Kitcatt & Ors MMS UK Holdings Ltd & Ors

This action arises from an “earn out” dispute under a share sale and purchase agreement dated 1 March 2011 (“the SPA”).

Kitcatt Nohr Alexander Shaw Ltd (“KNAS”) was an advertising agency specialising in direct and digital marketing which was owned and managed by the claimants. By the SPA the claimants sold their shares in KNAS to the first defendant, MMS (UK) Holdings Ltd (“MMS”), a subsidiary of the second defendant, Publicis Groupe SA (“Publicis”). Publicis is a major global advertising and public relations company with headquarters in Paris. Its chairman and chief executive officer is Mr Maurice Lévy.

Following the acquisition KNAS was to be merged with the United Kingdom division of Digitas, a marketing agency also specialising in digital work which was part of the Publicis group. The UK division, which I shall call Digitas UK, was a division of DM Europe Ltd, a subsidiary of Digitas Inc, an advertising agency headquartered in Boston in the United States which had been acquired by Publicis in January 2007. The merged agency was to be known as Kitcatt Nohr Digitas (“KND”) and was to be managed by the claimants.

The acquisition was structured to include an “earn out”, under which part of the total consideration that the claimants were to receive for their shares (the “Deferred Consideration”) was tied to the performance of KND in the years following the acquisition. As the performance of the new merged agency would depend in part on the performance of the business inherited from Digitas UK, the claimants negotiated warranties designed to protect them against matters of which they were unaware which might have an adverse effect on the future performance of Digitas UK’s business.

It is the claimants’ case that MMS was in breach of one such warranty because circumstances existed which could reasonably be expected to have a material adverse impact on the future earnings of KND. More than half of Digitas UK’s earnings at the time of the acquisition came from a single client, Procter & Gamble. However, it is the claimants’ case, not only that the continuing flow of work from Procter & Gamble was fragile because Digitas UK’s access to such work was controlled by other agencies within the Publicis group, but that those other agencies were already taking steps to cut Digitas UK out of such work, all of which was known to the individuals whose knowledge was relevant for the purpose of the warranty but was not disclosed to the claimants. In the event the Procter & Gamble work which the new agency was able to obtain was dramatically reduced with what the claimants say were disastrous consequences for KND, so disastrous indeed that, applying the relevant contractual provisions apart from any warranty claim, nothing was payable to the claimants by way of Deferred Consideration.

The claimants were unhappy about the loss of Procter & Gamble work. They considered that this was something they should have been told about before the SPA was concluded. They argued that they should be entitled to have their right to Deferred Consideration calculated in a way which removed the impact of the loss of this work. They maintain in this action that the validity of their grievance was accepted by Digitas senior management, and also eventually by Mr Maurice Lévy himself, and that an agreement was reached in two stages. The first stage was an agreement in principle in October 2012 that adjustments would need to be made for the purpose of calculating the claimants’ entitlement to Deferred Consideration. The second stage was an agreement as to what adjustments would be necessary – in effect, an agreement on a revised formula which would apply to the calculation of Deferred Consideration. That agreement is said to have been made in December 2012 although it was only possible to calculate the monetary Deferred Consideration to which the claimants would be entitled in accordance with this revised formula after the conclusion of the 2013 year.

In this action the claimants’ primary case is a claim for breach of warranty which, put at its highest, is now alleged to entitle them to the sum of £4.85 million by way of Deferred Consideration although at an earlier stage much higher figures were put forward. Their alternative case is that they are entitled, pursuant to the agreement made in December 2012, to Deferred Consideration in the sum of £3.6 million, alternatively £2.6 million.

The defendants deny that they were in breach of the warranty. Indeed they go further and contend that the warranty in question was unenforceable. They deny also that any agreement to compensate the claimants was concluded, either in principle or in detail. They say also that the claimants’ calculation of their warranty claim takes no account of the impact of factors other than any breach of warranty on the performance of KND during the relevant years and that such factors had a material impact on the failure of the claimants to achieve their “earn out”.

The claimants were the shareholders in KNAS, which was launched in 2002 by the first four claimants. Mr Marc Nohr was the Managing Partner, responsible for the general management of the business. He was the individual principally involved in negotiating the SPA and became the Chief Executive Officer of the merged agency KND. He was the claimants’ principal witness.

The Chief Creative Officer of KNAS and in due course of KND was Mr Paul Kitcatt. Ms Yvonne Alexander was the Client Partner, responsible for client relationships and, after return from extended maternity leave, performed a similar role at KND as well as running an agency associated with it known as Duke. They both gave evidence which was essentially supportive of Mr Nohr’s evidence. As Mr Nohr had been extensively cross examined, the cross examination of these witnesses was limited in the case of Mr Kitcatt, while Ms Alexander was not cross examined at all. Mr Jeremy Shaw was the Chairman of KNAS. He was responsible for financial matters at KNAS and subsequently at KND. Sadly he died in December 2015 before exchange of witness statements in this action although his solicitor’s attendance note of a meeting with him was in evidence.

Publicis was formerly a European advertising agency but by the time with which this action is concerned had become the third largest multinational advertising and communications company in the world. Its growth had occurred as a result in part of a series of significant acquisitions. These acquisitions were mainly handled by the M&A department in Paris which reported to Mr Lévy. Mr Johann Dupont was an Associate Director of this department with responsibility for the acquisition of KNAS. Neither Mr Lévy nor Mr Dupont gave evidence. Mr Stephane Estryn from the M&A department did give evidence, but although he was a frank and helpful witness, his knowledge of the matters in issue was limited.

Included in the Publicis group were a number of well-known agencies such as Publicis Worldwide, Saatchi & Saatchi, Leo Burnett and Digitas. The agencies within the group operated as distinct businesses and competed against each other, although all reported ultimately to the group’s headquarters in Paris.

Individuals within Digitas who are of particular importance in this case include Mr Stephen Beringer (former Chief Executive Officer of Digitas International), Mr Joe Tomasulo (former Chief Financial Officer of Digitas) and Ms Charlotte Frijns (former Finance Director of Digitas UK). These were the individuals whose knowledge is relevant for the purpose of the warranty. All of these gave evidence. Mr Beringer and Ms Frijns attended the hearing. Mr Tomasulo, who is no longer employed within the Publicis group, gave evidence by video link from the United States.

Ms Anne Davis was formerly a Senior Vice President and Global Marketing Lead at Digitas UK. She was responsible for the Procter & Gamble work in a number of markets, including the United Kingdom. Although a witness statement from her was adduced, she is no longer employed within the Publicis group and, in the event, it did not prove possible to take her evidence by video link. Accordingly her evidence consisted of a witness statement untested by cross examination. She was potentially an important witness and it is unfortunate that the defendants did not make arrangements to ensure that she could give oral evidence. I reject the suggestion made on behalf of the claimants that the defendants were actively seeking to shield Ms Davis (and Mr Tomasulo) from cross-examination and recognise that her availability was not entirely within their control. Nevertheless, the fact that her evidence is untested means that it can be given only limited weight.

MMS is not an advertising agency. It is not an operating company at all, but is simply a special purpose vehicle established as an intermediate holding company for UK companies within the Publicis group. In practice, when an acquisition is contemplated, it will be promoted by an entity within the group which, together with the M&A department, will negotiate with the shareholders of the target company and carry out the appropriate due diligence, with a final decision being made by Mr Lévy. The identity of the contracting party will be regarded as a detail, to be inserted in the contract at a late stage.

That is what happened in this case. The initial approach came from Digitas in the United States which was dissatisfied with the existing management of Digitas UK and recommended to Mr Lévy a merger with KNAS as a way of improving growth and profitability in the United Kingdom. This led to a Letter of Intent dated 10 September 2010 on Publicis’ notepaper. It stated that:

“I am pleased to submit the following offer whereby Publicis Groupe Holdings B.V. or one of its affiliates (“Publicis” or “Publicis Groupe”) will, based on your interactions with Digitas teams and the information S.I. Partners and yourselves provided to Publicis … acquire 100% of the securities … [of KNAS]. … ”

It was therefore clear from an early stage that the contracting party which was to buy the shares in KNAS would need to be identified in the final documentation. The identification of MMS as the “Buyer” occurred at a late stage prior to finalisation of the SPA. However, as the Letter of Intent made clear, the affiliate in question (in the event MMS) would enter into the SPA based on the “interactions” between the claimants and representatives of Digitas and Publicis.

The SPA, together with its schedules, is a complex document running to over a hundred pages which was drafted with City lawyers advising both parties.

It provides for the sale to MMS of the claimants’ shares in KNAS, to be followed as soon as reasonably practicable by a merger of KNAS and Digitas UK to form KND “in accordance with the Integration Plan”, the latter being defined as “the integration plan in the Agreed Form setting out the implementation of the Merger”. The consideration for the sale of the shares was to consist of an “Initial Consideration” of £5 million (which the claimants have received) together with “Deferred Consideration” to be calculated in accordance with a formula set out in clause 14.1 and developed in detail in Schedule 5. The precise details do not matter. In outline, the amount of the Deferred Consideration was dependent on the compound annual growth rate in Revenue over the period up to and including 2013 and the Average Operating Income Margin over the period from 2012 to 2013. However, no Deferred Consideration would be payable if the Average Operating Income Margin for 2012 and 2013 was less than 10% or if growth in Revenue did not reach a certain level. Once those levels were reached, there could be increases in the Deferred Consideration as further steps in Average Operating Income Margin or Revenue for 2012 and 2013 were achieved.

“Revenue” is a defined term in the SPA. The definition provides that it means the total revenue of KND in a relevant period, determined in accordance with various accounting principles, but it provides also for a reduction for calculation purposes of the “Base Year Revenue” in the event of the loss of two particularly important clients. The relevant part of the definition in clause 4.1(p) of Schedule 5 reads as follows:

“In the event that Digitas international network does not retain:

(i)

the current worldwide (excluding France) Procter & Gamble client account on such worldwide basis (such that the account is not retained in the United Kingdom), Base Year Revenue shall be reduced by a sum equal to the amount of such Base Year Revenue that was attributable to Procter & Gamble and/or

(ii)

the current European Nissan client account on such European basis (such that the account is not retained in the United Kingdom), or the current European Nissan account in the UK only at Nissan’s request on the grounds of conflict or dissatisfaction with the implemented conflict avoidance arrangements, Base Year Revenue shall be reduced by a sum equal to the amount of such Base Year Revenue was attributable to Nissan.”

This “carve out” of Procter & Gamble revenue in the event of Procter & Gamble being lost as a client by the Digitas international network did not reflect any perceived likelihood of such a loss occurring, any more than the equivalent carve out in the case of the Nissan account being lost. There was no reason to think that the Nissan account would or might be lost. Similarly, the issues arising in this action as a result of the loss of the Procter & Gamble account arise out of the Buyer Warranty referred to below.

“Operating Income” is also a defined term. It refers to Revenue less operating expenses (after making various adjustments) and would therefore more naturally be described as profit rather than income.

It is common ground that if the provisions of clause 14.1 and Schedule 5 are applied to the actual Revenue and Operating Income achieved by KND during the relevant years, the Deferred Consideration would be nil. That is so even after applying the reduction in “Base Year Revenue” for which the definition clause provides. The claimants’ claim in this action depends upon proving either a breach of the Buyer Warranty or an agreement which goes further than the mechanism provided in paragraph 4.1(p) of the definition clause.

Clause 10 of the SPA contains various warranties, most of which were to be given by the claimants who are described in the clause as the “Warrantors”. These include an overarching warranty in clause 10.1 to the effect that “except as Disclosed, each of the Warranties is true and accurate at Completion”. The term “Disclosed” is defined as referring to disclosure in a “Disclosure Letter” to be provided by the claimants. Clause 10.2 provides:

“A matter shall be regarded as disclosed by the Disclosure Letter only to the extent that information about that matter is disclosed in the Disclosure Letter fairly which expression shall mean that a matter has been disclosed only to the extent that it has been disclosed with sufficient detail to enable a reasonable buyer to identify the likely nature and scope of that matter.”

The standard of disclosure so far as disclosure by the claimants was concerned was, therefore, that the disclosure had to be “in sufficient detail to enable a reasonable buyer to identify the likely nature and scope of” the matter disclosed – or to put the same point in another way, the disclosure had to be sufficient to enable a reasonable buyer to understand its significance. A passing reference to some problem would not be good enough.

Clause 10.11 provides for warranties described as “Buyer Warranties” to be given by MMS. The warranty on which the claimants rely is clause 10.11(a) of the SPA, although it is useful to see that paragraph of the clause in its context. This is as follows:

“The Buyer warrants to the Sellers that save as disclosed by the Buyer to the Sellers from time to time prior to Completion each of the following warranties (‘Buyer Warranties’) is true and accurate at Completion:

(a)

none of the Buyer, Stephan Beringer, Joseph Tomasulo or Charlotte Frijns is aware of any facts or circumstances that could reasonably be expected to have a material adverse impact upon the Operating Income and/or Revenue in 2012 or 2013 (being a reduction of at least 20% in the case of Operating Income and 10% in the case of Revenue) including, without limitation:

(i)

the resignation or expected loss of any client of Digitas;

(ii)

any significant current or threatened litigation involving Digitas

(b)

the historic financial information relating to Digitas has been supplied to the Sellers has been properly prepared in accordance with good accounting practice, is not misleading and fairly represents the financial position of Digitas as at the date to which such information was prepared.

(c)

the financial projections for Digitas for the period ending 31 December 2011, which have been provided to the Sellers, were, when prepared, properly compiled in good faith and with due and careful consideration on the basis of reasonable assumptions and such functions were fair and reasonable and there were and remain as at the date of this agreement no material facts known to the Buyer, Stephan Beringer, Joseph Tomasulo or Charlotte Frijns which either were not taken into account in the preparation of such projections or which could reasonably be expected now to have a material effect on them.”

I shall refer to Mr Beringer, Mr Tomasulo and Ms Frijns as the “Named Individuals”. There is an issue as to whose awareness counts as the awareness of “the Buyer”.

The three paragraphs of clause 10.11 deal with the financial position of Digitas UK at different stages, with warranties of different scope at each stage. Paragraph (b) deals with historic financial information. Paragraph (c) deals with a projection for the then current financial period ending 31 December 2011. Paragraph (a) deals with facts and circumstances that could reasonably be expected to affect the financial position of the merged agency in 2012 and 2013. All of these matters were capable of affecting the Deferred Consideration to which the claimants would be entitled in due course.

There are issues as to the construction of clause 10.11(a) and whether the clause is enforceable which I shall consider in a later section of this judgment.

Clause 10.12 provides that the sole remedy for a breach of these warranties is an adjustment to Revenue and/or Operating Income for the purpose of calculating the Deferred Consideration:

“The Sellers’ sole remedy for breach of the Buyer Warranties shall be an adjustment to Revenue and/or Operating Income for the purposes of calculating the Deferred Consideration if and to the extent that the Deferred Consideration is adversely affected by a matter that gives rise to a breach of the Buyer Warranties.”

There will only be an adjustment to Revenue or Operating Income “if and to the extent that the Deferred Consideration is adversely affected” by the facts or circumstances which have not been disclosed. Any number of factors may affect Revenue or Operating Income in 2012 or 2013, but it is only to the extent that the undisclosed facts or circumstances affect these results that an adjustment may be made for the purpose of calculating the Deferred Consideration.

Accordingly two broad questions arise in any breach of warranty claim. The first is whether the warranty (if enforceable) has been broken. The second is what if any adjustment is appropriate.

Advertising agencies will generally be engaged by their clients either on a retainer or for particular projects. A retainer means that the agency will be instructed on a regular basis during the period of the retainer and will be paid either a monthly or an annual fee. Such an arrangement provides security of income, at any rate for the period of the retainer. For project work the agency will pitch for a specific project and, if successful, will be paid in accordance with whatever payment terms are agreed, but the relationship is limited to the particular project in question. When the project is concluded, there will not necessarily be any further work from that client. A retainer may indicate a more secure relationship with the client, but the industry is highly competitive and in neither case is future work guaranteed beyond the period of the retainer or the conclusion of the project. However, whether engaged on a retainer or a project basis, the agency has an opportunity to develop a relationship with the client which, if all goes well, is likely to lead to future work. I accept the evidence of Mr Nohr that in general the development of good working relationships and client satisfaction with the agency’s performance had in the past given KNAS a good track record in retaining clients and could be expected to do so in the future. Mr Nohr and his colleagues had demonstrated at KNAS and were to demonstrate at KND an impressive ability to win new clients and build good relationships with them.

In the course of negotiating the SPA, the claimants sought information about Digitas UK’s existing clients. On 27 September 2010, in response to a question about the contract lengths and notice periods with Digitas UK’s existing clients, Mr Joseph Tomasulo, Digitas Inc’s Chief Financial Officer, advised that:

“Our clients are fundamentally project based so projects expire and new projects are added so notice periods are not the same as on the traditional side. However, although a project based relationship, the growth in P&G (from zero a few years ago) has the confidence of a retainer as P&G is the largest client of the Groupe and a top tier client of both Digitas and VivaKi. Good work has allowed for significant growth and continued delivery will enable growth in the future.”

Procter & Gamble, mentioned in this email, was Digitas UK’s most important client, which in 2010 had accounted for 52% of its revenues. Although its work would represent a lesser proportion of the revenues of the merged agency KND, it would remain KND’s largest single client.

The claimants placed considerable emphasis on this message, in particular on the phrase “the confidence of a retainer”. It is important to keep in mind that this is not a claim for misrepresentation, whether fraudulent or negligent. Nevertheless I accept that this message is a relevant part of the background against which to consider whether (what the claimants say was) the fragility of Digitas UK’s relationship with Procter & Gamble was disclosed to the claimants before the conclusion of the SPA. Certainly there was nothing in this message to suggest that the Procter & Gamble work was under significant or any threat.

For many but not all of its brands, Procter & Gamble operated through a structure known as the “Brand Agency Leader” (or “BAL”) model under which it would designate one agency, the BAL, as a single point of contact in relation to any particular brand. The BAL was responsible to Procter & Gamble for all advertising activity undertaken in relation to the relevant brand. It was the BAL which effectively controlled the client relationship with the Procter & Gamble executive(s) responsible for marketing and advertising the brand in question. The BAL might decide to undertake work itself in-house or to refer or sub-contract it to another agency which had particular expertise. When that occurred, the agency in question would not have a direct relationship with the client and the BAL would in practice be able to exercise a significant degree of control in determining the scope of any work to be sub-contracted, as well as the way in which the work performed by the agency which it had sub-contracted was presented to the client. Indeed there was some evidence that one of the BALs, Saatchi & Saatchi, was adding a mark-up to Digitas UK’s estimates in order to make it appear more expensive to the client than it really was.

