Makdessi v Cavendish Square Holdings BV & Anor

The issue in this appeal is whether two clauses in an agreement for the sale of shares are unenforceable on the ground that they are penalties.

By 2008 Mr Talal El Makdessi (“Mr Makdessi”) was a key figure in the advertising and marketing world of the Middle East. He had founded a Group which became the largest advertising and marketing communications group in that region. It operated in more than 15 countries via a network of around 20 companies with more than 30 offices. Mr Makdessi was one of the most influential business leaders in the Lebanon and his name was identified with the business of the Group. He had very strong relationships with clients of the Group and with senior employees.

In 2008 the holding company of the Group was Team Y&R Holdings Hong Kong Ltd, the second claimant/respondent (“the Company”), which was a new holding company formed for the purposes of the acquisition. The shareholders were Mr Makdessi and Mr Joseph Ghossoub (“Mr Ghossoub”). In 2008 Young & Rubicam International Group B.V. (“Y & RIG”), a company in the WPP group of companies, held 12.6% of the shares in the Company.

By an agreement dated 28 February 2008 (“the Agreement”) Mr Makdessi and Mr Ghossoub (together “the Sellers”) agreed to sell to Y & RIG 474 shares in the company, being 47.4 % of its shares then in issue. Y & RIG transferred its shares in the Company to Cavendish Square Holdings B.V. (“Cavendish”), the first claimant, and by a novation agreement of 29 February 2008 Cavendish was substituted for Y & RIG as a party to the Agreement. Cavendish is a holding company within the WPP group, which is the world’s largest market communications services group. Thus Cavendish came to hold 60% of the Company and the Sellers retained 40%.

The Agreement had been the subject of extensive negotiations over six months between the Sellers and Cavendish. Both were represented by highly experienced lawyers – Allen & Overy in the case of Cavendish and Lewis Silkin in respect of the Sellers.

The price payable to the Sellers, pursuant to the Agreement and a First Amendment deed of 11 June 2008, for the 474 shares was to be paid in instalments as follows:

OPAT was the subject of an elaborate definition. It was, in essence, the relevant audited consolidated operating profit after tax. The “M” figure for the final payment was to vary between 7 and 10 according to the extent of the compound annual growth rate in profits in respect of the period 2006 to 2011. The Agreement provided – by clause 3.3 - that the maximum of all payments would be $ 147,500,000.

Clause 3.2 of the Agreement provided that if the Interim Payment and/or the Final Payment turned out to be a negative figure then it/they should be treated as zero, but there was to be no claw back of the earlier payments

The Sellers warranted that the Net Asset Value (“NAV”) of the Company at 31 December 2007 was $ 69,744,340. It is apparent from the figures:

Clause 11 provided as follows:

Clause 11.2 of the Agreement then contained a number of restrictive covenants, which the learned judge helpfully set out, incorporating the relevant definitions taken from Schedule 12 to the Agreement as follows:

“11.2.

Until the date 24 months after the Relevant Date [being the later of the date of termination of employment by the Group (not relevant in the case of the Defendant), the date that he no longer holds any Shares or the date of payment of the final instalment of the Option Price pursuant to Clause 15.5(b) – see below], no Seller will directly or indirectly without the Purchaser’s prior consent:

(a)

carry on or be engaged, concerned or interested in competition with the Group, in the Restricted Activities [being the provision of products and/or services of a competitive nature to the products and/or services provided by the Group Companies in the twelve months prior to the date of the alleged breach of Clause 11.2] within the Prohibited Area [being any countries in which the Group Companies have carried on the business of marketing communications and ancillary services at any time in the period of twelve months prior to the date of the alleged breach];

(b)

solicit or knowingly accept any orders, enquiries or business in respect of the Restricted Activities in the Prohibited Area from any Client [being a client or potential client of any Group Company which has placed any order in connection with the Restricted Activities during the twelve months prior to the date of the alleged breach or which was in discussions with any Group Company in relation to such provision in such period];

c)

divert away from any Group Company any orders, enquiries or business in respect of the Restricted Activities from any Client; or

(d)

employ, solicit or entice away from or endeavour to employ, solicit or entice away from any Group Company any senior employee or consultant employed or engaged by that Group Company.”

