable to try to analyse them by reference to suggested assumptions which in my view have no substantial basis.

For the reasons I have given ING’s claim fails.