Ultimately of course it was for Procter & Gamble to decide which agencies it wanted to engage. It could have insisted that (for example) Digitas UK should undertake digital advertising work on one of its BAL brands and overruled any objections by the BAL. In practice, however, that was not how the model worked. As operated by the BALs within the Publicis group, the BAL had significant and (as it was to prove) often decisive influence in being able to retain work for itself to the exclusion of other agencies within the group, regardless of the quality of work undertaken by those other agencies.

The BAL model was unique to Procter & Gamble. It was not a secret that Procter & Gamble used this model, but the claimants had never heard of it. They had not worked with Procter & Gamble before and there is no reason why they should have done.

Digitas UK’s work in relation to four of the Procter & Gamble brands was obtained through this model via other agencies within the Publicis group. The brands in question and the BALs for each brand were as follows:

In the case of Head & Shoulders, the contractual terms on which Digitas Worldwide was engaged enabled Saatchi & Saatchi to terminate Digitas’s (including Digitas UK’s) engagement on 30 days notice. They provided also that Saatchi & Saatchi would retain ownership of all creative work, complete or incomplete, undertaken by Digitas UK. At the time when this contract was first concluded, the weak position in which it left Digitas UK was fully appreciated by Charlotte Frijns and Anne Davis who registered their concerns with senior management. However, they were told that the contract should be signed anyway. Later, in October 2010 when it became apparent that Saatchi & Saatchi proposed to invoke these terms, they were described as having given rise to “a number of BIG concerns”.

The terms on which Digitas UK was engaged for the other three BAL brands were not in evidence, but it is a reasonable inference that they left it in a broadly similar weak contractual position.

Although some of the defendants’ witnesses, for example Mr Tomasulo, suggested that these were standard terms within the industry irrespective of the BAL model, I do not accept that evidence. It is difficult to reconcile with the undoubted fact that the terms of the Head & Shoulders contract were a matter of particular concern within Digitas.

These four BAL brands represented an important part of Digitas UK’s Procter & Gamble revenue before the merger, but there was also revenue from other non-BAL brands. The figures for 2010 were as follows:

It can be seen that the four BAL brands together accounted for £2,076,000 out of a total of £3,573,000 (or 58%) of Procter & Gamble revenue in 2010.

At this point it is necessary to mention some of the many issues which arose during the trial as to the scope of the pleadings. Although some of the claimants’ submissions did not distinguish between Procter & Gamble revenue which was obtained from work undertaken pursuant to the BAL model and work which was obtained directly from Procter & Gamble, the Particulars of Claim made clear that the claimants’ case was that “a significant proportion of the business that Digitas UK had previously generated from P&G” or even “the great majority” of Digitas UK’s work for P&G “did not originate from any direct relationship between Digitas UK and P&G, but had instead been obtained on a referral basis from other Publicis-owned advertising agencies that had direct ‘lead agency’ relationships with P&G relating to specific P&G brands”. The only such brands identified in the claimants’ pleadings were the four BAL brands identified above. There was no mention of any other Procter & Gamble brands, let alone of any loss of such brands as a result of anything done by any BAL.

Two conclusions follow. First, the claimants have not pleaded a case that all of Digitas UK’s Procter & Gamble revenue was obtained through the BAL model but only that a significant proportion of it was. Second, the claimants’ pleaded case as to revenue obtained by Digitas UK through the BAL model is limited to the four BAL brands. There is no pleaded basis on which to conclude, nor (perhaps more importantly) is there any evidence which I accept, that the revenue from other Procter & Gamble brands mentioned in the table above (e.g. Eukanuba) came via BALs or even that there was a BAL for those brands.

In their closing submissions the claimants drew attention to a paragraph of the Particulars of Claim which referred in more general terms to “Digitas UK’s P&G business” having been obtained on a referral basis from other Publicis-owned agencies holding BAL roles for specific brands. However, when the paragraph is read in the context of the pleading as a whole, it is clear that this refers to the four named BAL brands.

On the other hand, the defendants contended at the trial that part (or even a majority) of the revenue from the four BAL brands came to Digitas UK otherwise than through the BAL model. In particular, they relied on a spreadsheet which appeared to show that in 2011, when total Oral B revenue amounted to £967,663, only £160,760 (or 16.6% of the total) was BAL revenue, although (as the claimants pointed out) another spreadsheet calls that conclusion into question. In my judgment, however, it is not open to the defendants on the pleadings to contend that any significant part of the revenue from the four named BAL brands was obtained otherwise than through the BAL model. No case to that effect was advanced in their Defence. On the contrary, the Defence pleaded that “the H&S, Tampax and Oral B brands were subsequently retained in-house by their relevant BAL agencies in 2011 and 2012” without drawing any distinction between different revenue streams for such brands. If such a case had been pleaded, the claimants would have had an opportunity to adduce evidence to deal with it. As it is, although there is some limited evidence that not all Oral B work came to Digitas UK through the BAL model, there is no reliable evidence (at least none to which I was referred) to enable a conclusion to be reached as to the proportion of Oral B work or of work from the other BAL brands which was obtained through the BAL model from time to time.

Accordingly I proceed on the basis that all of the revenue from the four BAL brands (Ariel, Head & Shoulders, Oral B and Tampax) was obtained through the BAL model, but that none of the revenue from other Procter & Gamble brands was obtained in this way. Given the state of the pleadings and the way in which the parties prepared for trial, that is the only fair basis on which to proceed.

A comparison of the 2010 figures with those for the succeeding years shows that the new agency KND suffered a dramatic reduction in revenue from Procter & Gamble work:

It can be seen that the revenue from work on Ariel and Head & Shoulders, both BAL brands, reduced significantly, beginning in 2011, and that by 2013 such work had reduced to nothing. Revenue from Oral B and Tampax had also ceased by 2013. Indeed, by 2013 (and with the exception of work recorded simply as “Other”), all Proctor & Gamble work had reduced significantly by 2012 and had ceased by 2013. As explained below, the fact that Ariel work would be lost was known by Digitas UK before the date of the SPA. Whether or to what extent it was known that other BAL work would cease is in issue.

The claimants put their case in two ways. First they say that MMS was in breach of the Buyer Warranty in clause 10.11(a) such that they are entitled to an adjustment for the purpose of calculating their right to Deferred Consideration pursuant to clause 10.12. This is advanced as their primary case. Alternatively they say that an agreement was reached in 2012 that certain (alternatively reasonable) adjustments would be made for the purpose of calculating the Deferred Consideration in order to ensure that their right to Deferred Consideration was not prejudiced by the loss of the Procter & Gamble business.

The claimants’ case is that there was a breach of the Buyer Warranty in clause 10.11(a) of the SPA by reason of non-disclosure to them of the following facts and circumstances:

As the claimants made clear in the course of their opening submissions, this case does not relate to work undertaken by Digitas UK in respect of any specific Procter & Gamble brand, but to the entirety of Procter & Gamble work undertaken through the BAL model. However, it is important that the breach of warranty case is confined to the loss of BAL revenue and its consequences. There is no pleaded case that the defendants were aware of any facts or circumstances which could reasonably be expected to lead to the loss of revenue from Procter & Gamble obtained otherwise than through the BAL model (for example, Eukanuba revenue), or indeed to the loss of revenue from any other client.

Accordingly (and on the assumption that clause 10.11(a) is enforceable) the following issues arise:

If these questions are answered affirmatively, the claimants will have established a breach of the Buyer Warranty.

As I shall explain, it is not necessary, in order to make good a case of breach of warranty, for the claimants to prove bad faith on the part of the defendants or their witnesses. Clause 10.11(a) is concerned with the actual awareness of facts and circumstances which could reasonably have been expected to have a material adverse impact, whether or not that impact was in fact appreciated at the time. Nevertheless the claimants chose to advance a case of bad faith on the part of the defendants. A flavour can be seen in the opening paragraph of their written opening submissions, although it permeated much of their approach to the case:

“The story is framed in two acts, and goes right to the top of the Publicis group. The first act revolves around a Publicis subsidiary withholding critical information from the former shareholders of KNAS, in circumstances so appalling that Publicis’ own employees have subsequently reacted with embarrassment and guilt.”

There is no justification for this hyperbole. It was not in the defendants’ interests deliberately to withhold critical information from the claimants which would inevitably come to light after the merger and which would leave the senior management of KND disincentivised and distrustful. Nor did they do so. Indeed, it was in the interests of the defendants as well as the Named Individuals themselves that the acquisition of KNAS would be a success. I shall need to consider whether the Named Individuals or MMS itself were aware of facts or circumstances which, looking at matters objectively, could reasonably have been expected to have a material adverse impact on KND’s business in 2012 or 2013, but I say at once that I do not accept that there was a deliberate withholding of such information in bad faith.

If a breach of warranty is established, the question will then arise whether the claimants are entitled pursuant to clause 10.12 to an adjustment to Revenue or Operating Income for the purpose of calculating the Deferred Consideration. That will depend on whether and to what extent it has been proved that the Deferred Consideration was adversely affected by these matters. In this regard the claimants say that the loss of the BAL revenue was catastrophic for KND, which not only lost the anticipated revenue but was forced to restructure its business, cutting costs and reducing its workforce to reflect the fact that it was now a much smaller business than had been envisaged. The result was that it was unable to service existing clients and win new ones as it would have done if the BAL revenue had continued. Accordingly the claimants contend that the consequences of breach of warranty are to be measured, not merely by reference to the lost BAL revenue, but taking account of the wider consequences for the KND business as a whole.

Alternatively, the claimants say that an agreement was reached in two stages. The first was an agreement to the principle of an adjustment to protect them from the impact of the loss of the Procter & Gamble business in the calculation of their entitlement to Deferred Consideration (or, as it was put in an email of 26 October 2012, “the principle of an adjustment of Kitcatt Nohr Digitas P&L to carve-out P&G impact is accepted”). The second stage was an agreement, said to have been confirmed in a telephone conversation on 17 December 2012 between Mr Nohr and Mr Tomasulo and an email of the following day, on the precise adjustments which would be necessary to give effect to this principle.

The defendants say that although the parties discussed the question of an adjustment, and got as far as agreement on the part of Mr Johann Dupont of the Publicis M&A department in Paris that the adjustments proposed by the claimants were acceptable, this was as far as any agreement went. They say that it was well understood that no binding agreement could be made without the approval of Mr Lévy which was never forthcoming.

It is common ground that if a binding agreement was made in December 2012 as the claimants contend, the result is that they are entitled to Deferred Consideration of either £3.6 million or £2.6 million and that which of these is the correct figure depends on an issue as to the treatment of fees charged by the Publicis Shared Service Centre or “SSC”.

The claimants’ pleaded case is that the agreement reached was binding on both MMS and Publicis. The way they put it is that:

“a binding agreement was concluded in 2012 between [the claimants] (on the one hand) and MMS UK/Publicis (on the other hand) (‘the 2012 Agreement’) by which MMS UK agreed, further or alternatively Publicis agreed to procure, that certain (alternatively, reasonable) adjustments would be made to the calculation of the [claimants’] entitlement to deferred consideration in order to ensure that this entitlement would not be prejudiced by the P&G losses referred to above”.

An issue arises as to the relationship between the two ways in which the claimants put their case.

It is not disputed that if the claimants can establish that a binding agreement was made as alleged in December 2012, the claimants are entitled to recover £3.6 million (or £2.6 million as the case may be) regardless of whether they are able to prove a breach of the Buyer Warranty and regardless of whether the Buyer Warranty is enforceable at all. The defendants say that if (which they deny) such a binding agreement was concluded, that can only have been by way of compromise of any claim for breach of warranty or, alternatively, that the agreement satisfies or exhausts any remedy they may have pursuant to clause 10.12 for loss of the Procter & Gamble revenue. Accordingly the defendants say that as a matter of logic, the claimants’ alternative claim pursuant to this agreement should be their primary claim and that it is only if the claimants fail to prove the making of a binding agreement that any question of a claim for breach of warranty can arise.

The claimants on the other hand say that the December 2012 agreement was a freestanding agreement which did not compromise or otherwise affect their ability to advance a claim for breach of the Buyer Warranty and that they are entitled to elect for whichever of their claims gives them the greater recovery, always provided that there can be no double recovery for the same loss. They acknowledge, however, that this argument only arises if the December 2012 agreement did not amount to a compromise of their breach of warranty claim.

These arguments only emerged at a relatively late stage of the proceedings. The claimants’ reliance on an agreement being concluded in December 2012 was only introduced by an amendment made in January 2017, shortly before the trial. The defendants’ point that any such agreement would necessarily extinguish any pre-existing claim for breach of warranty was first made in their written opening submissions. The claimants’ response that the December 2012 agreement was a freestanding agreement and that they were entitled to elect between remedies was advanced only in the course of the trial, after Mr Nohr had given evidence.

It is relevant to examine the way in which the claimants put the case in their pleading. This was that as a result of the December 2012 agreement:

“the question of how the loss of KND’s P&G business would affect their entitlement to Deferred Consideration under the SPA had been resolved”

and that:

“MMS UK agreed … that certain adjustments would be made when calculating the Former Shareholders’ entitlement to Deferred Consideration under the SPA in order to ensure that this entitlement would not be prejudiced by the P&G losses referred to above.”

Mr Nohr’s written evidence was to the same effect. It was that the purpose of the discussions which resulted in the December 2012 agreement was to address:

“the specific question as to how the loss of KND’s P&G business would affect the former shareholders’ entitlement to a deferred consideration payment under the SPA.”

It is therefore the claimants’ pleaded case, supported by their own evidence, that the December 2012 agreement did address, and resolved by agreement, the precise issue raised by their breach of warranty claim. In those circumstances I accept the defendants’ submission that, if the claimants establish a binding agreement made in December 2012, there is no remaining scope for any additional recovery pursuant to a breach of warranty claim. That would be so regardless of whether the December 2012 agreement is correctly characterised as a compromise of the breach of warranty claim although, as I shall explain, I consider that any agreement is correctly so characterised.

As will be seen, there was at least a potential dispute: see Foskett on Compromise, 8^th Edition (2015), para 2-01. Certainly that is how the claimants understood the matter, albeit that for understandable tactical reasons they chose to keep their breach of warranty claim in the background during the negotiations which led up to December 2012. They were, however, well aware of the possibility of a claim for breach of clause 10.11(a) and regarded themselves as negotiating to secure an agreement which would avoid the necessity for such a claim. It would be remarkable if the defendants understood the position differently. An agreement can compromise a potential dispute even if the parties do not use the express language of compromise, language such as “in full and final settlement”. What matters is the substance.

Nevertheless, and whether or not the breach of warranty claim is correctly regarded as the claimants’ primary claim, it is convenient to address this first.

Before dealing with the facts in detail, I should say something about the way in which Digitas UK forecast its future revenue. This was described by Charlotte Frijns, the Finance Director for the UK business. Each year in September or October she would produce a projection of revenue and profitability for the calendar year ahead. This was known as “the Commit”. She described it as “an aspirational target”, but one which was “reasonably expected and attainable”. It would be broken down by brands and would include projections of new business as well as existing business. Once the Commit was finalised and approved, performance would be measured against it during the year.

In addition rolling forecasts would be produced in March, June, September and November of each year. These would be more accurate as the year progressed, as they would include actual revenue achieved as well as forecast future revenue. These too were intended to be reasonably expected and attainable revenues. They were produced in consultation with the business leaders responsible for each brand on which Digitas UK was working.

Ms Frijns’ particular concern was that if projected revenues were falling below the targets set in the Commit, it would be necessary to adjust Digitas UK’s cost base in order to maintain profitability. In practice this would mean making adjustments to salary costs, either by reducing freelance staff, cancelling or reducing new hires, or freezing salaries.

This was by its nature a process which focused on the short term. The Commit looked at revenue and profitability only one year ahead while the rolling forecasts were concerned only with the current calendar year. It was important to Ms Frijns to be aware of forecast reductions in revenue which would affect the figures in the current Commit and rolling forecasts and which might require cost adjustments to be made, but forecasts which looked further ahead would not affect these figures and were therefore of much less interest to her. To the extent that she considered such projections at all, they did not affect the figures in the Commit or rolling forecasts.

This was the process followed during 2010 and 2011. In September or October 2010, the 2011 Commit was produced. This was followed by rolling forecasts for 2011 produced in each of March, June, September and November of that year. These forecasts for the Procter & Gamble brands, together with the 2011 actual revenue figures, are shown in the table below:

The table shows that between the date of the Commit, prepared in September or October 2010, and the March 2011 rolling forecast there was a small decline in the expected 2011 revenue from the BAL brands which included the loss of the Ariel brand, but the figures for Head & Shoulders and for Oral B remained broadly steady. No revenue from Tampax had been forecast at the date of the Commit, although by March 2011 revenue of £215,000 during the year was expected, not all of which materialised as the year progressed. There was, however, a marked reduction in forecast Head & Shoulders revenue by the date of the June and September rolling forecasts which show essentially that by the date of these forecasts there would be no future work on this brand.

Between the same dates, September/October 2010 and March 2011, the forecast revenue from non-BAL brands increased and, as 2011 proceeded, some of the revenue forecasts increased further.

However, these figures say nothing about what was expected, reasonably or otherwise, in 2012 or 2013. They are solely concerned with forecast revenue for 2011. In contrast, clause 10.11(a) is concerned with facts and circumstances which may reasonably be expected to have an impact in 2012 or 2013.

There was no evidence to suggest that Digitas UK had any system for forecasting revenue more than one year ahead. In other words, by the date of the entry into the SPA, there would have been no forecasts predicting the revenue from particular clients or brands which could be expected to be achieved during 2012 or 2013.

I set out now a chronological narrative of the events which are relevant in this action. In making the findings of fact set out below I have relied mainly on the contemporaneous documents but have also taken into account my assessment of the evidence given by the various witnesses.

As indicated above, the 2011 Commit produced in October 2010 included £4 million of revenue from Procter & Gamble, of which some £2.6 million was from BAL brands.

However, by October 2010 Digitas senior management in the United States had recognised that the BAL model left Digitas vulnerable as a result of BALs taking over digital work hitherto done by Digitas. Documents prepared for a planning meeting on 18 October 2010 (which were sent to Mr Beringer and Mr Tomasulo, among others) included a “Key Client Action Plan” relating to Procter & Gamble which identified issues and threats as follows:

“Key issues

Partner agency workload (BAL) with Saatchi/Leo/Publicis continues but scope/fee negotiations with them is driving more investment/less margin …

Biggest threat: Partner agencies looking to take our digital work as part of their overall relationship.”