By Clause 7.5 (“the Carat Clause”) the Sellers agreed that within four months after completion they would dispose of any shares held by them in Carat Middle East S.a.r.l (“Carat”), and procure that a joint venture agreement of 19 December 2003 to which Group Carat (Nederland) BV and Aegis International B.V, on the one hand and Mr Makdessi, on the other, were parties, would be terminated. Carat was a joint venture company established under this agreement, of which Mr Makdessi was to have 245 shares (i.e. 49%) and the other parties were to have 255 (i.e. 51%) . Carat is defined on its website as “the world’s leading independent media planning and buying specialist…Owned by global media group Aegis Group Plc… [with] more than five thousand people in seventy countries worldwide”. It is admittedly a competitor of the Claimants.

Clause 11.7 contained a further covenant restrictive of the Purchaser:

“The Purchaser recognises the importance of the Group to the Sellers and to the value of the Interim Payment and the Final Payment. Accordingly, the Purchaser commits as set out below to ensure that the interest of each Seller in that goodwill is properly protected.

The Purchaser will not (and will procure that no other member of the WPP Group will) at any time until the end of the last financial period relevant to the calculation of the Consideration without the Sellers’ prior written consent other than within the Group Companies, trade in any of [twenty three identified] countries…using [specified] names.”

The learned judge held that the restrictions in clause 11.2 were not in unreasonable restraint of trade, and there is no appeal from that decision.

Clause 5 is headed “Default”. Clause 5.1. provides:

“If a Seller becomes a Defaulting Shareholder he shall not be entitled to receive the Interim Payment and/or the Final Payment which would other than for his having become a Defaulting Shareholder have been paid to him and the Purchaser’s obligations to make such payment shall cease”.

The definition of Defaulting Shareholder includes “a Seller who is in breach of clause 11.2”.

Clause 5.6 provides:

“5.6.

Each Seller hereby grants an option to the Purchaser pursuant to which, in the event that such Seller becomes a Defaulting Shareholder, the Purchaser may require such Seller to sell to the Purchaser (or its nominee) all (and not some only) of the Shares held by that Seller (the Defaulting Shareholder Shares). The Purchaser (or its nominee) shall buy and such Seller shall sell with full title guarantee the Defaulting Shareholder Shares… within 30 days of receipt by such Seller of a notice from the Purchaser exercising such option in consideration for the payment by the Purchaser to such Seller of the Defaulting Shareholder Option Price [defined as “an amount equal to the Net Asset Value [NAV] on the date that the relevant Seller becomes a Defaulting Shareholder multiplied by” the percentage which represents the proportion of the total shares the relevant Seller holds].”

Mr Makdessi’s remaining shares were 21.552% (40% x 53.88%) of the whole.

Clause 15 contains a Put Option which the judge summarised as follows:

The result of these provisions is that, if a Seller is in breach of clause 11.2 in any respect he becomes disentitled to his share of the Interim and Final Payments if his breach precedes the Due Date of each, or of the Final Payment if his breach falls between the Due Dates for the two payments. In the former case, if both shareholders default, they could lose up to $ 82 million ($ 147.5 – $ 65.5) between them or $ 44,181,600 in the case of Mr Makdessi alone. A defaulting shareholder is also liable to have all his retained shares in the Company called for in return for the appropriate percentage of the NAV of the entire share capital at the date when he became a Defaulting Shareholder. He will thus (a) receive for those shares nothing in respect of goodwill; and (b) will not be able to exercise the Put Option, otherwise available in 2011 and subsequent years, which would give him a price, not exceeding $ 75 million, which reflected goodwill. The difference between a goodwill and a NAV valuation can be very sizeable (as the terms of the Agreement themselves demonstrate), although it would be possible for a valuation at the Option Price to be less than the NAV if the default of the Defaulting Shareholder had a major impact on future profits. Cavendish has calculated what it claims to be the Defaulting Shareholder Option Price as $ 9,519,000 – but this valuation is itself in dispute. It is true that the valuation does not take into account any impact on the goodwill of the breach which causes Mr Makdessi to be a Defaulting Shareholder but that is because it takes no account of goodwill at all.

By the terms of a Service Agreement signed contemporaneously with the Agreement, Mr Ghossoub agreed to remain (as he still is) an employee and director of the Company. During the course of the negotiations leading up to the Agreement, Mr Makdessi made it clear that he did not wish to remain an employee. What he did was to sign, also contemporaneously, an agreement (“the Appointment”), by which he was appointed a non-executive director of the Company and non-executive Chairman, for an initial term of eighteen months which was renewable. Under this he agreed, in Clause 2.2 that “in addition to the requirements of all directors [his] role as a non-executive director has the following key elements”, being certain specific obligations by way of ongoing support of the Company. The Appointment and Mr Ghossoub’s Service Agreement were specifically provided for in the Agreement (Schedules 9 and 12), as documents which were to be delivered on Completion.