A similar document sent at about the same time by Mr Tomasulo to Laura Lang, the Global Chief Executive Officer of Digitas, expanded on the nature of this threat:

“Biggest threat: Partner agencies looking to take our digital work as part of their overall relationship as roll [sc. role] of digital only agency diminishes at P&G

•

BAL’s positioning Digitas as an execution shop

•

Digitas voice is getting lost in the BAL model and P&G not feeling the direct benefit of Digitas in the equation.”

These documents illustrate several factors which in due course were to loom large. These included the fact that it was the BAL which had the relationship with Procter & Gamble, that Digitas was not receiving credit with Procter & Gamble for the work which it was doing and was portrayed by the BALs as doing no more than executing the ideas of others, and the threat that other agencies within the Publicis group which had the status of BAL would be able to take for themselves work which had hitherto been done by Digitas. Mr Tomasulo sought to play down these concerns in his evidence, but I accept that they were real and significant concerns at the time.

At about this time or later in the same month it became apparent that these fears were well-founded so far as the Ariel and Head & Shoulders brands for which Saatchi & Saatchi was the BAL was concerned. Procter & Gamble was keen to establish what it described as “an integrated ad agency digital team … with an integrated BIG idea”. Although it appears that Procter & Gamble saw this as a way in which the best digital capability from the Publicis group as a whole would be assembled to work on its brand, Saatchi & Saatchi (and in particular Mr Vaughan Emsley who was the Saatchi executive responsible for the relationship with Procter & Gamble) saw this as a way of obtaining digital work for itself and excluding Digitas. Anne Davis (who distrusted Mr Emsley and felt that he was not being straight with her) was very concerned about this, particularly when she learned that this proposal was not confined to Ariel (the first brand on which this model was to be adopted) but was intended to apply also to the Head & Shoulders brand. She regarded it as a “land grab … within the groupe for our current P&G business”. When she reported the news to Colin Kinsella, the Chief Executive Officer of Digitas US, his response was that it was “troubling – and unfortunately, confirms our worst fears”.

Ms Davis’s concerns, as reported in an email to Mr Kinsella dated 24 October 2010, extended to the impact on forecast revenue for 2011 if the Head & Shoulders brand were to be lost to Saatchi & Saatchi. Mr Kinsella responded that this was “the tip of the iceberg”. Digitas began, therefore, to consider how it would respond to this new situation. A “P&G 2011 Strategic Account Review” presentation of October 2010 referred to Digitas as being “under attack and need to evolve fast”. The attack in question was said to be as a result of the BAL model:

“Attack coming from our frenemies

•

BAL model drives partnership mentality; lead spot enables brand shops to take credit for all thereby driving perception that digital ideas can come from brand shops.”

The “frenemies”, enemies who should have been friends, were the BALs (or “brand shops”) which were other agencies within the Publicis group.

There is no doubt that at this time this issue was seen as one which was likely to have a serious impact on Digitas UK’s profitability. Emails exchanged among the Digitas senior management, including Mr Beringer and Mr Tomasulo, demonstrate the frustration which was felt. As Ms Davis put it in one email, “This whole thing sucks”. One email on which Mr Beringer and Mr Tomasulo were copied referred to the loss of Ariel and Head & Shoulders as a result of the BAL providing “the digital talent” itself as distinct from engaging other agencies such as Digitas as having “huge implications on our P&L in the UK” and as being “a big deal”. I see no reason to doubt that this was indeed how the matter was viewed. Mr Beringer claimed in his evidence to have no recollection of any major issue concerning Procter & Gamble at this time. He too played down the significance of what was described in the email, claiming that it was nothing unusual. However, there is no evidence that anybody reacted by saying that Ms Davis’s concerns were exaggerated or that there was nothing to worry about. Plainly there was, although the timing of any such impact was uncertain. While the loss of Ariel was seen as something which was to happen shortly, it was understood that this would represent a pilot scheme and that the loss of other brands such as Head & Shoulders would be further in the future. This was in part because Digitas took the view, encouraged by Mr Emsley, that Saatchi & Saatchi did not yet have sufficient digital capability to undertake all this work and would wish for the time being to continue working with Digitas.

The first definite change came on 5 November 2010 when Saatchi & Saatchi advised Ms Davis “that our strategic and creative services were not required any further on Ariel”. This decision was recognised in the Monthly Brand Report for October 2010, sent to Mr Tomasulo among others in early November, the Procter & Gamble section of which was prepared by Ms Davis:

“Procter & Gamble (P&G)

Ariel WE [Western Europe] have decided to work exclusively with S&S on advertising and digital creative campaigns moving forward, London agency currently in discussions with S&S London on revenue management. Impact on 2010 will be a £50k shortfall, £500k in 2011. The active plan is to replace revenue through UK MDO and Wella Global new business development. H&S and Oral Care teams are on alert for any similar activity from the BAL advertising agencies. London team is working with C Kinsella/D Beder on strategies to re-position Digitas at P&G.”

The report recognised the severe impact of the loss of the Ariel business (the loss of expected revenue of £500,000 in 2011) and the risk of further losses of the Head & Shoulders and Oral B brands. Although the report referred to an “active plan” to replace the lost revenue, it appears that this was no more than an aspiration. Ms Davis’s evidence about this (untested by cross examination) was that:

“Put simply, when an account closed it was not acceptable to simply shrug your shoulders and hope that revenue would increase elsewhere. Rather, you always had to make up any reduction and there would have been pressure on the account team to do just that.”

No doubt there would have been pressure and perhaps even hope, but there is no basis on which to conclude that there was any realistic basis for such hope.

The loss of the Ariel business was referred to in a spreadsheet produced in November 2010 which indicated that no revenue from this brand was forecast for 2011 and that the brand had moved to Saatchi & Saatchi. The spreadsheet indicated also, under the heading “client history”, that there had been a “poor client relationship” in the case of the Ariel brand. In contrast, the spreadsheet indicated a strong forecast for Head & Shoulders (revenue of £982,000 in 2011) and Oral B (revenue of £935,000) and described the level of client satisfaction in both cases as “high”. The total revenue from Procter & Gamble for the year was forecast to be £3.7 million, which represented a reduction of £300,000 from the figure of £4 million in the Commit.

This spreadsheet was provided to Mr Nohr as an attachment to an email dated 23 November 2011 which referred to Digitas as having “relationships across a wide spectrum” with Procter & Gamble. It did not say anything about the BAL model.

Mr Nohr’s evidence was that he would have done his best to study this spreadsheet, although he had no recollection of having read it. He denied that any explanation of these figures had been provided. Mr Beringer confirmed this. Mr Nohr agreed, however, that a fair reading of the spreadsheet would indicate that the Ariel account had been permanently lost to Digitas. However, he pointed out also that there was nothing in the spreadsheet to indicate that the loss of the account was attributable to predatory action on the part of Saatchi & Saatchi.

The expected timing of further moves became clearer following a conversation between Mr Colin Kinsella (Ms Davis’s superior) and Mr Elmsley of Saatchi & Saatchi on 6 December 2010. Mr Kinsella’s report of that conversation was to the effect that:

“Timing may be a bit slow as they need to sort out process but feels everything will be shifted by 2012. He couldn’t speak for the timing of the other BAL agencies, but felt it would be similar.”

This report recognised that although Digitas could expect to retain Head & Shoulders work during 2011, not least because (according to Mr Elmsley) Saatchi & Saatchi was not yet ready to take over from it, by 2012 (and even more so by 2013) that work would have been lost. It represented also a warning that the other BAL agencies, Leo Burnett and Publicis Worldwide, were also likely in Mr Elmsley’s view to take over for themselves the work which had previously been done by Digitas. Mr Kinsella concluded with the gloomy comment that “It is clear our role will shrink across P&G if we don’t reverse this perception” (i.e. the perception that digital agencies could not deliver creatively) and set out some ideas for next steps for Digitas to take in order to reverse this perception and win further Procter & Gamble business.

While it does not appear that Mr Tomasulo saw this report, it is clear that he was aware of the concerns at the highest levels of Digitas that Saatchi & Saatchi was taking active steps to get Digitas “pulled off” Procter & Gamble business. In an email dated 9 December 2010 he urged Ms Lang that something needed to be done:

“Apparently Saatchi is taking the position that Digitas is a ‘competitor’ and is taking active steps to get us pulled off clients like Mead Johnson and PG. We need a strategy to deal with this (do we go on the offensive for example?). The troops need direction.”

The Procter & Gamble section of the Monthly Brand Report for November 2010, again prepared by Ms Davis and sent to Mr Tomasulo in early December, included the following:

“Procter & Gamble (P&G)

London is reviewing the 2011 forecasts and headcount in light of the loss of Ariel creative responsibility to Saatchi & Saatchi under a client global pilot to integrate the digital and advertising work. The £500k shortfall is being replaced through additional H&S global assignments, UK MDO assignments and Prodigious UK MDO retainer fee negotiations. …”

While the loss of £500,000 of forecast 2011 revenue was now a fact, the replacement of this revenue with other Procter & Gamble work remained no more than an aspiration. The list of “new business” won in November contained in a different section of the report did not include any new Procter & Gamble business. It is relevant to notice also that according to this report the loss of the Ariel work meant not only that forecasts would need to be reviewed, but that the “headcount” would also need to be reviewed – in other words, that staff numbers might need to be reduced.

The review of 2011 financial forecasting continued into January 2011. At that stage the question arose, in the light of the loss of the Ariel business, what protection if any Digitas UK had against the loss of brands where Saatchi & Saatchi and Publicis Worldwide were the BALs (i.e. Head & Shoulders and Oral B). Although it is not apparent from the documents what answer was given to this question, the only realistic answer would have been that there was no such protection. I infer that this is the conclusion which must have been reached. It was recognised, including by Mr Tomasulo, that the potential loss of this Procter & Gamble business represented a “potential issue” to the proposed deal with KNAS, although at this stage nobody was thinking in terms of what was eventually to be agreed as the Buyer Warranty.

On 21 January 2011 KNAS was proposing to pitch for new business with L’Oreal, a competitor of Procter & Gamble. Recognising the importance of Procter & Gamble to what would become KND, Mr Nohr asked Mr Beringer for permission to speak to Ms Davis about this. Eventually, on 7 February 2011, Ms Davis advised that it would not be acceptable to Procter & Gamble if KND also had L’Oreal as a client and, in view of this, KNAS decided not to pitch. Nothing was said to Mr Nohr about any vulnerability of the status as a Digitas UK client of Procter & Gamble. Ms Davis’s email listed the Procter & Gamble brands on which Digitas UK worked. The list included (without explanation) all four BAL brands, including Ariel (although the 23 November 2010 spreadsheet had indicated that this brand had been lost). Strictly, it was still true that Digitas UK was undertaking work on Ariel, but this was only the runoff of existing work after notice had been given that there would be no new work. However, the fact that the Ariel work was limited in this way was not explained in Ms Davis’s email.

By this stage, therefore, Mr Nohr had been told two conflicting things about Ariel – first, that the account had been lost to Saatchi & Saatchi because of a poor client relationship, and second, that it had not. In neither case was any explanation provided going beyond the contents of the documents in question. Not surprisingly Mr Nohr could not recall whether he had appreciated at the time that contradictory information had been provided. Having considered his evidence and the inherent probabilities, I find that Mr Nohr’s probable understanding at the time of entry into the SPA was that Ariel remained a client. That was the most up-to-date information, provided to him by Ms Davis who was responsible for the running of the Procter & Gamble business for Digitas in Europe, the Middle East and Africa. Although the purpose of his inquiry had been to establish whether there would be a conflict of interest if KNAS were to pitch for L’Oreal work, his actual inquiry had been to ask what Procter & Gamble brands were being handled by Digitas UK. The answer had been that Ariel was one such brand. It may be that the claimants could have noticed the contradiction in what they were being told about Ariel and asked for further information, but the fact is that they did not. This was not unreasonable. In any event, even if the claimants had understood that the Ariel account had gone, that would not have alerted them to any structural weakness in the Procter & Gamble work more generally.

By late January 2011 KNAS had raised the question of including the Buyer Warranty as a term of the proposed SPA, although the final terms of clause 10.11(a) had not yet been agreed. At this stage a warranty in general terms was being considered, to the effect that neither Mr Beringer nor Mr Tomasulo (and perhaps also Ms Frijns) was aware of facts that would have a material impact on the Digitas UK business or on its financial statements. There was at this stage no mention of 2012 or 2013 and no definition of what would constitute “a material impact”. It is apparent that some consideration was given within Digitas, including by Mr Tomasulo, to the question whether something should be said about “the issue with P&G” which was described to him as “a threat”. Mr Tomasulo commented that:

“i agree. we should discuss with Johann [i.e. Mr Dupont] first so we have something to offer KNAS if this works gets transferred within the Groupe.”

Transfer “within the Groupe” refers to the loss of work to another Publicis company which was the BAL for the brand in question.

It is not apparent from the evidence what if any discussions did take place with Mr Dupont, but at all events no such disclosure was made. Mr Tomasulo did not know why not. So far as he was concerned, he said, he was some way away from the detail of the negotiation and, having made his point, took no further action.

On 29 January 2011 Digitas received some more encouraging news. In a conversation between Mr Tony Weisman (the President of Digitas in the US) and Mr Sandy Kolkey of Leo Burnett (the BAL for Procter & Gamble’s Tampax brand), Mr Kolkey assured Mr Weisman that Leo Burnett was eager to build a strong partnership with Digitas. In another conversation between Mr Weisman and Catherine Guthrie who was responsible for Procter & Gamble at Leo Burnett, Ms Guthrie said in response to a direct question that she had no plan to take away Digitas’s work. However, comforting as the news from Leo Burnett may have been, Tampax was a very new client of Digitas UK and its work represented only a small proportion of its total Procter & Gamble work.

So far as the brands where Saatchi & Saatchi was the BAL were concerned, confirmation that Digitas’s days were numbered came from a conversation between Mr Kinsella and Mr Elmsley on 3 February 2011. Mr Elmsley indicated that although the pace of transition would vary, “fundamentally all digital will move to the BALs”. Mr Kinsella’s bleak assessment was that very soon there would be no new Procter & Gamble work for Digitas on any brand where Saatchi & Saatchi was the BAL.

On 4 February 2011 Ms Frijns sent to Mr Shaw at KNAS a document described as the “Walkforward” which contained Digitas UK’s 2011 revenue forecast. The forecast showed total revenue of £7.3 million, the figure in the Commit. Mr Shaw asked some questions about this document, including a request for a breakdown of 2010 revenue by client and the revenue projections by client for 2011. Ms Frijns responded on 7 February 2011 by sending a spreadsheet, together with the covering comment that:

“We are currently in the process of reforecasting, as some clients will be lower than expected, some higher than expected.”

This of course was normal. It is natural that forecasts need to be adjusted in the light of experience and updated information. However, there was no suggestion of any structural problem with Procter & Gamble revenue.

Ms Frijns’ spreadsheet showed total 2011 forecast revenue for Procter & Gamble as being £4 million as set out in the Commit, together with the comment that:

“The mix between brands has changed since putting the Commitment together, but the total number is still £4m according to Anne Davis.”

This spreadsheet was relied on by the defendants as demonstrating that whereas the Commit forecast of £4 million had been reduced to a forecast of £3.7 million by the time of the 23 November 2011 spreadsheet referred to above, the forecast had now increased back up to £4 million, and that this represented an increase of £300,000 in the forecast 2011 revenue since November. I do not accept this interpretation of the documents. Neither Ms Frijns nor Ms Davis explained these documents in this way in their evidence. There was no evidence of any basis on which such an increase in the forecast between late November 2010 and early February 2011 could be justified. While I see no reason to think that the Walkforward forecast for Procter & Gamble revenue was deliberately overstated, I do not accept that there was a valid basis on which to conclude that the Commit figure of £4 million was still a reasonable expectation. It was no more than an expression of hope, resulting in part from Ms Davis’s over optimism and in part from a reluctance to convey bad news to the group’s headquarters in Paris.

The competition from other Publicis agencies was recognised in Digitas’s “2011 Growth Discussion” document dated 7 February 2011 which referred to “Relationship vulnerability at four key clients”, one of which was Procter & Gamble, and to “Extreme pressure from agency partners (Saatchi, Leo, Publicis) targeting our digital work”. These references do not appear to be specifically concerned with vulnerability as a result of the BAL model, which did not apply in the case of the other three clients mentioned, GM, Kraft and Samsung. Nevertheless, where Digitas UK was vulnerable because it was the BAL which had the relationship with the client, it was to be expected that the BAL would exploit that advantage in targeting work which had hitherto been done by Digitas UK.

Later in February 2011 the question whether any disclosure should be made pursuant to the Buyer Warranty was raised again. By this stage the warranty was approaching its final terms, but materiality was defined as consisting of “a reduction of at least £[*] in the case of Operating Income and £[*] in the case of Revenue”. It was, therefore, impossible to tell what was required to be disclosed. Nevertheless each of the Named Individuals was asked by Mr Dupont whether in the light of this proposed warranty there was any specific disclosure to be made. Their responses (or non-responses) were as follows:

The SPA was finally concluded on 1 March 2011. The night before, on 28 February 2011 at 10:42 pm, Ms Davis sent an email to Mr Nohr which, in response to his request for an early meeting with Digitas UK’s clients, referred for the first time to the BAL model. It was in these terms:

“The European clients are in Geneva and Frankfurt for the most part so if there is an opportunity for you to meet with them if they are in London we can arrange it. They haven’t tended to get involved with the UK office leadership as such given many of the relationships are via the BAL structure of the holding company and therefore they deal with their global teams. It’s a convoluted structure and I will take you through the account background and SAR for 2011 so you can get a better view of what and who and where. We manage both in the UK and Geneva/Frankfurt over 60 client relationships, depending on brand, region, BAL, type of project, favourite colour, etc, etc.”

“SAR” stood for Strategic Account Review. There was no explanation of what the acronym “BAL” stood for. I accept the evidence of Mr Nohr that he did not know what “BAL” referred to and had no reason to think that this was something which he needed to understand before the SPA was concluded. Ms Davis had explained that the structure was convoluted and that it would be explained to him in due course. She had not suggested that it was anything which might affect the parties’ expectations for the future or that Mr Nohr ought to worry about it.

This was followed by another email from Ms Davis early on the morning of 1 March 2011 which also referred to the BAL leads:

“I need to advise Dan Beder, who leads P&G in NA, what is happening so that he can advise Cincinnati-based clients and the US based BAL leads. Can I do that today so that he can coordinate communication at the same time as I advise purchasing and European-based BAL leads?”

This took the matter no further.

Very soon after completion of the acquisition on 1 March 2011 the previously anticipated loss of the Head & Shoulders brand became definite. On 14 April 2011 Saatchi & Saatchi gave notice that from July 2011 it would be retaining all new digital work in-house. This was despite a high level of satisfaction with Digitas UK’s work and despite Digitas UK’s belief that this was not what Procter & Gamble had wanted to achieve. The immediate result was a projected loss of income for 2011 of between £400,000 and £510,000. Ms Davis commented:

“Whilst this change is to be expected, Vaughan [Mr Elmsley] told both Tony and myself that h&s would not move on Europe/Global until 2012/13 and that the US would be the first market they would change.”

It is clear, therefore, that to Ms Davis at any rate the loss of the Head & Shoulders business was not a surprise. On the contrary, it had been expected. What was a surprise was that it happened so soon. She had expected it to happen in 2012 or 2013. That did not lessen her disappointment and anger, not least as she understood that Saatchi & Saatchi had been marking up Digitas UK’s estimates when presenting them to Procter & Gamble in order to make the agency appear more expensive than it really was. She described the behaviour of Saatchi & Saatchi as “utterly disgraceful” and “banditry”.

Mr Beringer was not impressed with this news. He responded:

“This is ridiculous! What can we do to fight back? The impact is huge, how can SnS come to such an agreement ‘behind our backs’?”

It is notable that Mr Beringer described the loss of a single brand, Head & Shoulders, as “huge”.

Naturally Mr Nohr was also very disappointed. Principally this was because of the impact on KND’s business, which as the new Chief Executive Officer he naturally wanted to succeed. But it was also because, as he immediately realised, this would jeopardise the claimants’ “earn out”. He commented to Mr Shaw that it would be necessary to look at the wording of the SPA. It is clear that, among other things, he and Mr Shaw had in mind the terms of clause 10.11(a), although the latter warned that it would be necessary to “tread carefully” in invoking that clause. He suggested that Mr Nohr should attempt to find out discreetly whether the loss of Head & Shoulders and any further losses which might follow had been known or anticipated prior to completion.

It is apparent, therefore, that from a very early stage the claimants were alive to the possibility of a claim under clause 10.11(a) and were seeking to obtain evidence in support of such a claim.

A question to Ms Davis prompted the response that the issue “had been on the table since last July” but that the timing “was not anticipated at this speed”. She added that Publicis Worldwide (the Oral B BAL) “remain[ed] a strong and supportive partner”.

It was immediately obvious to Ms Davis and Mr Nohr that in view of this development it would be necessary to make a significant reduction in KND’s costs. The result was that within a few weeks of the merger, instead of planning for growth, Mr Nohr found himself in an unhappy situation where he was forced to discuss redundancies. I accept that redundancies were necessary because of the loss of Procter & Gamble business, including in particular the loss of BAL business, that this was a foreseeable consequence of the loss of such business, and that this had an impact on KND going wider than the mere fact that expected revenue would not now be earned. However, I do not accept that all of the redundancies which KND was forced to implement were attributable to the loss of Proctor & Gamble business. In particular, a much later email from Mr Nohr to Mr Shaw, dated 29 August 2012, addressed precisely this question. Mr Nohr asked:

“All these redundancies – can we finger p&g for any of them [so] they don’t hit our earn out. Laz is the obvious candidate.”

Mr Shaw replied:

“… I suspect it will be hard to make much of a case given the decline in non-Digitas legacy income as well, and the issues with new business. …”

Although Mr Nohr insisted in his evidence that the loss of Proctor & Gamble business caused redundancies which left KND without valuable skilled employees and therefore made it harder to win new business and retain existing clients, a point which I accept, both he and Mr Shaw would have been alive to this point when discussing whether it would be possible to “finger p&g” for any of the redundancies being discussed in 2012. (The context for these discussions was that the claimants were seeking to agree adjustments to the “earn out” calculation which would eliminate the effect of the loss of Proctor & Gamble business). Nevertheless, Mr Shaw thought that it would have been hard to make such a case. I see no reason to doubt that expression of opinion by a man who was aware of the arguments and had detailed knowledge of the business. It will therefore be necessary to be cautious in determining the extent to which redundancies and other cost-cutting measures were attributable to the loss of Proctor & Gamble business.

Further bad news was to come. On 27 June 2011 Leo Burnett advised that Tampax work was immediately being taken in-house in Chicago. Again Ms Davis observed that this would have “a revenue and staffing impact for London”. It was described as “a BAL decision”. It was suggested on behalf of the defendants that this was a decision taken on geographical grounds (a desire to centralise Tampax work in Chicago) and not because Leo Burnett as the BAL had decided to take work away from KND in order to do such work itself. However, the fact was that it was able to do this because of its position as the BAL regardless of the quality of KND’s work.

KND also lost the Eukanuba business for Procter & Gamble in November 2011. However, Eukanuba was not a BAL brand and this had nothing to do with the BAL model. It occurred because Digitas in the United States was unable to get on the approved supplier list for Procter & Gamble website work and, as a result, the Digitas agency worldwide (which included KND) was ineligible to compete for this work.

Finally, in February 2012 Publicis Worldwide announced that it was taking all Oral B work in-house as from 1 July 2012. Mr Kinsella commented in an email to Mr Nohr, Ms Davis and Mr Beringer that:

“Even though we could see this coming a year ago we are stunned at the lack of partnership and transparency. What P&G wants, and has asked for, is a better Groupe solution – give us the best team. What they are being given is the brand agency with some new digital hires.

Not a recipe for success – for either party.”

Neither Ms Davis nor Mr Beringer took issue with the suggestion that Digitas had “seen this coming a year ago”, that is to say before the date of the SPA. I find that they had indeed seen it coming.

However, it is important to note that there were other accounts where the loss of expected revenue had nothing to do with the Procter & Gamble BAL model. That applies to revenue from Procter & Gamble itself as well as from other clients. For example, and quite separately from the example of Eukanuba to which I have already referred, in January 2012 Procter & Gamble decided that in the year of the London Olympics its advertising budget would be shifted away from digital advertising in favour of printed media. Other clients also reduced their expected advertising spend. For example, clients in the financial services sector such as National Savings & Investments and Nationwide were tightening their belts in the then current financial climate. Work for other significant clients, such as Shell and Delta, was also lost for reasons unrelated to the BAL model, as Mr Nohr accepted. Indeed some clients, such as Delta, were lost for reasons unconnected with KND at all.

These developments led Mr Nohr to write to Mr Beringer and Mr Tomasulo on 4 March 2012. His email referred to the fact that KND’s latest forecast would involve the presentation of some “very challenging figures” consisting of a significant revenue downturn. He explained:

“This revenue decline is partly the result of the downturn in our UK financial services clients, all of whom have tightened their belts. But largely as a result of the contraction/loss of international clients (Samsung, Asus) and heritage Digitas business (P&G, BOA, Shell, Delta). By way of context, compared to 2011, the heritage Digitas clients have declined by £2.7m. And compared to forecast Samsung and Asus have declined by over £1.5m. Fortunately, the agency’s strength in new business has taken the edge of [sc. off] these revenue losses. But new business alone is not enough to scotch this trend.”

It is notable that the loss of Procter & Gamble business was described as merely one among a number of factors which had caused this situation. Mr Nohr did not suggest in this email that the loss of the Procter & Gamble business had caused the other problems, even indirectly. His email went on to describe two matters which followed from this situation. The first was a need to cut staff, the only question being how deeply to cut. He pointed out that cuts which were sufficient to restore KND’s Operating Income (that is to say, its profit margin) would seriously impede its ability to win new business and would require a fundamental restructuring. The second was the impact on the “earn out”. Here he pointed out that it was now unlikely that any Deferred Consideration would be achieved and that this would pose significant challenges in keeping the London management team focused. He proposed various ways in which this might be dealt with, either by amending the Deferred Consideration formula, extending the period during which it could be earned, or coming to a settlement with the claimants. He concluded:

“It is a cause of great concern to the London management team, who feel increasingly alarmed about the business we inherited and I would very much like to be able to reassure them that we are looking at solutions which will work for the group as well as the selling shareholders.”

This was the beginning of the dialogue which was in due course to lead to what the claimants say was the December 2012 agreement. Mr Nohr continued over the next few months to make these and similar points.

The senior management of Digitas was sympathetic to the claimants’ position, recognising that the loss of Procter & Gamble business was not the claimants’ fault and that it had occurred despite good work done for Procter & Gamble, initially by Digitas UK and subsequently by KND. As Mr Nohr reported in an email to his fellow claimants dated 6 March 2012:

“In short the business we have inherited appears, with the exception of Nissan, to be vulnerable in the extreme. And the people with whom we did that deal feel so bad about that they want to help make amends.

… we got the full support of the Digitas global team to put a revised deal to Maurice Levy (along with the reassurance that Levy has agreed to such revisions before).”

I accept, as already indicated, that Digitas’s senior management was sympathetic to the claimants’ position and wanted to make amends for the fact that the inherited Digitas UK business had proved to be vulnerable, but I do not accept the claimants’ submission that they “reacted with embarrassment and guilt”.

The concerns caused by the loss of BAL work were elevated within the Publicis group to Mr Lévy himself. On 8 March 2012 he expressed his own thoughts about this in an email to all the agencies concerned which referred to the BAL agencies having decided to “fire” Digitas on some accounts and take the work inside their own agencies. Among other things, he asked for information, including precisely why this had to be done. It is likely that each of the agencies replied to this email (a direct request by Mr Lévy would not have been ignored), but their replies were not in evidence.

By August 2012 no progress had been made and the claimants were becoming frustrated. They realised that in order for anything to be achieved, it would be necessary first to persuade the Publicis headquarters in Paris and ultimately to obtain the agreement of Mr Lévy. Senior employees within Digitas were sympathetic to the claimants’ position, recognising that the loss of Procter & Gamble business as a result of the actions of the BALs was not the claimants’ fault and was nothing to do with the quality of Digitas UK’s or KND’s work for Proctor and Gamble, which was of a high standard. Those senior employees were also concerned to maintain the motivation of Mr Nohr and his colleagues in growing the KND business, recognising that the earning of Deferred Consideration was a very significant factor for them. It was therefore in the interests of Digitas management to resolve this issue. For a combination of both these reasons senior Digitas executives such as Mr Tomasulo and Mr Bob Lord (who had replaced Ms Laura Lang as Global Chief Executive Officer of Digitas) assisted Mr Nohr in making a case to the Publicis M&A department in Paris, and ultimately to Mr Lévy, that an adjustment ought to be made. It was recognised by all concerned that any solution would need to be approved, or at any rate authorised, by Mr Lévy and that for this purpose the support of the M&A department in Paris (in particular Mr Dupont) would be important.

On 6 August 2012, Mr Tomasulo sent Mr Dupont a draft note. He praised the success of the merger in general but drew attention to “an unexpected and material event” which had “created an issue with the Earnout, which we must address”:

“The KN Acquisition to date has been a success in terms of the objectives we set out at the time of the acquisition. The combination of the KN CRM and branding background with the Digitas digital business has made an impact on the UK market: the business has won new business, been recognized within the UK market as an integrated marketing leader in both the award space and in terms of new client wins, and the management team has provided leadership to the RD International network. Financially, we have achieved the integration savings we expected and organic growth was trending where we anticipated. However, an unexpected and material event has created an issue with the Earnout, which we must address.”

On 4 September 2012 Mr Nohr sent an email to Mr Beringer reviewing the loss of (or at any rate the failure to achieve) the expected revenue from inherited Digitas UK clients and summarising the claimants’ argument:

“Our argument is that we entered into the merger in good faith, and were not told about any of the revenue threats, some of which would have been foreseen at the time. But rather than now invoking the warranties in the deal (10.11.a) and asking the Groupe why these revenue threats were not made clear to us, we would like to appeal to Paris’s commercial self-interest. By revising the terms of reference of the deal to once more make it viable, the Groupe can incentivise us to grow London revenues. We would then benefit from having an earn-out that works, and Digitas will benefit from having regrown its depleted London income.”

It is apparent that the claimants had in mind the possibility of a claim under the Buyer Warranty and wanted the defendants also to be aware of that possibility. Specific reference was made to clause 10.11(a). However, the claimants indicated that they would prefer to reach a negotiated agreement which would restore to them the prospect of earning Deferred Consideration (“having an earn-out that works”). This was expressly stated to be an alternative to a breach of warranty claim. No doubt the claimants took the view (and they were probably right) that invoking a legal claim for breach of warranty at this stage would sour relationships and render the negotiation of an amicable agreement difficult or even impossible. But, however veiled, the threat of a warranty claim if agreement could not be reached was unmistakably there. Mr Beringer must have understood this perfectly well. So too must the Publicis M&A team in Paris who would have realised that (as is common ground) there was no other basis in the SPA on which the claimants could seek the adjustments which they were seeking. Accordingly it must have been apparent to all concerned that, in addition to the important commercial advantages which would ensue if agreement could be reached, the parties were negotiating to avoid the necessity for a possible warranty claim and that, if agreement were reached, that would preclude the possibility of any such claim in the future.

The impact which a failure to address this issue could have on KND was described by Mr Beringer in a commentary sent to Mr Tomasulo on 14 September 2012:

“Due to the loss of P&G and subsequently the exit of the core digital team there is a real risk the business will default back to what KND was prior to the merger; however, that part of the industry (traditional CRM [Customer Relationship Management]) is in decline.

The motivation and the moral [sic.] of the agency leadership is now at risk too given the delay of the EO adjustment process and the fear that there will not be a resolution. Marc Nohr is expecting first resignations in Q4 should the situation not be resolved. The word on the ‘floor’ is that the agency got ‘screwed’ by Publicis.”

Getting “screwed by Publicis” was a crude way of referring to the fact that the claimants had not been told things that they ought to have been told before the merger took place. Again, therefore, the possibility of a warranty claim formed part of the background to these discussions. Any agreement reached was plainly intended to avoid the necessity for such a claim.

Mr Nohr was also in direct contact with Mr Lévy about the “earn out” issue. On 1 October 2012 he had summarised the situation in an email to Mr Lévy personally, describing the “huge fallaway in Digitas inherited income” as being “principally (but not entirely) the result of the loss of P&G to ATL agencies within the network”, the result of which (among other things) was that “our earn-out is no longer achievable”, and asking for Mr Lévy’s guidance on a range of issues. Mr Lévy’s response was that he was not aware of the “discussions on earn out” and did not wish to become involved in that issue as he did not wish “to cut through the organisation”. Mr Nohr responded on 5 October 2012 that he would look to Mr Lord (who had advised that he had discussed the matter with Mr Lévy) to discuss and resolve these issues and asked Mr Lévy for his support, in the event that Mr Lord raised the issue with him, in ensuring that Mr Nohr could keep the KND leadership team focused and incentivised.

On 2 October 2012, Mr Tomasulo sent Mr Dupont an email setting out recommendations for dealing with this issue, together with supporting calculations, and requesting that the matter be raised with Mr Lévy on the following day. He explained that “the impact of the loss of P&G on the KND acquisition in the UK has had a material negative impact on the business as a whole”. The language of “material negative impact” echoes the language of clause 10.11(a) (“material adverse impact”) and suggests that, on the Publicis side, the possibility of a breach of warranty claim if agreement could not be reached was well in mind.

The matter was raised with Mr Lévy as requested. On 3 October 2012 Mr Lord sent an email to Mr Tomasulo and Mr Dupont, reporting on this discussion:

“I discussed this issue briefly with ML [Mr Lévy] today and he asked me to construct an email with a recommendation? Johann [Mr Dupont] do you support what Joe [Mr Tomasulo] is proposing? It would be great if we all agreed on the approach before I construct a recommendation to ML? I’d really like to close this issue ASAP for the team’s sake.”

On 4 October 2012, Mr Tomasulo sent an email to Mr Dupont responding to a number of questions from Mr Dupont earlier that day:

“- P&G work was taken away from KND by the BALs (PWW, Saatchi, LB) effective 1 July 2011 as P&G is on a June 30 Fiscal year. The KND scores at P&G were good. …

-

We have restructured what we could (client facing talent) and we will reflect that savings into 2013 (no severance, lower PC cost for talent we needed to exit) but losing the high level of revenue represented by P&G materially impacts our overhead leverage (rent, non-billable talent, mgmt) and profitability as P&G was a profitable business so it is these adjustments that will continue.

-

The P&G impact disturbed the post acquisition KND business significantly. As part of the integration, KND restructured its business around the large scaled P&G client and losing it was disruptive and costly. Given the materiality, they had to stop moving forward and deal with this loss so it cost them time and effort not to mention undoing the positioning they had spent months putting in place. While we could have quantified that cost (and KND asked), in the end we agreed with KND that it was best to focus the adjustments on P&G alone.”

It is apparent from this email that the impact of the loss of Procter & Gamble business was wider than merely the loss of revenue and that, whatever other factors may have been operating, it had required significant restructuring of the KND business. There is no reason to doubt the accuracy of what Mr Tomasulo said about this. He accepted in evidence that it had hampered the claimants’ ability to grow and run KND as a healthy business.

On 8 October 2012 Mr Tomasulo confirmed again, in response to a further question from Mr Dupont, that “the BAL leaders (LB, Saatchi, PWW) pulled the business, we were not fired by PG”.

On 15 October 2012 Mr Dupont sent a report to Mr Lévy, in response to which Mr Lévy acknowledged that the loss of Procter & Gamble business was not KND’s fault but was the result of the actions of other agencies within the Publicis group. He indicated that he was willing to consider an agreed solution, but wanted to examine matters further.

Mr Dupont reported this to Mr Tomasulo in an email sent on 17 October 2012:

“I received the feedback from Maurice. If he agreed on the adjustment concept, he wants to make sure that the calculations are accurate. I will share with him the assumptions and hopefully, that’s enough for him to give him comfort. One other request is that he wants some proof that the acquisition was a success before the P&G issue. Can you please provide input/argumentation on this last piece?”