Mr Makdessi resigned as non-executive Chairman of the Company in April 2009. On 1 July 2009 he resigned, at the Company’s request, as non-executive director of all companies in the group, save the Company itself. He was removed from the board of the Company on 27 April 2011, after the commencement of these proceedings.

In or by December 2010 Mr Makdessi was found by Cavendish to have acted in breach of his duties to the Company as a Director and, as Cavendish contends, thus also in breach of his obligations under Clause 11.2 of the Agreement. On 13 December 2010 Cavendish gave notice of the exercise of its Call Option under clause 5.6.

These proceedings were commenced on 15 December 2010. In them Cavendish sued for breach of the Agreement, and the Company sued for breach of fiduciary duty. In the Re-amended Particulars of Claim (RAPOC) of 1 June 2012 the Claimants contended that in breach of his fiduciary duties and the restrictive covenants Mr Makdessi had throughout 2008 and 2009 in Lebanon and Saudi Arabia (both within the Prohibited Area):

On 10 October 2012 Mr Makdessi submitted a draft Re-Amended Defence and Counterclaim, by which he admitted that after 28 February 2008 he had had an ongoing, unpaid involvement in the affairs of Carat pending the appointment of a replacement CEO and that such involvement placed him in breach of fiduciary duty to the Company with effect from 1 July 2008: paragraph 9 (a); and that, if contrary to his case, the covenants in 11.2 were valid and enforceable (as the judge has held them to be) his involvement in the affairs of Carat rendered him a Defaulting Shareholder within the meaning of the Agreement: paragraph 17 (b). On 1 October 2012 he made a Part 36 offer to the Company in respect of the Company’s claim for breach of fiduciary duty in the sum of $500,000. On 8 October 2012 the Company accepted the offer and payment was made on 19 October 2012.

Cavendish claimed that Mr Makdessi’s admissions of breach of fiduciary duty showed that he was in breach of clause 11.2 in relation to (at least) his continued involvement in Carat. It sought a declaration that he was a Defaulting Shareholder, was not entitled to payment of the Interim Payment or the Final Payment as a result of Clause 5.1, and was obliged, as of the date 30 days after the service of its notice exercising the Call Option, namely 14 January 2011, to sell to Cavendish all his shares in the Company at the Defaulting Shareholder Option Price, and it sought specific performance of the latter obligation.

In paragraph 30 of the RAPOC the Company claimed to have suffered loss and damage by virtue of Mr Makdessi’s breaches of fiduciary duty in that it had lost the custom of some of its clients and the services of some of its employees. In paragraph 31 Cavendish claimed to have suffered loss and damage in the form of a loss of value of its shareholding in the Company “by virtue of [the Company’s] losses referred to in paragraph 30...” Mr Makdessi’s response - in paragraph 21 of his Re-Amended Defence and Counterclaim - was that the whole of Cavendish’s pleaded case (in paragraph 31) was by reference to the loss of value in its shareholding in the Company by virtue of the Company’s losses included in paragraph 30, and that such loss was irrecoverable as “reflective” loss: in accordance with the principles laid down by the decisions of the Court of Appeal in Prudential Assurance Co. Ltd v Newman Industries Ltd (No. 2) [1982] Ch 204 and the House of Lords in Johnson v Gore Wood and Co. [2002] 2 AC 1). After the exchange of skeleton arguments in November 2012 Cavendish abandoned its claim for damages in these proceedings.

Mr Michael Bloch QC for Mr Makdessi contends that both clauses 5.1 and 5.6 are penal. Clause 3.1 of the Agreement provides that the Purchaser shall make, inter alia, the Interim and the Final Payment. That is an unconditional obligation to make a payment in the future. Clause 5.1 is a clause which operates on breach since it takes effect when a Seller becomes a Defaulting Shareholder which occurs, relevantly, on a breach of clause 11 and, if valid, disentitles Mr Makdessi from receiving a sum which would otherwise be due to him. Clause 5.6 also operates on breach for the same reason. The doctrine of penalties is applicable to provisions of both types.

Both clauses are penal. In consequence there are no grounds for withholding the Interim or the Final Payment; nor was Cavendish entitled to exercise the Call Option in respect of Mr Makdessi’s remaining shares; and he is entitled to exercise the Put Option in accordance with the terms of clause 15. The reasons why they are penal are as follows.