As a result Mr Nohr provided some information to Mr Tomasulo, highlighting successes which KND had achieved since the merger, which was sent on to Mr Dupont. The result was an email dated 26 October 2012, sent by Mr Dupont to Mr Nohr (copying, among others, Mr Tomasulo and Mr Lord). Mr Dupont explained that he had discussed the issue with Mr Lévy, who had given his agreement to the principle that an adjustment of the sort that had been discussed by the parties over the preceding weeks and months would be made to the calculation of the claimants’ earn-out:

“I discussed the issue with Maurice and the conclusion is the following: the principle of an adjustment of Kitcatt Nohr Digitas P&L to carve-out P&G impact is accepted. The process will be as follows: when 2013 financial accounts are finalized, you will submit a revised P&L for 2012 and 2013 in collaboration with Joe/Digitas team, with relevant substantiation of the figures. We will then review and audit the submitted numbers. Would that be agreeable to you?”

There is no reason to doubt that this was an accurate account of what Mr Lévy had agreed. I find that it was and was so understood. There was, therefore, agreement in principle by Mr Lévy to an adjustment to the formula by which the Deferred Consideration was to be calculated. The adjustment was to “carve out” the “P&G impact”. In the context of the parties’ previous discussions, this referred to the impact of the loss of Procter & Gamble business as a result of the decisions by the BALs within the Publicis group to take that business away from KND and to retain it themselves. The adjustment was to cover both years, 2012 and 2013. However, the precise way in which it would work was to be left for further discussion after the 2013 financial accounts had been finalised. The email concluded with a question whether this process was acceptable to the claimants.

The claimants’ case is that at this point a binding agreement was concluded that reasonable adjustments would be made to the calculation of their entitlement to Deferred Consideration under the SPA in order to ensure that this entitlement would not be prejudiced by KND’s loss of Procter & Gamble work. They do not say that agreement was reached at this stage on the precise adjustments that would be made, but they do say that as a matter of law a binding contract was concluded that reasonable adjustments would be made.

The claimants did not give an affirmative answer to the question with which Mr Dupont’s email had concluded. Although Mr Nohr was pleased that the principle of an adjustment had been accepted, he wanted to go further. As he put it in evidence, it “was encouraging but it was not sufficient detail for us”. By an email to Mr Tomasulo and Mr Lord dated 28 October 2012 Mr Nohr indicated that he was encouraged by Mr Lévy’s agreement, but concerned that “the decision is only ‘in principle’.” He said that the claimants’ position would be better protected if the defendants agreed not only the principle that adjustments would be made, but also the precise adjustments that would be made. In that way, he said, if KND performed well in 2013 and submitted robust figures to Paris, “the deal will hold”. Mr Lord and Mr Tomasulo supported this approach.

Accordingly on 30 October 2012 Mr Nohr replied to Mr Dupont (copying Mr Tomasulo among others):

“I would like to clarify one matter if I may before we proceed.

We would like to avoid any surprises down the line, on either side. For that reason we have laid out the assumptions behind our model in the documentation we sent. We would therefore like your endorsement of the assumptions, so that at the conclusion of 2013 we can simply submit our numbers for you to audit without any concerns about the basis of calculation. …”

On 31 October 2012, Mr Dupont replied, stating that although the procedure he had suggested in his email of 26 October 2012 had contemplated the review of both figures and assumptions at the end of the process, the assumptions provided by Mr Tomasulo seemed generally acceptable, with the exception of one assumption relating to the treatment of overheads in 2013. However, it is apparent that he continued at this stage to envisage a process whereby the detailed basis on which any adjustments would be made would only be determined after the 2013 figures were available.

Mr Nohr was not happy about this. He indicated to Mr Lord and Mr Tomasulo that the idea of leaving matters unresolved for another 15 months would be unacceptable to the KND leadership team. As he put it in one email to Mr Lord and Mr Tomasulo:

“… For reasons I know you both understand, I can’t tell the leadership team that we will work it through in early 2014 – as they want to know whether or not the earn-out remains viable.”

Mr Tomasulo explained to Mr Dupont that this further clarity was critical to the KND team and, over the following weeks, continued to discuss the detail of the proposed adjustments with him. On 18 November 2012 Mr Nohr pressed Mr Tomasulo for news of progress, adding that he was “hoping we could have this wrapped up by now – but certainly this side of Xmas if we are to avoid team fallout”. Mr Tomasulo replied on the same day that:

“Johann [Dupont] is talking to Maurice [Lévy] about the adjustments, he agreed with our position and we found another positive one…”

Mr Nohr was anxious to pin this down. He asked:

“When you say ‘he agreed with our position’ what does that mean precisely?”

The answer from Mr Dupont was that so far as overheads and other costs were concerned, “it should be ok to apply the methods suggested by Joe [Mr Tomasulo] in 2013”. Mr Nohr commented that movement from an “in principle” agreement to “it should be ok” to apply the assumptions provided represented “real progress”, and added that the ideal would be to achieve an unambiguous and documented agreement. He chased again on 22 November 2012:

“I am sorry to chase you further – as I feel like I have been doing it since March. But I’d love to close this down and would value your advice on how to get this codified with Paris rather than rely on turns of phrase in informal emails with Johann. …”

The advice from Mr Lord and Mr Tomasulo was that Mr Nohr should contact Mr Dupont directly, with a note which could if necessary be forwarded to Mr Lévy. Accordingly Mr Nohr wrote to Mr Dupont (copying Mr Lord and Mr Tomasulo) on 22 November 2012 as follows:

“If I understand your last few emails correctly, we have an agreement in principle, and the method is also OK with you. So just for the avoidance of doubt what I would propose is that next week our Financial Director pulls together a list of all the key adjustment principles we have used in our calculations and send them to you for your formal approval.

There should be no surprises in that list, but once you have formally approved them we can put the list away until we submit our 2013 figures, confident in the knowledge that all the assumptions are fully agreed and that all we need to do is focus our leadership team on growing the business.”

It was apparent, therefore, that what Mr Nohr was asking for was formal agreement on the adjustments which should be made which would cover both years, 2012 and 2013. Mr Dupont replied to Mr Nohr on that same day, agreeing with the course that Mr Nohr had proposed:

“Ok, please ask your CFO to provide such a list and I’ll give you formal feedback on it.”

As it happened Mr Nohr conducted an interview with Mr Lévy at an event on 30 November 2012. At the conclusion of the interview Mr Lévy raised the subject of the earn-out, asking if things were progressing with Mr Dupont. Mr Nohr said that they were and that he hoped for a resolution soon. Mr Lévy said that he trusted Mr Dupont to sort it out. It is evident that Mr Lévy was not averse to the idea of concluding an agreement along the lines which Mr Nohr was seeking and that he had confidence in Mr Dupont’s handling of the issue. Mr Nohr reported this conversation to Mr Lord.

On the advice of Mr Lord, Mr Nohr emailed Mr Lévy on 4 December 2012 to follow up this conversation:

“I forgot to mention – I appreciate you bringing up the P&G issue. You will know from previous correspondence that it sits at the heart of whether we can make our earn-out work. So, as you suggested, I will liaise with Johann [Dupont] to lock down the issue and hope we can bring this to a resolution soon.”

Mr Lévy replied on the same day, indicating that he knew that Mr Dupont was working on the issue of what adjustments should be made:

“I know Johann [Dupont] is working seriously on the earn-out issue and P&G.”

On 6 December 2012, Mr Nohr sent Mr Dupont (copying Mr Tomasulo) two documents setting out some very detailed assumptions that the claimants sought to agree. However, these went further than dealing only with the loss of Procter & Gamble business. Mr Dupont was not at all impressed with the claimants’ proposals. On 12 December 2012 he wrote to Mr Tomasulo making clear what was needed:

“Did you review this? Half of the content is unnecessary or off topic. The intention was to agree on key principles, not to make assumptions on calculation of the earn-out, gap in OI to get to the 10% or estimated 2013 figures. Can you please coordinate with Marc [Mr Nohr] to clean this up and come back with a simple document where we can say yes, these will be the principles which we will apply when calculating 2012 and 2013 OI (in the spirit of your assumption sheet in your initial excel file)? …”

Mr Tomasulo responded on the following day, agreeing that what needed to be dealt with was the Procter & Gamble issue.

Accordingly there was agreement between the claimants and Mr Tomasulo on the one hand and Mr Dupont on the other as to what they were seeking to achieve – namely, a statement of the principles which would apply for the calculation of Deferred Consideration which took account of (and only took account of) the loss of Proctor & Gamble business as a result of the conduct of the BALs.

Finally, after further discussions, on 17 December 2012 Mr Nohr and Mr Tomasulo spoke on the telephone. Mr Tomasulo confirmed that the adjustments that the claimants had now submitted were agreed, at any rate by Mr Dupont. This much appears to be common ground. However, it is in dispute whether what was agreed by Mr Dupont had been approved or authorised by Mr Lévy or whether (as Mr Tomasulo maintained in evidence) all that had been achieved was an agreement by Mr Dupont that these adjustments were acceptable, which was understood to be subject to obtaining approval from Mr Lévy.

On the same day Mr Nohr sent an email to the former KNAS shareholders, confirming that agreement had been reached:

“I am pleased to be able to share the news that after 9 months of discussion we appear to have a way forward on the earn-out.

Paris have agreed both ‘in principle’ and ‘in practice’ to the reworking of the figures to compensate us for the loss of P&G. And this has been endorsed by the M&A team and Maurice [Lévy], whom I have spoken to in person.

You will know that the loss of this huge piece of business in Q1 pretty much made the earn-out impossible to achieve. But by agreeing to allow us to discount the revenue (and some of the associated costs) the deal becomes viable again. …

What next?

We have sent a theoretical model to Paris as the basis for the new arrangement. But in a few weeks we will submit our actual 2012 figures, along with the computations which show this P&G ‘subsidy’ and we [will] be looking for formal approval of those figures. We will then do the same thing again in early 2013 [sc. 2014]. In both cases the figures will be audited. …”

Mr Nohr’s reference to speaking to Mr Lévy personally was a reference to the interview event of the previous month. He had not spoken to him more recently. Clearly, however his understanding was that the agreement reached had been approved by Mr Lévy.

On the following day, 18 December 2012 Mr Nohr sent an email to Mr Tomasulo confirming his understanding of their conversation:

“Thanks for your time on the phone yesterday.

We are delighted that together with Johann [Dupont] you are comfortable with the principle of the P&G adjustment and the way we have calculated it. We will now submit the 2012 actuals in January, applying the agreed P&G adjustment and hope that this gets the green light as discussed.

We have shared this news with shareholders – all of whom are delighted at the progress after such a long period of discussion and all of whom are appreciative of your support.”

Also on 18 December 2012, Mr Tomasulo forwarded Mr Nohr’s email to Mr Lord who responded by congratulating Mr Tomasulo:

“Nice job Joe getting this across the line…”.

Mr Tomasulo’s evidence was that at this point they had succeeded in persuading Mr Dupont to agree to the assumptions proposed by the claimants, but that it was still necessary for these proposals to be taken to Mr Lévy for his agreement. He insisted that the only “line” which had been crossed was obtaining the agreement of Mr Dupont which, as everybody understood, did not bind anyone. He said that it was for Mr Nohr to take the matter forward to obtain Mr Lévy’s agreement. Without that agreement, what had been achieved so far was “meaningless”.

The claimants made no further attempt to obtain agreement to the adjustments. In particular, they did not approach Mr Lévy to obtain his agreement. Nor did they pursue the suggestion, which had been raised at an earlier stage, of a formal amendment to the SPA. It is apparent that the claimants believed that agreement had been reached and that nothing further was required. They would not otherwise have let the matter rest as they did.

During the later part of 2012 the Publicis group was considering a new acquisition. This was the acquisition of LBi, a digital agency which was much greater in size than KND and with which it was proposed that KND would be merged. This created further issues for the claimants’ earn out as they would now become part of a much larger agency to which it would be difficult to apply the terms of the SPA. The claimants did not relish becoming part of this larger agency in which KND would lose its distinctive identity. Moreover, the ability of KND to win and retain clients was affected by speculation about its future. In the event the acquisition of LBi was announced in early 2013 and led to the departure of Mr Shaw, who saw no role for himself in the new organisation.

On 29 May 2013 Mr Nohr forwarded a letter to Mr Lord and Mr Tomasulo which was described as representing the collective view of the claimants who were said to be concerned about events which they believed had put the earn out beyond reach. The letter complained that the possibility of a successful earn out had been fatally damaged by circumstances beyond their control, for which they sought recognition. Two such matters were identified, the first being “the loss of the Procter & Gamble income” and the second “the impact of the acquisition of LBi”. So far as the first was concerned, the letter stated that:

“… Suffice to say that prior to the conclusion of the Share Purchase Agreement we had been told that the P&G business was growing and there were no concerns about it – as was confirmed by the warranties in the SPA (which are arguably breached). Whereas after the merger it became clear that the P&G business in Digitas was under threat due to global realignment agreements within Publicis. …

Some compensation for this has been discussed (though still not actually agreed).”

The letter identified a number of factors which, it was said, would need to be factored in to any compensation. After referring to what was said to be the damaging consequences for KND of the LBi acquisition, the letter concluded:

“We believe that the remedies which have been discussed in relation to the P&G business are no longer viable, and that this should be acknowledged by both sides. To move forward, we can see two potential remedies: …”

The remedies identified were either negotiation of a settlement sum or an extension or restructuring of the earn out period.

Mr Nohr was asked about this letter. He did not accept the suggestion put to him in cross examination on behalf of the defendants that the claimants by this time wished to have an entirely different negotiation as a result of the proposed acquisition of LBi. He agreed, inevitably, that there was no suggestion in the letter that any agreement had already been reached. He agreed also that this was a letter which, even though drafted by Mr Shaw, he would have seen and approved. His explanation of the statement that compensation had been discussed but not agreed was that:

“It says that compensation for this has been discussed although not actually agreed. What, I think, is clear is that what had been agreed was the formula by which to arrive at it, but it would not be complete until we submitted the figures and they were in the financial years that were relevant. … A precise figure would not have been agreed. It could not have been agreed. All that was agreed was the 4 million and the assumptions. … I know what you are suggesting and I do not agree, but I think this is not a particularly useful sentence because it lacks precision.”

I do not accept this explanation. The letter is careful to say that nothing had been agreed. That is directly contrary to the claimants’ case in this action. Although Mr Nohr would not accept the suggestion put to him that the claimants by this time wished to have an entirely different negotiation as a result of the proposed acquisition of LBi, that explanation seems to me to be clearly correct. By this time the claimants wanted more than they had been seeking in December 2012. As the letter makes clear, they wanted to be compensated not only for the loss of Procter & Gamble business but also for the impact of the acquisition by Publicis of LBi.

Mr Lévy’s reaction, in an email sent to Mr Nohr on 25 June 2013, was that nothing should be done to change anything, so as to give the claimants every chance to earn the maximum earn out. He added that:

“To date we only have one pending issue which is the P&G account loss and we should not do anything which could impact your ability to earn the price complement.”

Mr Shaw then set out his views, in an email dated 3 July 2013. He complained, among other things, that although KND had provided adjusted accounts for 2012 at the beginning of the year, reflecting the impact of the loss of Procter & Gamble, it had so far had no response.

The end result of these further communications was that Mr Stephane Estryn of the Publicis M&A department in Paris (who was now dealing with the matter as Mr Dupont had left the Publicis group in April 2013) advised that Mr Lévy would make a decision on what adjustments should be made to take account of the loss of Procter & Gamble business after the end of 2013.

In January 2014, by which time KND had been merged into LBi and the combined agency was known as Digitas LBi, Mr Nohr decided that it was time for him to move on. He asked Mr Beringer to provide him with a reference and Mr Beringer did so. At this time Mr Nohr made two comments about the December 2012 agreement. The first was a text sent to Mr Beringer in which Mr Nohr stated that:

“We ended the year doing out [sc. our] numbers so if Maurice honours the adjustment he promised, the earn-out can still happen.”

The second was an email dated 31 January 2014 thanking Mr Beringer for his reference. Mr Nohr said that:

“If the chance offers itself to put in a good word with Maurice as he decides whether to honour the P&G correction agreed by you and Joe Tomasulo in the coming months I’d really appreciate it.”

The claimants relied on the first of these messages, which refers to an adjustment having been promised by Mr Lévy. The defendants relied on the second, which refers only to an agreement by Mr Beringer and Mr Tomasulo. I do not accept that these messages contradict each other quite so starkly. Nor do I accept the defendants’ suggestion that the second message had been deliberately withheld from disclosure by the claimants when the first message, the text, was disclosed during the course of the trial. So far as Mr Nohr was concerned, both ways of describing what had happened were correct. He believed that Mr Lévy had promised that adjustments would be made. But he also knew that this correction had been agreed by Mr Beringer and Mr Tomasulo, albeit on the basis that ultimately it was Mr Lévy’s agreement which mattered.

The reference which Mr Beringer provided was in glowing terms, praising Mr Nohr’s leadership of KND and his successes in winning new business and making the most of business opportunities. In evidence, however, Mr Beringer repudiated much of the content of this reference, not merely by referring to it as a “stretch” but actively criticising Mr Nohr as a poor leader who had failed to nurture client relationships and whose performance at KND had been in many respects disappointing. While I recognise the tendency of a reference writer to make the most (and sometimes rather more than the most) of an individual’s good points, the stark contradiction between the content of the reference and Mr Beringer’s evidence goes much further than this. Mr Nohr had a successful track record at KNAS (which was of course part of the reason why Publicis decided to acquire KNAS in the first place) and contemporaneous documents describe the successes which KND had achieved, for example Mr Tomasulo’s note of 6 August 2012 quoted above. In my judgment that material represents a fairer summary of Mr Nohr’s ability and achievements at KND. Mr Beringer’s determination to do Mr Nohr down calls the reliability of his evidence into question.

In early 2014 the claimants submitted a draft earn out calculation. This was forwarded to Mr Estryn by Ms Amy Murphy-Dowd (Mr Tomasulo’s successor, he having also left the group) with the comment that:

“Based on information that Joe Tomasulo shared with me before he left regarding the calculation, I believe that spreadsheet fairly represents the spirit of the agreements previously made with you relating to earnout adjustments.”

This was later supported by an email dated 1 July 2014 reiterating that such adjustments had been agreed. However, Mr Estryn’s response, which has remained the defendants’ position ever since, was that no Deferred Consideration was payable in accordance with the terms of the SPA, there was no breach of warranty by MMS, and no agreement had been concluded for adjustments to be made to the contractual formula.

I turn now to determine the issues in the case, beginning with the construction of clause 10.11(a). For ease of reference I set out its terms again:

“The Buyer warrants to the Sellers that save as disclosed by the Buyer to the Sellers from time to time prior to Completion each of the following warranties (‘Buyer Warranties’) is true and accurate at Completion:

(a)

none of the Buyer, Stephan Beringer, Joseph Tomasulo or Charlotte Frijns is aware of any facts or circumstances that could reasonably be expected to have a material adverse impact upon the Operating Income and/or Revenue in 2012 or 2013 (being a reduction of at least 20% in the case of Operating Income and 10% in the case of Revenue) including, without limitation:

(i)

the resignation or expected loss of any client of Digitas;

(ii)

any significant current or threatened litigation involving Digitas.”

I draw attention to the following points, which for the most part were not controversial and in any event seem to me to be clear:

Clause 10.11(a) applies only in the event of a reasonable expectation of a “material adverse impact” on Operating Income and/or Revenue in 2012 or 2013, which impact is defined as a percentage reduction of at least 20% or 10% respectively. In other words, there is only a breach if the facts not disclosed could reasonably be expected to cause a reduction of 20% of Operating Income or 10% of Revenue. The question arises, however, a reduction from what base? What comparison is required by the clause to be made in order to ascertain whether the warranty is broken?

It is common ground that in order for the clause to operate, it must be possible to determine at the date of the SPA whether there has been a breach. MMS must know, or at any rate have the means to determine, what disclosure it is obliged to make. That cannot be left to hindsight after the expiry of the years in question.

The claimants say that the relevant comparison is between:

The defendants say that the clause provides no “baselines” which enable any comparison be made and that it is therefore unenforceable. If there is no baseline, it is impossible to say whether there is, or even whether there is a reasonable expectation of, a reduction from the baseline of the specified percentage. This argument, as I understand it, arises because “Revenue” and “Operating Income” are defined terms in the SPA which refer to the actual Revenue and Operating Income in any given year (after making the accounting adjustments required by the definition in question). Thus “2012 Revenue” and “2013 Revenue” are defined to mean the “Revenue” of KND for the years in question, with “Revenue” itself being defined to refer to the actual Revenue for the year. Similarly in the case of “Operating Income”.

This means, according to the defendants, that the clause must be read as referring to a 20% or 10% reduction in what turns out to be the actual Operating Income or Revenue in 2012 or 2013, but that there is nothing to indicate the baseline by reference to which the reduction has to be calculated. They say further that even if a calculation were possible, it could only be undertaken after the expiry of the years in question, so that it would be impossible to say until after that time whether there had been a breach of the warranty back in March 2011 at the date of conclusion of the SPA, a consequence which cannot have been intended.

The defendants rely on Prophet Plc v Huggett [2014] EWCA Civ 1013 as illustrating that it is perfectly possible for what appears to be a carefully drafted clause to have no substantive content. That was a case where a proviso to a restrictive covenant in an employment contract which limited the restriction to work on the former employer’s own products had the effect that the clause did not prevent the employee from doing anything for his new employer whose products, although competitive, were not the products on which the employee had worked for the former employer. Rimer LJ accepted at [33] that if faced with an ambiguous clause, one interpretation of which gave rise to an absurdity while the other interpretation was commercially sensible, a court would be likely to favour the latter, but held that the clause in question was not ambiguous. He went on to say:

“35.

… I regard the drafting of the proviso as unambiguously clear and I have said what its sense is. I also agree that it is likely to have little or no commercial effect, and thus to leave Prophet with, for all practical purposes, a toothless restrictive covenant. I am not, however, persuaded that this is a case in which it is clear that something has 'gone wrong' with the drafting of the proviso. A vital part of the context in which it falls to be construed is that it was drafted by someone who was plainly sensitive to the likely voidness of the first sentence of clause 19 if it stood alone; and sensitive, therefore, to the need for the proviso to be drawn very tightly with a view to ensuring that no wider post-employment restriction was imposed on Mr Huggett than was reasonably necessary. I would therefore approach the proviso on the basis that it was a carefully drawn piece of legal prose in which the draftsman chose his words with deliberate and specific care. That is exactly what one would expect in the drafting of a restrictive covenant; and the fact that the proviso was tacked on to clause 19 with the plain intention of saving the validity of the clause as a whole merely serves to underline that.

36.

I do not, therefore, accept that this is a case in which the court can be confident that something has 'gone wrong' with the drafting. On the contrary, I regard the words of the proviso as reflecting exactly what the draftsman intended. Where, I accept, something probably did go wrong is that the draftsman did not think through to what extent his chosen restriction would be likely to achieve any practical benefit to Prophet upon Mr Huggett's departure to a competitor. In other words, he did not think through the concept underlying his chosen words. If he had done so, and had realised the potential practical futility of those words, I apprehend that he would have started again. As it is, I consider that it is not possible to read the proviso and conclude from it that, although it actually achieved result A, it is clear from its language as a whole, read in its context, that the draftsman really intended to achieve different result B – or, for that matter, C, D or E.”

The question for decision in this case is a question of construction – whether, properly construed, the clause provides a baseline which enables a meaningful comparison to be made when applying the clause’s definition of materiality. The modern approach to such questions was conveniently summarised by the Supreme Court in Arnold v Britton [2015] UKSC 36, [2015] AC 1619 at [15]:

“When interpreting a written contract, the court is concerned to identify the intention of the parties by reference to ‘what a reasonable person having all the background knowledge which would have been available to the parties would have understood them to be using the language in the contract to mean’, to quote Lord Hoffmann in Chartbrook Ltd v Persimmon Homes [2009] AC 1101, para 14. And it does so by focussing on the meaning of the relevant words … in their documentary, factual and commercial context. That meaning has to be assessed in the light of (i) the natural and ordinary meaning of the clause, (ii) any other relevant provisions of the [contract], (iii) the overall purpose of the clause and the [contract], (iv) the facts and circumstances known or assumed by the parties at the time that the document was executed, and (v) commercial common sense, but (vi) disregarding subjective evidence of any party’s intentions”.

The relevant background knowledge and overall purpose of the clause in the present case included the facts that:

In these circumstances there is in my judgment no difficulty in construing the clause, which is directly concerned with the reasonable expectations of the parties, as providing for a comparison between what would reasonably be expected on the basis of the information in fact provided and what would reasonably be expected if the facts and circumstances in question had been disclosed. That is the natural meaning of the clause which gives effect to its overall purpose and accords with commercial common sense.

The difficulties relied on by the defendants are caused entirely by a literal (I would respectfully say over literal) application of the definitions in Schedule 5 of the SPA to the bracketed words in clause 10.11(a) which themselves constitute a definition of materiality for the purpose of the clause. Mr Nigel Jones QC for the defendants accepted that if the bracketed words were not there, the problem of absence of a baseline would not arise. In that event the clause would simply refer to “a material adverse impact” but without defining what kind of impact counted as “material”, although its reference to the expected loss of a client would provide some guidance in that respect. (Mr Jones suggested that such a clause would be void for uncertainty, but I would not necessarily accept that: deciding what counts as “material” would involve an exercise of judgment and might in a marginal case be a point on which views could reasonably differ, but it would be an extreme conclusion to say that the absence of a definition rendered the clause void; however, that is not a point which need be decided).

However, it was plainly the intention of the parties that all three paragraphs of the Buyer Warranty would be enforceable and that it would be possible to ascertain whether there had been a breach as at the date of conclusion of the SPA and not merely after the expiry of 2012 and 2013. If incorporating the definitions in Schedule 5 into the terms of the warranty (and thereby reading it as referring to what turns out in due course to be Operating Income and Revenue in 2012 for 2012 or 2013) renders the clause unenforceable and unworkable, that is a strong indication that the parties did not intend this to be done. A definition should be the servant of clarity, not a dictator of absurdity.

Indeed, the “Definitions and Interpretation” clause of the SPA is introduced by the words:

“In this agreement the following words and expressions have the following meanings unless the context requires otherwise: …”

In clause 10.11(a) the context clearly does require otherwise.

I would accept that there may be cases, albeit probably rare, where an unambiguous clause is devoid of meaning or content. Prophet Plc v Huggett provides an illustration although, as Rimer LJ observed at [35], “a vital part of the context” in that case was the fact that restrictive covenants in employment contracts will sometimes be unenforceable as a matter of public policy and that drafting efforts to save such a clause by limiting its scope may have the effect of drawing its teeth entirely. That kind of consideration does not apply here.

I would be reluctant to conclude that a clause which was obviously intended to provide an important protection to one party and which underpinned the whole structure of the deal (including the allocation of Initial and Deferred Consideration) was meaningless and therefore unenforceable. That would in my view be a last resort (cf. the cases cited by Leggatt J in Astor Management AG v Atalaya Mining Plc [2017] EWHC 425 (Comm) at [64] and [65]). As it is, however, there is no need to reach such a conclusion or to resort to special rules of construction (e.g. supplying additional words to make sense of an otherwise nonsensical clause) in order to avoid having to do so. The clause is well capable of being construed as I have indicated and to do so gives effect to its natural intended meaning.

It is therefore necessary to consider what the parties’ expectations were as to the Operating Income and Revenue for 2012 and 2013 as at the date of the SPA. The starting point must be to identify their expectations for 2011. The warranty in clause 10.11(c) refers to “the financial projections for Digitas for the period ending 31 December 2011”, although that reference is to a document previously prepared for Digitas UK alone, not for the merged agency KND. Obviously the intention of all concerned was that KND would be greater than the sum of its parts and, under the leadership of the former KNAS managers, would grow accordingly.

Provision for the merger was made by clause 13.1 of the SPA which refers to an “Integration Plan”:

“The parties agree and acknowledge that as soon as reasonably practicable following the Completion Date the Company and Digitas and Duke will be operationally merged to form the Kitcatt Nohr Digitas Division in accordance with the Integration Plan.”

The Integration Plan was a document dated 16 February 2011 which was prepared with input from both Digitas UK and KNAS. It was a mutually agreed document. It comprised a series of slides which dealt with a variety of matters. One slide described the “Objectives of the Combined Businesses”. These were to “create a top tier integrated agency in the UK with digital at its core by combining KNAS and Digitas UK”. The “unique strength” of the new company was to be “created by a new combination of skills essential for the development of mass marketing solutions in a post mass media age”. Another slide, prepared by Mr Tomasulo, was headed “2011 Financials”. It included the following statements:

“s 2011 combined growth projected at 14.8%

s Projected 2011 Revenue – At present, 78% of 2011 Kitcatt Nohr Revenue is visible from current clients and 6% from new business won in Dec, 2010. The balance of 16% of its 2011 Revenue is expected from conversion of new business supported by its current pipeline. …

s We continue to support our 2012 and 2013 predictions out in the PTE [Preliminary Target Evaluation] of combined growth of 15% and average margin of 16%.”

Although this particular slide was lacking in detail, it was underpinned by the figures which the parties had shared. It represents a clear statement of their shared expectations and aspirations not only for 2011, but also for 2012 and 2013. I accept the claimants’ submission that it comprises the best evidence of the expected Revenue and Operating Income for 2012 and 2013 at the date of the SPA.

The defendants’ witnesses emphasised that it was critical for them to put forward projections in which they had confidence. Mr Beringer, for example, insisted that it was not in the interests of those individuals who recommended the acquisition of KNAS to recommend an acquisition which carried any risk of failure as they would be accountable (and their careers would be likely to suffer) in the event of any failure to meet the projections in the Integration Plan. Moreover, there would be no benefit to the defendants if, as a result of including unrealistic projections, the claimants had little or no prospect of earning any Deferred Consideration. That would merely leave the defendants with a demotivated senior management team at KND. Mr Tomasulo gave similar evidence.

It is true that the claimants’ initial pleading of their claim did not identify the Integration Plan as representing the parties’ (or their own) expectations at the time of the SPA. Initially they relied on projections prepared at a slightly earlier stage by their own advisers, SI Partners, which projected Deferred Consideration of up to £10 million depending on how KND performed. However, their expert Mr Cottle did not support reliance on these projections and, at a late stage before trial, the claimants amended their pleading to rely on the Integration Plan. The defendants made much of this change of case. However, the fact that the claimants only advanced this case at a relatively late stage does not detract from the fact that the Integration Plan represents the best evidence of the parties’ expectations, as well as having some contractual status in accordance with clause 13.1 of the SPA.

The accountancy experts agreed that the Integration Plan provided contemporaneous information as to the parties’ expectations as at the date of the SPA and that, in accordance with this plan, their expectations can be summarised as follows:

I find that this represented an accurate statement of the parties’ expectations at the date of the SPA and that these expectations were reasonable.

As already indicated, the facts or circumstances relied on by the claimants as giving rise to the claim for breach of warranty are:

The defendants do not accept that this is a fair characterisation of the position which actually existed. They say that the BAL model provided advantages to Digitas UK and that it was no less secure than any project based client relationship which, in a competitive market, might well come to an end at the conclusion of the project in question. Nor do they accept that such a deliberate strategy was being implemented by the BALs. Rather, all that had happened was that the Ariel account had been lost and that Procter & Gamble itself was interested in a more integrated approach.

I accept that the BAL model brought advantages as well as disadvantages. The way in which it had been operated in the past meant that Digitas UK had obtained access to some digital work effectively subcontracted to it by the BALs, which in all likelihood it would not otherwise have obtained. Clearly this was a benefit. Nevertheless, as events were to demonstrate, Digitas UK (and subsequently KND) was vulnerable to a change of attitude on the part of the BALs. In each case it was the BAL which in practice controlled the relationship with the client, Proctor & Gamble. The BAL was able, by virtue of that relationship, to ensure that Digitas UK was cut out of any future work on the brand in question. This vulnerability went beyond the ordinary risk that, once a project was completed, there might be no further work for that particular client. In such ordinary circumstances the advertising agency concerned would have the opportunity, by developing and nurturing its relationship with the client and by delivering work of high quality during the currency of the project, to put itself in a strong position to obtain new work once the current project was completed. In the ordinary course, a strongly performing agency could expect to obtain new commissions. But the BAL model operated (or was capable of being operated) differently. The work done by Digitas UK was of high quality, as Proctor & Gamble always recognised. Despite this, KND proved powerless to resist the loss of ongoing work once the BAL decided to take the work in-house.

Moreover, by the date of the SPA, the writing was already on the wall. The Ariel work had already been lost as a result of the influence of the BAL, Saatchi & Saatchi, albeit that the claimants had been given conflicting information about this, first being told that the Ariel work had been lost as a result of a poor client relationship and then that Ariel remained a client after all. It was already clear that Head & Shoulders work would not last much longer, although the precise timing of the loss of this work was uncertain. While it was clear that Saatchi & Saatchi had already decided to take in-house the work on brands for which it was the BAL, there was also a recognition that the other BALs, Publicis Worldwide and Leo Burnett, were likely to follow the same course.

The threat from “partner agencies looking to take our digital work” and the fact that, in practice, there was little or nothing that Digitas UK could do about this was recognised in numerous internal Digitas documents at a senior level. The frustration and even anger to which this gave rise within Digitas is apparent. It was viewed as an issue which was likely to have a serious impact on Digitas UK’s profitability.

In these circumstances I consider that the claimants are right to say that the fragility and vulnerability of BAL work generally, together with the fact that there had been a dramatic shift in the way that the BAL model was being operated within the Publicis group, was a fact or circumstance which was in existence at the date of the SPA. It is therefore necessary to consider (a) whether that fact or circumstance was disclosed to the claimants, (b) whether any of the Named Individuals or MMS itself was aware of that fact or circumstances, and (c) whether that fact or circumstance could reasonably be expected to have a material adverse impact upon Operating Income or Revenue as defined in clause 10.11(a).

There is in my judgment no doubt that the claimants were provided with no explanation of the BAL model before the conclusion of the SPA, let alone of the fact that the BALs were now operating that model in a way which would cut Digitas UK (or as it was to become, the new merged agency KND) out of future business. In fact the only written references to this model at all prior to the SPA were the unexplained references in Ms Davis’s emails sent the night before and early on the morning of completion. These described the “BAL structure” as something convoluted which would need to be explained. They did not come close to constituting disclosure for the purpose of clause 10.11(a) of the matters relied on by the claimants. The failure to disclose extended in my judgment to all four BAL brands.

Ms Davis suggested in her written evidence that she would be surprised if the subject of the BAL model had not come up during her discussions with Mr Nohr prior to the acquisition date of the SPA. However, I accept the evidence of Mr

The protection for which the claimants bargained was concerned with the awareness of the three Named Individuals, together with the awareness of “the Buyer” – that is, MMS – itself. I consider first the position of the Named Individuals. There was an issue as to whose awareness counts as the awareness of “the Buyer”, but that issue will only matter if none of the Named Individuals had the necessary awareness.

In considering these questions it is necessary to distinguish between awareness of the fact or circumstance – that is to say, awareness of the vulnerability of Digitas UK as a result of the BAL model and of the shift in the way that model was being operated – and appreciation of the impact of that fact or circumstance. As explained above, the former is subjective (was the individual in fact aware of the fact or circumstance in question?) while the latter is objective (could the fact or circumstance reasonably be expected to have a material adverse impact as defined?).

I accept the evidence of Ms Alexander that in a conversation with Ms Frijns at some point in the second half of 2012, the latter said words to the effect that the claimants “had been sold a pup”. Ms Frijns did not seriously deny having said words to this effect, although she thought that the word used might have been “lemon” rather than “pup”. However, while I find that a conversation along these lines happened, that does not necessarily say anything about Ms Frijns’ own knowledge of the position before the date of the SPA.

Ms Frijns agreed in her evidence that the loss of a substantial part of the Procter & Gamble business would have a substantial negative effect on the business of KND, at any rate if the lost business was not replaced by other business. Her attitude was that clients were won and lost all the time and that this was part of the ordinary course of business, although she agreed that to win new clients was hard work.

She denied knowing before the SPA that the Procter & Gamble business was on shaky foundations, but she accepted that with hindsight this was the case. In fact, however, it was Ms Frijns, together with Ms Davis, who had registered serious concerns about the terms on which Saatchi & Saatchi had engaged Digitas UK. It was suggested to her that Ms Davis (with whom she shared an office and who, in her words, “wears her heart on her sleeve”) would have explained her concerns about this, including the way in which matters were developing. She accepted this, but denied the suggestion that such a sharing of concerns had happened before the date of the SPA. Her reason for this denial was that if she had known before the SPA, she would have adjusted the figures in the March 2011 rolling forecast. As she had not done this, she concluded that she could not have been aware of Ms Davis’s concerns.