First, any loss which Cavendish may have suffered from a breach of clause 11 is irrecoverable because it is only reflective of a loss which the Company has suffered: see paragraph 25 above. Clause 5.1 provides for Mr Makdessi to be deprived of what may well be over $ 40 million. Clause 5.6 provides for a compulsory sale of shares at a value which may well be tens of millions below their true value. Neither of them can constitute a genuine pre-estimate of a loss which at law amounts to zero.

Second, even if one ignores this point, when the Agreement was made the amounts of the Interim and the Completion Payments was entirely unknown. So also was the likely value of the remaining shares at any time when the Call Option might be exercised. In those circumstances neither clause 5.1 nor clause 5.6 can have been designed to provide any form of estimate of the loss attributable to any breach of clause 11. In addition the exercise of the Call Option deprived Mr Makdessi of any entitlement to exercise the Put Option, the price of which would be based on the goodwill value of his remaining shares.

Third, the operation of the clauses is arbitrary and indiscriminate. Mr Makdessi becomes a defaulting shareholder on any breach of clause 11, whether material or not. The breach might be of any one of four separate covenants. Clauses 5.1 and 5.6 make no distinction between a breach which is comparatively insignificant and one which is very serious; or between a breach which is isolated and one which is continued. They operate in like manner whether Mr Makdessi makes one unsuccessful attempt to solicit the employment of a single senior employee or successfully poaches the Group’s most important executive or sets up a large rival business throughout the Prohibited Area. They make no distinction between a short lived establishment of a small rival concern in one country and the continuation of a major business throughout the Prohibited Area.

In addition, the clauses operate in a perverse way. It is likely that the more significant the breaches of clause 11 and their effect on the goodwill, the lesser will be the Interim and Completion payments (since they will largely be based on a valuation of the goodwill). The lesser those payments are the lesser will be the compensation in effect payable to Cavendish for the breaches in question, represented by the difference between what would be payable on a goodwill and a NAV valuation. So the worse Mr Makdessi behaves, the less he suffers. But, if compensation is the aim, the position should be the reverse.

Fourth, there is a duplication of prejudicial provisions. If Mr Makdessi becomes a defaulting shareholder he will be liable to lose the Interim and the Completion Payment and to have his remaining shareholding purchased at NAV. In addition the Company may sue him for breach of fiduciary duty (as it has done). There is no restriction on it so doing, nor any requirement to give credit for any recovery the company may make. In addition, Cavendish may sue for breach of clause 11. I discount this latter consideration since, as Mr Makdessi contends, the claim produces a nil recovery.

Miss Joanna Smith, QC, for Cavendish submits that these contentions rest on an erroneous base. Although clauses 5.1 and 5.6 operate upon a breach of contract, they are not provisions which purport to provide compensation for breach. For provisions such as these the appropriate test is whether they have a commercial purpose and a commercial justification and lack a predominant intention to deter a breach.

These provisions were not intended, at any rate predominantly, as a deterrent against breach. Clause 5.1 fulfils the commercial purpose of adjusting the consideration in the light of a clause 11 breach. It was wholly justifiable for Cavendish as buyer to reduce the price that it was prepared to pay – a price substantially based on goodwill – if Mr Makdessi showed himself unprepared to abide by the covenants he gave not to do that which would be likely to reduce, perhaps very greatly, the value of that goodwill.

Clause 5.6 fulfilled the commercial purpose of decoupling Mr Makdessi from the company in circumstances where he had shown himself unprepared to abide by those covenants. That was something that would need to be done swiftly. The basis of valuation was a matter for the parties to bargain about – there was nothing unjustifiable in their deciding on one basis rather than another. A NAV valuation is an accepted method of valuation which could be speedily arrived at. That would avoid what might be the difficult task of determining a value based on goodwill in circumstances where that which triggered the need to do so was action on the part of Mr Makdessi which was likely to have damaged it with consequences that might be difficult to predict.

In any event the law of penalties is not applicable to a clause such as clause 5.6.

The judge was not persuaded that clause 5.6 was a penalty clause, the onus being on Mr Makdessi to show that it was. He found that its purpose was not to deter but to decouple the parties on a speedy and conventional basis. Part of its purpose was also to adjust the consideration between the parties. Cavendish was entitled to assess a “competition breach” by reference to “the greatest loss that could conceivably be proved to have followed from the breach”. Any such breach could have a very substantial impact on the goodwill such as not to render the adoption of a NAV valuation inappropriate. The clause was the subject of thorough negotiation and was not shown to have been oppressive.