However, that explanation would only hold good if Ms Davis had advised that the loss of business would affect the 2011 figures. If, as Ms Davis appears to have believed, the impact of Saatchi & Saatchi’s approach would only be felt in 2012, there would have been no need for Ms Frijns to adjust the 2011 forecasts. Her evidence was that what concerned her was the accuracy of the financial forecasts which were her responsibility, namely the Commit and the rolling forecasts produced during the year. By their nature, however, these forecasts were concerned with the short term only, and in particular with whether urgent adjustments needed to be made to Digitas UK’s costs base. Ms Frijns did not need to apply her mind in the period before the SPA to the question whether any expected future developments would affect revenue or profit in 2012 or 2013. Nor did she (or the other two Named Individuals) ever see, let alone apply her mind to, the final terms of the clause 10.11(a) warranty as to her state of awareness which MMS was to give.

In these circumstances I find on the balance of probabilities that Ms Frijns was aware of Ms Davis’s concerns, including in particular the fragility of the BAL model so far as Digitas UK was concerned, the highly probable future loss of the business for which Saatchi & Saatchi was the BAL, and the risk that other BALs would follow suit. I accept that she did not regard these matters as likely to affect significantly the 2011 figures. As she said, she would have had no reason to conceal anything from Mr Nohr or Mr Shaw who, in her mind, would be her new bosses in a few weeks’ time. However, she did not apply her mind to the likely impact of the loss of BAL work in 2012 or 2013.

Mr Tomasulo agreed that from October 2010 onwards there was real concern at a senior level within Digitas about the way in which the BAL model would affect future prospects with Procter & Gamble, albeit that (as he said) nobody really knew how matters would turn out. His evidence, however, was that in his role as Chief Financial Officer he would only have taken notice of a development which found its way into a financial forecast. For example, he insisted that a statement that action by the BALs would have “huge implications on our P&L in the UK” would have no impact on him unless it resulted in a negative change to a financial forecast. I do not accept that evidence, which in any event represents the same short-term thinking as found in Ms Frijns’ evidence. It is moreover difficult to reconcile with some of the emails which Mr Tomasulo sent, for example his email of 9 December 2010 expressing the urgent view that “the troops need direction”.

I accept Mr Tomasulo’s evidence that he, like Ms Frijns, had no reason to conceal relevant information from the claimants, but it does not follow from this that he was not aware of matters which, objectively, ought to have been disclosed pursuant to clause 10.11(a). Certainly Mr Tomasulo received a number of documents which on their face made plain the probable significance (even if not the timing) of the change in the way in which the BAL model was being and was likely to be operated within the Publicis group. He was an astute businessman. I have no doubt that he appreciated the significance of the concerns which were being expressed to him. He too, therefore, was aware of the facts and circumstances on which the claimants rely.

Similar considerations apply to Mr Beringer, who also received documents expressing in clear terms the threat to Digitas UK’s revenue as a result of decisions by the BALs to take away work from Digitas UK and the “huge implications on our P&L in the UK” which this represented. Although in his case there are fewer such documents than in the case of Mr Tomasulo, I find that he too was aware of the real risk which existed. His comment in April 2011, in response to the decision by Saatchi & Saatchi to give notice terminating Digitas UK’s further involvement, that “the impact is huge”, recognised the significance of this issue. While that comment might suggest that the Saatchi & Saatchi decision came as a surprise to Mr Beringer, I note also that he did not challenge the suggestion by Ms Davis in February 2012, commenting on the loss of Oral B work, that “we could see this coming a year ago”. Overall I consider that Mr Beringer was aware of the facts and circumstances on which the claimants rely, even though he never applied his mind (and was probably never given the opportunity to apply his mind) to whether this was something which ought to be disclosed to the claimants in accordance with the final terms of clause 10.11(a).

I have concluded that each of the Named Individuals was aware of the facts and circumstances relied on by the claimants, although it would be sufficient for the claimants’ purposes that any one of those individuals was so aware. In the circumstances the question whether MMS as “the Buyer” was also aware of these matters is of little or no importance. I will therefore deal with it briefly.

The claimants contend that, for the purpose of clause 10.11(a), the term “the Buyer” (although defined as meaning MMS unless the context required otherwise) should be construed as encompassing anybody providing information to the claimants on behalf of Publicis and/or Digitas UK in connection with the transaction who, if aware of facts and circumstances satisfying the requirements of the clause, would have been expected to disclose them to the claimants. The purpose of this contention was to enable the claimants to rely on the awareness of Ms Davis.

The claimants support this contention with an argument that elsewhere in clause 10.11(a) the term “Buyer” must be broadly construed. They say (and I accept) that when the clause refers to matters being “disclosed by the Buyer”, that cannot be limited to disclosures made by MMS itself, partly because MMS was only introduced into the transaction at a very late stage and in any event because it would be absurd if disclosures by Digitas personnel with whom the claimants were negotiating (including the Named Individuals themselves) did not count as disclosures “by the Buyer”. Therefore, they say, reference to “the Buyer” elsewhere in the clause must similarly be given a broad construction.

I do not accept this. The bargain struck by the parties was that, in addition to MMS itself, the awareness of the three Named Individuals was what mattered for the purpose of the clause. These were individuals who had been personally involved in the transaction. Applying ordinary principles of attribution, their knowledge would not count as the knowledge of MMS but they could reasonably be expected to have knowledge of any facts or circumstances which ought to have been disclosed to the claimants. It made sense, therefore, for MMS to be required to give a warranty which extended to their awareness of relevant facts and circumstances. However, to construe “the Buyer” in the phrase “none of the Buyer, Stephan Beringer, Joseph Tomasulo or Charlotte Frijns” as encompassing anyone with relevant knowledge who had been involved in providing information to the claimants would render pointless the identification of the three Named Individuals whose knowledge was to count for the purpose of paragraph (a) of the clause. If the claimants had wished to be in a position to rely on facts and circumstances known to Ms Davis (which were not known to the three Named Individuals), they should have bargained for her inclusion as an individual named in the clause.

There is in my view no equivalence in the use of the term “the Buyer” in the different places where it appears in the clause 10.11. The phrase “save as disclosed by the Buyer” in the opening words of the clause plainly refers to disclosures made in the course of negotiations on behalf of whichever corporate entity was to become a party to the contract. However, there is no reason to construe the term “the Buyer” in paragraph (a) itself as extending more widely than MMS and, in view of the naming of three identified individuals, there is every reason not to do so.

The claimants rely on the fact that MMS itself was simply a holding company with no operational business, inserted into the transaction as the contracting party at a late stage, with directors and officers who were most unlikely to be aware of any relevant facts or circumstances. They suggest that this supports their extended construction of “the Buyer” because, if this expression is limited to MMS itself, it is “inconceivable” that MMS would have been aware of any relevant facts or circumstances. To my mind, however, the fact that MMS itself is unlikely to have had any relevant knowledge points in the opposite direction. It was for this reason that it made sense for the parties to identify as Named Individuals those persons whose knowledge was relevant for the purpose of the clause. That would enable MMS to understand the scope of the warranty which it was giving and to make enquiries of the Named Individuals (notwithstanding that in fact it only did so in a cursory way by reference to an incomplete draft version of the clause). If the warranty extended to the knowledge of anyone who had been involved in providing information to the claimants it was potentially of very uncertain scope.

It is necessary to consider next whether the facts and circumstances of which the Named Individuals were aware could reasonably be expected to have a material adverse impact on KND’s Operating Income or Revenue in 2012 or 2013. That depends on whether it could reasonably be expected that the BALs’ new strategy of retaining Proctor & Gamble work for themselves and cutting out Digitas UK or KND would cause a reduction of at least 20% of Operating Income or 10% of Revenue compared with what was otherwise expected.

As already stated, it is obvious that the loss of a major client was capable of having – and could reasonably be expected to have – an adverse impact on future performance. That is expressly contemplated by the terms of the clause. As Mr Tomasulo accepted, the relationship with Procter & Gamble was a key relationship for Digitas UK. It was, as he said, “a huge part of the business”. Indeed it accounted for over 50% of Digitas UK’s Revenue in 2010, the greater part of which came through the BAL model. It was expected to be KND’s largest single client after the merger. If the loss of such a significant part of Proctor & Gamble’s business could not reasonably be expected to have a material adverse impact within the meaning of the clause, it is hard to see what would. Indeed, Mr Tomasulo accepted that if Digitas UK had thought that the BAL model was not going to work successfully for it in the future because the BALs were going to move work in-house, it would have told KNAS about this because that would have been a material change in the relationship.

Any answer to the question whether the loss of a major client would cause a reduction of the specified percentages of Operating Income or Revenue must take account of two factors. The first is the loss of revenue, which is relatively straightforward. The projected revenue from that client will no longer be received. The second, however, is the effect on profitability. This involves wider considerations. Thus it was entirely foreseeable that the loss of a client such as Procter & Gamble would result not only in the loss of revenue but also in cost-cutting measures which, whatever their effect in the short term, would affect KND’s ability to grow the business in the longer term. For example, as Ms Frijns explained, any projected failure to achieve the revenue targets set in the Commit would require adjustments to the cost base, in practice by reducing salary costs. That is what happened when the Procter & Gamble work fell away after completion of the SPA. The foreseeable consequences were that KND was less able to service existing clients and win new ones, that additional costs were incurred in the form of redundancy payments, that fixed overheads remained payable despite the reduced client base, and that management time which could have been applied to growing the business had instead to focus on managing the fallout from the loss of this work. These were precisely the arguments which Mr Nohr put forward, which Mr Tomasulo supported and which (at least) Mr Dupont accepted, as explained above.

Taking these matters into account, I have no doubt that it could reasonably be expected at the date of the SPA that the BALs’ new strategy of retaining Proctor & Gamble work for themselves and cutting out Digitas UK or KND would cause a reduction of at least the specified percentages of Operating Income and Revenue during 2012 and 2013 compared with what was otherwise expected.

The defendants sought to resist this conclusion by pointing out that it was a feature of the industry that clients would come and go and that, when one client might be lost, another could be won. They relied on the evidence of Ms Davis quoted above that when one account closed, the reduction in revenue had to be made up elsewhere. I accept that some turnover of clients was to be expected in the ordinary course of business. However, the loss of a major client (and a fortiori the loss of KND’s principal client) as a result of facts which were not disclosed to the claimants is a matter expressly within the scope of clause 10.11(a) and is of a different order of magnitude. It is no answer to say that clients may be lost in the ordinary course of business or that, when they are, an effort can be made to gain new clients to replace them.

For the reasons set out above I have concluded that the claimants have established a breach of the Buyer Warranty in clause 10.11(a) on the part of MMS. There is of course no question of any breach by the second defendant, Publicis, which was not a party to the SPA.

I defer consideration of the remedy to which the claimants would be entitled under clause 10.12 until after I have dealt with their alternative case that an agreement was reached in either October or December 2012.

The legal principles to be applied in determining whether the parties reached a binding agreement are not in dispute. They were summarised by Lord Clarke giving the judgement of the Supreme Court in RTS Flexible Systems Ltd v. Molkerei Alois Muller GmbH [2010] UKSC 14, [2010] 1 WLR 753, at [45]:

“The general principles are not in doubt. Whether there was a binding contract between the parties and if so, upon what terms depends upon what they have agreed. It depends not upon their subjective state of mind, but upon a consideration of what was communicated between them by words or conduct, and whether that leads objectively to a conclusion that they intended to create legal relations and had agreed upon all the terms which they regarded or the law requires as essential for the formation of legally binding relations. Even if certain terms of economic or other significance have not been finalised, an objective appraisal of their words and conduct may lead to the conclusion that they did not intend agreement of such terms to be a precondition to a concluded and legally binding agreement.”

Lord Clarke went on at [49] to approve the well known summary of the relevant principles by Lloyd LJ in Pagnan SpA v. Feed Products Ltd [1987] 2 Lloyd’s Rep 601 at 619, adding that the same principles apply where the question is whether a contract was concluded in correspondence as well as by oral communications and conduct. These principles include that:

“In order to determine whether a contract has been concluded in the course of correspondence, one must first look to the correspondence as a whole...”

Hamblen LJ explained the rationale of this approach in Global Asset Capital Inc v Aabar Block Sarl [2017] EWCA Civ 37 at [30]:

“The rationale of this approach is that focusing on one part of the parties’ communications in isolation, without regard to the whole course of dealing, can give a misleading impression that the parties had reached agreement when in fact they had not.”

It is therefore necessary to have regard, not only to the parties’ communications up to 26 October and 17 December 2012 (the dates on which the claimants say that binding contracts were concluded) but also to their communications after those dates. The claimants do not suggest that binding contracts were concluded after those dates, but later communications may cast light on the question whether contracts were concluded on those dates as the claimants allege.

Applying those principles to the communications described above, I have no doubt that no binding contract was concluded on or about 26 October 2012. The claimants’ case is that a binding contract was concluded that reasonable adjustments would be made to the calculation of Deferred Consideration in order to ensure that they would not be prejudiced by the loss of Procter & Gamble work. However, it is clear that the claimants were not satisfied with and did not accept the proposal that had been made in Mr Dupont’s email of that date. They were expressly asked whether the process described in that email was acceptable to them and their response was that it was not. They treated the proposal as a useful step on the way to the agreement which they hoped to achieve, which no doubt it was, but they made it perfectly clear that it was not enough. Looking at the matter objectively, it is apparent that the parties did not intend to create legal relations as a result of their communications on or about this date. As it happens, this objective conclusion accords with the parties’ subjective understanding. As a result, they continued their discussions.

The real question is whether the process which culminated in the telephone conversation between Mr Nohr and Mr Tomasulo on 17 December 2012 and which was then confirmed in Mr Nohr’s email of the following day was intended to result in a legally binding agreement. Again, the matter must be looked at objectively.

Although Mr Nohr had no independent recollection of the telephone call, I accept that his email to Mr Tomasulo dated 18 December 2012 was an accurate statement of what Mr Tomasulo had said on the telephone (although it was not and did not purport to be a verbatim record) and that Mr Tomasulo had accurately passed on the results of his communication with Mr Dupont. If that had not been the case, Mr Tomasulo would have made the position clear.

The claimants point to three factors in particular as demonstrating that the parties intended a legally binding agreement. First, what the claimants had been seeking to achieve was agreement on the adjustments which would need to be made for the purpose of calculating Deferred Consideration. Those were now agreed by Mr Dupont. Second, the terms of Mr Nohr’s email to Mr Tomasulo clearly refer to the conclusion of a process and not merely the reaching of yet another interim stage. The references to sharing the news with shareholders and to the delight with which the shareholders had received the news strongly support this. As Mr Nohr had always made clear, what he and the other claimants had wanted was to “close this down” – in other words, to secure agreement on the adjustments which would be made so that all that remained was to feed in the figures, once they were known, to the new formula. Third, the claimants made no further attempt to obtain agreement to the adjustments, although up to this point they had been persistent in seeking such agreement. They would not have given up at this stage. Having pressed hard for a binding agreement for some months, it was obvious to all concerned that the claimants would not have abandoned the race just short of the finishing line.

I accept that these factors, taken together and viewed objectively, represent a strong case that the parties did intend a binding agreement at this stage.

The claimants rely also on Mr Lord’s comment, “Nice job Joe getting this across the line …” However, although this comment is highly relevant to a different aspect of this issue, it was not a communication between the parties and therefore cannot affect an objective evaluation of their exchanges.

The defendants suggest that the terms of Mr Nohr’s email, and in particular the expression “hope that this gets the green light as discussed”, fall some way short of the language of binding contract. They rely also on the terms of his email to the former KNAS shareholders (although that too was not a communication between the parties) which spoke of appearing “to have a way forward on the earn-out” rather than of having a legally binding contract. However, Mr Nohr was not a lawyer and these linguistic arguments carry little weight. In particular, I do not accept that his reference to getting “the green light” once figures were provided constituted an acknowledgement that agreement had not yet been reached. Rather, his point was that now that the revised formula had been agreed, it should be a straightforward matter to apply that formula to the financial results for 2012 and 2013.

The next question is whether the agreement by Mr Dupont to the adjustments proposed by the claimants had been approved by Mr Lévy, it being common ground that his agreement was required for any agreement binding on MMS. It is at this stage that Mr Lord’s congratulation to Mr Tomasulo for “getting this across the line” is relevant. Mr Lord was the most senior executive within Digitas and was well placed to have an accurate understanding of the position reached. His image is one of completing a race, not of being left with another hurdle still to jump. The comment makes perfect sense if it was understood that the agreement had been approved by Mr Lévy so that the matter was now concluded. But it makes no sense if Mr Lévy had not given his approval, as (if he had not done so) that approval could certainly not be taken for granted.

Mr Tomasulo insisted in his evidence that the only “line” which had been crossed was obtaining the agreement of Mr Dupont, but that Mr Dupont’s agreement did not bind anyone and it was for Mr Nohr to take the matter forward to obtain Mr Lévy’s agreement. I found that evidence thoroughly unconvincing and reject it. If Mr Tomasulo had indeed thought at the time that this was all that had been agreed, he would have encouraged Mr Nohr to conclude the matter by seeking Mr Lévy’s agreement. That would have been consistent with his contemporaneous advocacy of the claimants’ position which was in part at least the result of his strong wish to keep the senior management of KND properly motivated. I accept the evidence of Mr Nohr (supported to some extent by Mr Estryn) that Mr Dupont’s habit was to make clear when he needed to obtain Mr Lévy’s approval of what he was saying. The fact that he did not do so on this occasion indicates that he had already obtained it.

I note also that neither Mr Lévy nor Mr Dupont was called by the defendants to give evidence. Nor was any explanation given why they could not have been. It is true that the way in which the claimants put their case as to a December 2012 agreement only emerged after the date for exchange of witness statements and at a relatively late stage before trial, but I can see no reason why evidence could not have been adduced, certainly from Mr Lévy, if he had anything to say which would assist the defendants’ case. Mr Dupont has now left the Publicis group, but again it ought to have been possible to adduce evidence from him.

If it were legitimate to stop at this point, I would therefore conclude that a binding agreement was reached as a result of the 17 December 2012 telephone call and its email confirmation. However, as the cases cited above make clear, that is not a legitimate approach. Later communications cannot be regarded in this case as forming part of a continuous negotiation as nothing further was said on the subject until May 2013. However, later communications are in principle capable of showing that in fact no agreement had been reached in December 2012 and must be examined in order to see to what extent they do cast light on that question.

It must be acknowledged that, taken at face value, the claimants’ letter dated 29 May 2013 casts serious doubt on the existence of any binding agreement made in December 2012. The letter asserts in terms that no such agreement had been made. The claimants could hardly complain if they were held to that conclusion. However, for the reasons explained above, I consider that the assertion in the letter that nothing had been agreed, and that there had only been discussions, was a deliberate misstatement because, by this time, the claimants were seeking to improve on what had been agreed in December 2012 and hoped to obtain compensation also for what they saw as the adverse impact of the acquisition of LBi. To say the least, that reflects no credit on the claimants, but it does not in the end affect my assessment of what was or was not agreed in December 2012.