As to clause 5.1 the learned judge considered the contention that no loss was recoverable by Cavendish because it was reflective of that of the Company and observed that there was no authority for the proposition that:

“a clause said, in order not to be a penalty, to amount to a genuine pre-estimate of loss can be justifiable as a genuine pre-estimate of irrecoverable loss. Diplock LJ in Robophone Facilities at 1446H referred to "a sum greater than the measure of damages to which he would be entitled in common law", which suggests that the loss must be recoverable …. “

However, in his view, as the law now stood, the relevant questions were:

As he found, there was a plain commercial justification. The function of the clause was to adjust the consideration in the event of a competition breach. The existence of detailed negotiations by experienced solicitors negatived oppression. It was significant that there was a series of covenants to protect goodwill and substantial delayed payment therefor, because it meant that the earlier the default (e.g. before the Interim Payment was due) the more damaging it was likely to be, and the more it would detract from the value of the goodwill purchased, which would justify the cancellation of both the Interim and the Completion Payments.

The judge also considered that clauses 5.1 and 5.6 were not, in combination, penal. However he went on to find that, not only was there no provision that the Company would not pursue Mr Makdessi for damages in addition to the operation of the two clauses, but that the Company had in fact done so to the tune of $ 500,000. He considered this to be “double counting” which, as he held, “renders” clause 5.1 a penalty resulting in Cavendish receiving an extravagant return. He therefore gave Cavendish the opportunity to agree to repay the $ 500,000 as a term of the Court making any declaration that it was not liable to make any payment under clause 3 and ordering specific performance of the Call Option in clause 5.6. The $ 500,000 was to be added to the price of the shares. Cavendish has had to accept this position.

In this latter respect both sides contend that the learned judge was in error. I agree. The question whether a clause is a penalty must be judged as at the date of the agreement in which it is contained. If, as the judge in effect holds, the clauses were not penal as at that date, they cannot have become so as a result of subsequent events.

Mr Bloch contends that the fact that the judge regarded the Company’s recovery of the $ 500,000 as rendering the clause penal shows that it was in fact penal ab initio. The ability of the Company to sue without Cavendish having to give credit for the proceeds was built into the Agreement. Moreover at the date of the Agreement it was inherently likely that the Company would sue (as in the event it did).

The law of penalties is a blatant interference with freedom of contract. In Robophone Facilities Ltd v Blank [1966] 1 WLR 1428 Diplock LJ said that he made “no attempt where so many others have failed to rationalise this common law rule”. The rule is traceable to the preparedness of a court of equity to restrain the enforcement at common law of penal bonds i.e. bonds providing for the payment of a sum of money upon the non-payment of principal and interest due under another instrument. The common law courts adopted the doctrine and applied it to situations other than penal bonds.

In Workers Trust & Merchant Bank Ltd v Dojap Investments Ltd [1993] AC 573 Lord Browne Wilkinson described the scope of the law of penalties as follows:

“In general a contractual provision which requires one party in the event of his breach of contract to pay or forfeit a sum of money to the other party is unlawful as being a penalty, unless such provision can be justified as being a payment of liquidated damages being a genuine pre-estimate of the loss which the innocent party will incur by reason of the breach. One exception to this general rule is the provision for the payment of a deposit (customarily 10% of the contract price) on the sale of land…”

Thus the paradigm case in which the law of penalties is engaged is where a party agrees that, upon a breach of contract, he will pay the innocent party a sum of money. But decisions of this Court have held the doctrine to be applicable to a clause which (a) disentitles the contract breaker to receive a sum which would otherwise be due to him; or (b) requires a transfer of property from the guilty to the innocent party either for nothing or at an undervalue.

That clauses of the former kind engage the law of penalties is common ground in the light of the decisions of this court in Gilbert Ash (Northern) Ltd v Modern Engineering (Bristol) Ltd [1974] AC 689; and Firma C-Trade SA v Newcastle Protection & Indemnity Association (“The Fanti” and “The Padre Island”) [1989] 1 Lloyd’s Rep 239 . Mr Bloch contends and Ms Smith disputes that the same applies to a clause of the latter kind in the light of the decision of this court in Jobson v Johnson [1989] 2 WLR 1026.