Still later communications do not take matters much further. Mr Nohr’s exchanges with Mr Beringer in January 2014 are at least consistent with an agreement having been reached (“if Maurice honours the adjustment he promised”) even if the later message refers only to Mr Lévy deciding “whether to honour the P&G correction agreed by you and Joe Tomasulo”. In both cases the language of “honour” suggests a degree of commitment which, as Mr Nohr knew, would only exist if Mr Lévy had in fact agreed the adjustment. Moreover, it appears from Ms Murphy-Dowd’s comment to Mr Estryn of the Publicis M&A department that Mr Tomasulo had told her that there were “agreements previously made with you relating to earnout adjustments”. That must refer to agreements made with the approval of Mr Lévy. The comment would otherwise be pointless.

Taking account of the correspondence as a whole, I conclude that agreement was reached in December 2012 as to the adjustments which would apply to the calculation of Deferred Consideration and that these adjustments were intended to be applied to both the relevant years, 2012 and 2013.

The defendants’ witnesses – in particular Jean-Michel Etienne who was a director of MMS – accepted that Mr Lévy had authority to bind MMS. There is as I understand it no dispute that an agreement made with the approval of Mr Lévy would be binding on MMS. Since what was being negotiated was a variation of the SPA, I conclude that the agreement reached in December 2012 was an agreement between the claimants and MMS.

I do not accept, however, that Publicis undertook any legally binding commitment to the claimants. Although they plead that there was an agreement by Publicis to procure that the adjustments agreed on behalf of MMS would be made, this being in fact the claimants’ only basis for seeking to hold Publicis liable in this action, such an agreement is an artificial and unnecessary concept. It was never suggested by the claimants in the negotiations that they needed something more than an agreement by MMS to vary the terms by which the Deferred Consideration would be calculated or that any agreement would need to include a legally binding commitment by Publicis. It is to my mind highly unlikely that the claimants would ever have made such a suggestion expressly. I strongly suspect that it would have been regarded by Mr Lévy as insulting and that the claimants (if they thought about it at all) would have been well aware of this. That being so, there is no room to find any implied commitment on behalf of Publicis. Accordingly the claim against Publicis must fail.

I conclude also, if it matters, that the agreement constituted a compromise of any breach of warranty claim. I have referred above to the exchanges which show that this was certainly in the claimants’ minds and is highly likely to have been in the defendants’ minds also. The claimants resist this conclusion, relying on the fact that neither the defendants’ pleadings nor their witness evidence advanced a case that the breach of warranty claim had been compromised so as to prevent such a claim from being advanced. However, in circumstances where the defendants and their witnesses have throughout denied that any agreement was concluded, this submission by the claimants has little or no weight.

I should record that at one time the defendants pleaded that any agreement would fail for want of formality, relying on a clause in the SPA which required any variation to be in writing and signed. However, that point was abandoned in the light of the decision of the Court of Appeal in Globe Motors Inc v TRW Lucas Varity Electric Steering Ltd [2016] EWCA Civ 396 (which was confirmed as a matter of ratio by MWB Business Exchange Ltd v Rock Advertising Ltd [2016] EWCA Civ 553).

It is common ground that if an agreement was concluded in December 2012, as I have found that it was, the adjusted Deferred Consideration to which the claimants are entitled is either £3.6 million or £2.6 million and that which of these is the correct figure depends on the application of a clause dealing with fees charged by the Publicis group’s Shared Service Centre or “SSC”. I deal with that issue now.

The SSC was an entity within the Publicis group which supplied services to the various agencies within the group at a centrally charged cost. In principle that would appear to promote economies of scale, but it did leave agencies potentially vulnerable to fees whose level they could not control. An agency would not be able to shop around to obtain services from another provider, but would have to pay the fees which were centrally determined within the group. There was therefore a risk, recognised by the parties, that the claimants’ entitlement to Deferred Consideration, which depended on the profitability of KND, could be affected by the level of fees charged by the SSC.

The parties dealt with this in the SPA by including within the definition of “Operating Income” a requirement that a specific adjustment be made to take account of SSC fees. Clause 4.1(m)(iii) of Schedule 5 to the SPA provided that:

“shared service fees charged by the SSC (provided that for the purpose of calculating Operating Income, such shared service fees, excluding any related to general IT services and general legal services (which will be billed on a per usage basis), shall not exceed 1.05% of the difference between Revenue and Operating Income (before charging any shared service fees) for such period) and Intercompany Coordination & Creative Fees shall be deducted in accordance with schedule 6.”

The effect of this was that for the purpose of calculating Deferred Consideration, there was to be a cap on the SSC fees which would count as costs incurred by KND. The SSC could still charge KND for its services in the same way and at the same rates as it charged other agencies within the group, but any excess fees over the level of the cap would not count for the purpose of the calculation. However, this cap did not apply to “general IT services” and “general legal services” which, as explained in parentheses, were to be “billed on a per usage basis”. The fees charged for those purposes, therefore, would simply count as costs incurred by KND.

No issue arises as to legal costs. The issue is as to the treatment of fees for general IT services which, the claimants say, were not “billed on a per usage basis” as envisaged by the clause.

In the case of some general IT services, it would be straightforward to charge on a per usage basis. An example would be for use of an IT helpdesk where, if assistance was provided, a charge could be made, but not otherwise. However, as Ms Frijns explained, that is not how matters were dealt with at Publicis. Rather, agencies were charged a fee which covered all aspects of the provision of IT:

“The way we get charged by SSC for IT costs is not just the helpdesk; it pays for the infrastructure, for the service, the servers, and I am not an IT expert, but ensuring the whole infrastructure is safe, we cannot get hacked, all those things, are part of the service IT delivers.”

The fee for those services was spread across the Publicis organisation based on a headcount of employees. Thus if an agency employed (say) 5% of the total number of employees within the group, it would be charged 5% of the total fees charged by the SSC.

In these circumstances the claimants submit that, as a matter of construction, (1) the exclusion from the cap of general IT services applies only to the extent that such services were in fact billed on a per usage basis, and (2) the fees charged in this case were not so billed and accordingly are subject to the cap. The defendants, however, submit that the exclusion from the cap of general IT services is clear and that in any event the services encompassed within this expression, identified above, can be regarded as being charged for on a per usage basis in circumstances where no other method of charging was practicable. They rely also on documents exchanged between the parties in the course of concluding the SPA which, they say, made clear that IT services were and would be charged for “per user”.

In my judgment it is clear that the clause in question provides that general IT services are to be excluded from the cap. That exclusion is not conditional on the services being billed on a per usage basis, albeit that the clause contemplates that they will be charged for in this way. However, it seems to me that charging on the basis of headcount can also be regarded as charging “per usage” when the services in question are utilised every time an employee switches on his or her computer. Although Ms Frijns was prepared to accept in her evidence that there is a distinction between charging “per usage” and “per user”, and in general I accept that this is so, it is not a distinction which is readily applicable to the use of services such as the provision of IT infrastructure. I conclude, therefore, that the IT services charged by the SSC should be excluded from the cap for the purpose of calculating the claimants’ entitlement to Deferred Consideration.

In reaching this conclusion I have not needed to rely on the pre-contractual documents to which the defendants referred. It seems to me that in order to do so it would be necessary to venture into the disputed territory on the boundary between factual background and pre-contract negotiations, the latter being inadmissible as an aid to construction (Chartbrook Ltd v Persimmon Homes Ltd [2009] UKHL 38, [2009] 1 AC 1101). I record, however, that those documents demonstrate that the claimants were aware that IT services would be charged for on the basis of an employee headcount.

It is common ground that, on the basis of my conclusions that (1) the December 2012 agreement was concluded in the manner described above and (2) general IT services should be excluded from the cap, the sum to which the claimants are entitled pursuant to that agreement is £2.6 million. I have concluded also that the effect of the agreement is that there is no remaining scope for a separate breach of warranty claim. Although this conclusion does not depend on characterisation of the December 2012 agreement as a compromise of a breach of warranty claim, I have found that the agreement can and should be so characterised, in which case the claimants accept that no separate breach of warranty claim can be made. In any event, a claim to a remedy for breach of the Buyer Warranty in the form of an adjustment under clause 10.12 would only assist the claimants if it would enable them to recover more than £2.6 million.

However, in case I am wrong in the conclusions reached so far, I consider next the adjustment to which the claimants would be entitled pursuant to clause 10.12.

I set out again for convenience the terms of the clause:

“The Sellers’ sole remedy for breach of the Buyer Warranties shall be an adjustment to Revenue and/or Operating Income for the purposes of calculating the Deferred Consideration if and to the extent that the Deferred Consideration is adversely affected by a matter that gives rise to a breach of the Buyer Warranties.”

The claimants recognise that they are only entitled to an adjustment “if and to the extent that the Deferred Consideration is adversely affected by a matter that gives rise to a breach”. There is, therefore, a question of causation which can only be addressed after the event, when it can be determined to what extent, if at all, the Deferred Consideration has been so affected. However, while the question of causation must be addressed, it must be recognised (and would always have been apparent) that disentangling the various factors which may affect Revenue and Operating Income in any given year in order to isolate the effect of a breach may be a complex process. It was not to be expected that mathematical precision would be attainable. I consider, therefore, that the parties would have contemplated a flexible approach with a view to achieving a fair outcome.

The claimants submit that in the circumstances of this case the appropriate adjustment should be based on the substitution of the reasonably expected Operating Income and Revenue for 2012 and/or 2013 as at the time of Completion in place of the actual Operating Income and Revenue for 2012 and 2013 – in other words, that for the purpose of calculating the Deferred Consideration it should be assumed that, but for the breach, KND would have achieved the Operating Income and Revenue which was reasonably expected at the date of the SPA as set out in the Integration Plan. The justification for this approach, which the claimants accept would not necessarily apply in every case of breach of warranty, is that (they say) the loss of the Procter & Gamble BAL business that had previously been undertaken by Digitas UK was of such catastrophic severity, having regard not only to the purely numeric loss of the expected revenue itself but also to the wider consequences for the business (the diversion of the attention of KND’s senior management who now had to focus on cost cutting and a drastic re-structuring of the business, instead of nurturing existing client relationships and winning new business), that the actual performance of KND in 2012 and 2013 is not a reliable guide as to how KND would have performed if there had not been any breach of the warranty. The loss of this business was, in the claimants’ words, “a game-changing event”.

As I have already made clear, I accept that the loss of the Procter & Gamble BAL business had an impact on KND going beyond the loss of predicted revenue. Nevertheless, determination of the extent to which a failure to achieve projected Revenue or Operating Income is attributable to an undisclosed fact or circumstances which gives rise to a breach of warranty remains a question of fact. It is necessary to consider whether the claimants’ proposed assumption is credible. As explained below, I consider that it is not.

Both parties adduced expert evidence from forensic accountants on this issue. Their evidence was careful, fair and (so far as it went) helpful. Nevertheless both experts were severely constrained in the evidence which they could give by the instructions which had been given to them.

The claimants’ expert was Mr Andy Cottle of BDO LLP. He was instructed to calculate the Deferred Consideration “based on the substitution of the (reasonably expected) OI and/or revenue for 2012 and/or 2013 as at the time of Completion and on the basis of the information it had in fact been provided to the Former Shareholders … in place of the actual revenue and/or OI for 2012 and 2013”. Thus his calculation assumed the correctness of the claimants’ proposed approach, with the assumed consequence that, but for the breach of warranty, the actual Revenue and Operating Income would have been as projected in the Integration Plan. He calculated on this basis a Deferred Consideration of £4.85 million. The defendants accept that, if the approach is valid, the calculation is correct.

However, as Mr Cottle accepted, the calculation which he had performed in accordance with his instructions (1) assumed that every single pound by which the actual results fell short of the projected results was attributable to the breach of warranty, and (2) did not take into account any other factors which may have been operating on the actual results.

I am confident that this is not a valid approach, even applying a flexible approach to the clause. This is no criticism of Mr Cottle, who performed the calculation which he was instructed to perform, but the assumption underlying his instructions does not correspond with reality. As is apparent from the matters set out in the narrative section of this judgment, there were numerous factors affecting the actual results achieved which had nothing to do with the breach of warranty. These included, by way of example, the loss of other Procter & Gamble business (including the important Eukanuba account and the decision to switch away from digital advertising during the 2012 London Olympics) which had nothing to do with the BALs and which has not been proved to be a consequence of the breach of warranty alleged by the claimants, the reduction in work from financial services clients and the loss of other significant clients such as (but not limited to) Shell and Delta.

It is apparent that 2012 and 2013 were years in which the business climate for KND was challenging for reasons unrelated to the loss of the four Procter & Gamble brands which were subject to the BAL model. These adverse business conditions were not what the parties had expected. Referring to the fact that in 2012 and 2013 KND was a much smaller business than had been envisaged at the date of the SPA, the claimants themselves went no further in their closing submissions than describing the loss of Procter & Gamble business as “a major contributory factor”. It is obvious that other factors also played their part. It follows that the claimants’ primary claim for £4.85 million, based on Mr Cottle’s calculation, must fail. He has not made (and in fairness was not instructed to make) any assessment of the extent to which the loss of the BAL brands caused the failure to achieve the results projected in the Integration Plan.

The claimants seek to avoid this conclusion by arguing that apart from the loss of the Procter & Gamble BAL business, the other work lost by KND represented the normal and expected coming and going of clients in the ordinary course of business. They contend that, even if these particular clients had been lost, but for the loss of the BAL business KND would not only have made good any shortfalls in revenue but would have achieved the rates of growth projected in the Integration Plan. I regard that, however, as speculation and wishful thinking. It is in my judgment an attempt to “finger p&g” for the loss of such business which is just as lacking in merit as the suggestion that it might be possible to “finger p&g” for redundancies which had nothing to do with the loss of Procter & Gamble work.

Mr Cottle also put forward some alternative figures. His first alternative resulted in a Deferred Consideration of £4.175 million. However, this was predicated on (among other things) acceptance of the claimants’ approach to the issue of SSC charges for IT services, which I have rejected. His final alternative resulted in a Deferred Consideration of £3.05 million. This was predicated on the defendants’ approach to the SSC charges issue, but it also assumed that severance costs of £249,000 incurred in 2012 were all attributable to the loss of the Procter & Gamble BAL business, with little margin for error. What this means is that even if only a small proportion of such costs were not so attributable, the effect of the Deferred Consideration formula in the SPA would invalidate this figure. Despite the evidence of Mr Steven Ireland, KND’s Head of Finance, I consider that it is not possible, even adopting the kind of flexible approach which I have accepted to be necessary, to have confidence that all of these severance costs were attributable to the loss of the Procter & Gamble BAL business. Mr Ireland’s evidence does not appear to distinguish between the loss of BAL business and the loss of Procter & Gamble business more generally. Accordingly I do not accept either of Mr Cottle’s alternative figures.

On the other hand, the approach of the defendant’s expert, Mr David Stone of Stone Turn UK LLP, was limited to consideration of the effect of the loss of all projected Procter & Gamble revenue. Adopting this approach, his calculations showed that the effect on revenue of the loss of these brands was such that the expected Operating Income would be reduced by less than 10%, with the consequence that (applying the contractual formula) no Deferred Consideration would be payable. However, the flaw in this approach is that it takes no account of what I have described as the wider consequences of the loss of a major client, which undoubtedly existed and were foreseeable. For his part, however, Mr Cottle thought that such wider consequences would quite likely have existed (I accept that they did) but were difficult or impossible to quantify. In these circumstances it would not be safe to conclude that the claimants suffered no loss as a result of the breach of warranty. Having regard to the importance of this business, it is probable that they did.

In those circumstances, what conclusion should be reached? One possibility, which would be sufficient for present purposes, would be to conclude that even though the claimants have suffered some loss, they have failed to prove that they are entitled to an adjustment which would result in Deferred Consideration greater than the agreed figure of £2.6 million. I make that finding.

However, it is possible to go further and to conclude, as I do, that £2.6 million represents the Deferred Consideration to which the claimants are entitled after making appropriate adjustments. Despite the difficulties which exist in determining the extent to which the calculation of Deferred Consideration was affected by the matters which give rise to the breach of warranty claim, the fact is that the parties did reach agreement about this in December 2012. They did so as parties who were knowledgeable about the KND business and the factors impacting upon it, who were seeking to isolate the consequences of the loss of Procter & Gamble BAL business but were aware of the market and other factors which had affected the business, and who were able to arrive at a series of adjustments which (as they agreed) would fairly reflect those consequences. In those circumstances (and subject only to one point) the agreement which they reached represents the best evidence of the adjustments which would be appropriate pursuant to clause 10.12. I consider this a much more reliable guide than the various theoretical scenarios put to Mr Cottle in cross examination. That is the position regardless of whether the agreement reached was legally binding. On any view, even if Mr Lévy’s approval was lacking so that the agreement did not attain the status of a binding contract, there was nevertheless agreement by Mr Dupont who was charged with the responsibility of detailed consideration of the issue and who evidently gave it careful study.

The one qualification is that, as pointed out by Mr Cottle, the agreement which the parties reached was too favourable to the claimants in one respect, in itself relatively minor, concerning the treatment of overheads. However, on the premise (which I have accepted) that the defendants’ approach to the SSC charges is accepted, it is common ground that the qualification makes no difference to the Deferred Consideration to which the claimants are entitled. Moreover, even if the claimants’ approach to the SSC charges issue were correct, the effect of correcting the treatment of overheads in accordance with Mr Cottle’s evidence would be that instead of Deferred Consideration of £3.6 million which would otherwise apply, the correct figure would remain at £2.6 million. Ultimately, therefore, Mr Cottle’s qualification makes no difference.

I should record that a similar point arose in relation to the treatment of wasted overhead costs. Mr Cottle and Mr Stern adopted two different approaches, each of which can be regarded as reasonable. The difference between them would only matter in the event that (1) no agreement was reached in December 2012 and (2) the claimants’ approach to the SSC issue is correct. The conclusions which I have reached mean that the issue does not matter. Accordingly I will say only that because of the way that the Deferred Consideration formula worked, Mr Cottle’s figure would only have to be very slightly too high in order for the Deferred Consideration payable to be reduced from £3.6 million to £2.6 million. It seems to me that his figure must be subject to at least some margin for error and accordingly that, on these premises, the correct Deferred Consideration figure would still be £2.6 million.

For the reasons explained above I conclude that:

There will be judgment for the claimants against MMS accordingly. The claim against Publicis is dismissed.