In the latter case two brothers agreed to sell shares representing 44.914% in a football club for £ 40,000 to the buyer’s nominee. By a side letter to the agreement the buyer agreed to pay an additional sum of £ 311,968 in six equal half-yearly instalments. Paragraph 6 (b) of the letter provided that if the buyer defaulted in payment of the first instalment he would transfer shares amounting to not less than 44.9% of the capital of the club to the brothers subject to payment to him of £ 15,666.50 and if he defaulted in any subsequent instalment he would transfer the share to the brothers subject to payment to him of £ 40,000. The buyer paid the £ 40,000 but defaulted on the first instalment. A variation agreement was then entered into substituting £ 300,000 for £ 311,698 and providing for that sum to be paid by instalments. The buyer paid the first and second instalments but no others. The brothers assigned their rights to the plaintiff who sought to enforce the repurchase of the shares.

The Court of Appeal held that the provision for repurchase was a penalty clause and unenforceable to the extent that it provided for compensation for the innocent party in excess of its loss. A significant complication arose from the fact that the buyer’s claim for relief against forfeiture had been struck out. Nevertheless the Court said that the appropriate course was either to order a sale of the shares with payment to the plaintiff of the amount of the unpaid instalments and interest or to direct an inquiry to ascertain the current value of (i) the shares, (ii) the aggregate of the unpaid instalments, and (iii) the outstanding amount charged on the shares insofar as it exceeded £ 40,000 (and so could not be paid off out of the purchase price of £ 40,000), and to make an order for the retransfer in exchange for £ 40,000 if such value was less than the sum presently due from the buyer.

Dillon LJ held that the trial judge (Harman J) was right to hold that paragraph 6 (b) was a penalty clause because the re-purchase was to be at £ 40,000 if there was default of payment of any instalment regardless of how much had already been paid. As to the application of the doctrine he said (1034H-1035B) that:

“In principle, a transaction must be just as objectionable and unconscionable in the eyes of equity if it requires a transfer of property by way of penalty on a default in paying money as it requires a payment of an extra, or an excessive, sum of money. There is no distinction in principle between a clause which provides that if a person makes default in paying a sum of £ 100 on a certain day he shall pay a penalty of
£ 1,000 and a clause which provides that if a person makes default in paying a sum of £ 100 on a certain day he shall by way of penalty transfer to the obligee 1,000 shares in a certain company for no consideration. Again there should be no distinction in principle between a clause which requires the defaulter , on making default in paying money , to transfer shares for no consideration, and a clause which in like circumstances requires the defaulter to sell shares to the creditor at an undervalue. In each case the clause ought to be unenforceable in equity insofar as it is a penalty clause.”

Nicholls LJ agreed, holding that “there can be no difference between an obligation to pay a stipulated sum of money arising on a default and an obligation to transfer specified property arising on a default”.

Ms Smith urged on us the marked difference in circumstances between that case and this. That case concerned a buyer who failed to pay the agreed instalments and was being required to forfeit what he had bought in reliance on a clause intended to provide security for payment of the price. By contrast, in the present case the clauses operate on breach of restrictive provisions inserted in order to protect the goodwill whose value formed a major part of the consideration for what was being sold and their function is to adjust the consideration and to decouple. It seems to me, however, that the fact that the default does not lie in a breach consisting of a failure to pay but of a failure to abide by the terms of the restrictive covenants cannot affect the application of the principle to clauses providing for the transfer of property. That principle applies to clauses operating on breach, whatever be the nature of the breach.

Further the authorities do not suggest that, in cases where the doctrine of penalties applies to the withholding of entitlements or a requirement to transfer property, the question whether the effect of such clauses is to produce extravagant recompense for loss is irrelevant. Whether or not the doctrine should apply in such cases may be (or have been) debatable; but to deny the continued relevance of that question would be to wrench the doctrine from its equitable foundations.

Many of the authorities relate to the paradigm case and it is convenient to begin with them before considering more recent authority which suggests, inter alia, that whether a clause is penal should not, or need not necessarily, be answered by assuming a complete dichotomy between what is and what is not a genuine pre-estimate and treating as a penalty anything which does not fall within the former category.

The locus classicus is the decision of the House of Lords in Dunlop Pneumatic Tyre Company Limited v New Garage and Motor Company Limited [1915] AC 79. Dunlop sold tyres, covers and tubes to trade purchasers with discounts off list prices. The purchasers had to agree (a) not to sell to private customers at less than list prices or to trade customers at less than list prices after deducting certain discounts and (b) as agent for Dunlop to obtain from sub-purchasers who were trade customers a similar undertaking to observe Dunlop’s list prices and abide by other undertakings.

The New Garage company, a sub-purchaser, entered into an agreement with Dunlop (acting through the purchaser) whereby it agreed (i) not to alter, remove or tamper with the marks or numbers on Dunlop’s covers or tubes; (ii) not to sell or offer any Dunlop tyres covers or tubes, to any private customer at prices below those in the current price list; (iii) not to supply any such goods to anyone whose supplies Dunlop had decided to suspend; (iv) not to exhibit Dunlop goods at any Exhibition in the UK without Dunlop’s consent; and (v) not to export any such Dunlop goods to any country outside the UK without such consent. It agreed to pay Dunlop £ 5 for each tyre, cover or tube sold or offered in breach of the agreement. The New Garage started selling tyre covers at £ 3 12s 11d when the list price was £ 4 1s. Lord Dunedin’s speech sets out a number of propositions deducible from the authorities which he regarded as authoritative.

Proposition No 1 was that, whilst parties who use the words “penalty “ or “liquidated damages “ may prima facie be supposed to mean what they say, the expression used is not conclusive.

Proposition No 2 was that the “essence of a penalty is a payment of money stipulated as in terrorem of the offending party; the essence of liquidated damages is a genuine covenanted pre-estimate of damage”. Later authority has treated “in terrorem” as somewhat archaic in language or inapt (since parties may not be “in the least terrorised by the prospect of having to pay” penalties: Lord Radcliffe in Campbell Discount Co Ltd v Bridge [1962] AC 600, 622). “Intended to deter” is a modern equivalent. The expression “genuine” does not “turn upon the genuineness or honesty of the party or parties who made the pre-estimate. The test is primarily an objective one, even though the court has some regard to the thought processes of the parties at the time of contracting”; Alfred McAlpine Capital Projects Ltd v Tilebox Ltd [2005] EWHC 281 (TCC) per Jackson J. An analysis of the true character of the provision is critical.

Proposition No 3 is that “the question whether a sum is [a] penalty or liquidated damages is a question of construction to be decided upon the terms and inherent circumstances of each particular contract, judged at the time of the making of the contract, not at the time of the breach”. However, as Lord Woolf observed in Philips Hong Kong Ltd v The Attorney General of Hong Kong [1993] 61 BLR 41 at 59, "the fact that the issue has to be determined objectively, judged at the date the contract was made, does not mean that what actually happened subsequently is irrelevant. On the contrary it can provide valuable evidence as to what could reasonably be expected to be the loss at the time the contract was made”.

Proposition 4 is in four parts:

The House regarded the provision in Dunlop as not amounting to a penalty. Lord Dunedin observed that proposition 4 (c) had been overpressed in argument; that the damage from the various possible breaches was indirect; and that the data did not suggest that it would differ in magnitude. But, as he added in relation to Proposition 4 (c):

“if there are various breaches to which one indiscriminate sum to be paid in breach is applied, then the strength of the chain must be taken at its weakest link. If you can clearly see that the loss on one particular breach could never amount to the stipulated sum, then you may come to the conclusion that the sum is a penalty.”

The House took into account commercial considerations. Part of the reason for upholding the clause was that the terms were there to preserve the integrity of Dunlop’s wholesale distribution system (which required resale price maintenance and strict control of persons who were to be supplied).

The speeches (and later cases) display an intention to keep proposition 4 (c) within limits. Thus Lord Atkinson warned against treating an obligation to carry out a composite act (e.g. the completion of a house) as an amalgam of a variety of obligations and then claiming that a clause providing for a liquidated sum was excessive in respect of some of them. He regarded the obligation relied on by Dunlop as one, in effect, to sell and endeavour to sell Dunlop’s goods at the list price. He warned against requiring the damage resulting from different events to be almost uniform in order to avoid a sum being a penalty. He also regarded the presumption referred to in Proposition 4 (c) as rebutted by the fact that the very damage caused by each of the events, however varying in importance might be of such an uncertain nature that it could not accurately be ascertained.

Lord Parker described as “the really difficult cases” those, such as the present, where the court has to consider what presumptions or inferences arise from the number or nature of the stipulations on breach of which the sum in question is to be paid. He postulated, first the breach of single stipulation such as a covenant not to reveal a trade secret to a rival which can be broken once only and in only one way. In those circumstances there could be no inference or presumption that the sum payable on breach was not liquidated damages. Next he postulated a single stipulation, such as a stipulation not to solicit customers, where the damages would be nominal if the solicitation was unsuccessful and might vary greatly according to the value of the customer if successful. As to that he said:

“Still whatever damage there is must be the same in kind for every possible breach and the fact that it may vary in amount for each particular breach has never been held to raise any presumption or inference that the sum agreed to be paid is a penalty”.

He next considered the case where a single sum is agreed to be paid on the breach of a number of stipulations of varying importance. He drew a distinction between cases in which the damage likely to accrue from each stipulation was the same in kind and cases in which it varied in kind with each stipulation. Cases of the former class seemed to him analogous to those of breach of a single stipulation. But cases in the latter class were such that the prima facie presumption or inference was against the parties having pre-estimated the damage:

“The damage likely to accrue from breaches of the various stipulations being in kind different a separate pre-estimate in the case of each stipulation would be necessary and it would not be very likely that the same result would be arrived at in respect of each kind of damage. In my opinion however any such presumption or inference would be prima facie only and capable of being displaced by other considerations.”

He postulated a circumstance in which the parties had estimated the damage from a breach of stipulation A at £5 to £ 15 and of stipulation B at £ 2 to £ 12. An agreed sum of £ 8 for every breach would be enforceable.

Lord Woolf considered the same sort of problem in Philips Hong Kong Ltd v A-G Hong Kong [1993] 61 BLR 41. The clause provided for liquidated damages at a daily rate if certain Key Dates were not met and additional damages at another daily rate if the whole of the works was not completed within a specified time. Lord Woolf was fully alive to the danger of invalidating clauses on the ground that they might provide disproportionate damages in various hypothetical situations:

“Whatever the degree of care exercised by the draftsman it will still be almost inevitable that an ingenious argument can be developed for saying that in a particular hypothetical situation a substantially higher sum will be recovered than would be recoverable if the plaintiff was required to prove his actual loss. Such a result would undermine the whole purpose of parties to a contract being able to agree beforehand what damages are recoverable in the event of a breach of contract. This would not be in the interest of either of the parties to the contract since it is to their advantage that they should be able to know with a reasonable degree of certainty the extent of their liability and the risks which they run as a result of entering into the contract”.

He posed the question:

“Is it sufficient for a contractor to identify hypothetical situations where the effect of the application of the clause may be to produce a sum payable to the employer substantially in excess of the damage which the employer is likely to suffer in order to defeat the intended effect of a clause freely entered into by the parties providing for the payment of liquidated damages?”

In seeking to answer that question he cited with approval the decision of the High Court of Australia in AMEV UDC Finance Ltd v Austin (1986) 162 CLR 170:

“But equity and the common law have long maintained a supervisory jurisdiction, not to rewrite contracts imprudently made, but to relieve against provisions which are so unconscionable or oppressive that their nature is penal rather than compensatory. The test to be adopted in drawing that distinction is one of degree and will depend on a number of circumstances, including (1) the degree of disproportion between the stipulated sum and the loss likely to be suffered by the plaintiff, a factor relevant to the oppressiveness of the term to the defendants, and (2) the nature of the relationship between the contracting parties, a factor relevant to the unconscionability of the plaintiff’s conduct in seeking to enforce the term. The courts should not, however, be too ready to find the requisite degree of disproportion lest they impinge on the parties’ freedom to settle for themselves the rights and liabilities following a breach of contract. The doctrine of penalties answers, in situations of the present kind, an important aspect of the criticism often levelled against unqualified freedom of contract, namely the possible inequality of bargaining power. In this way the courts strike a balance between the competing interests of freedom of contract and protection of weak contracting parties; see generally Atiyah, The Rise and Fall of Freedom of Contract (1979), especially Chapter 22”.

Lord Woolf’s answer to his question was the following:

“Except possibly in the case of situations where one of the parties to the contract is able to dominate the other as to the choice of the terms of a contract, it will normally be insufficient to establish that a provision is objectionably penal to identify situations where the application of the provision could result in a larger sum being recovered by the injured party than his actual loss (Footnote: ¹). Even in such situations so long as the sum payable in the event of non compliance with the contract is not extravagant, having regard to the range of losses that it could reasonably be anticipated it would have to cover at the time when the contract was made, it can still be a genuine pre-estimate of the loss that would be suffered and so a perfectly valid liquidated damage provision. The use in argument of unlikely illustrations should therefore not assist a party to defeat a provision as to liquidated damages….

A difficulty can arise where the range of possible loss is broad. Where it would be obvious that, in relation to part of the range, the liquidated damages are totally out of proportion to certain of the losses which may be incurred the failure to make special provision for those losses may result in the “liquidated damages” not being recoverable (see the decision of the Court of Appeal on very special facts in Ariston SRL v Charly Records Limited (1900) The Independent 13 April 1990). However the court has to be careful not to set too stringent a standard and bear in mind that what the parties have agreed should normally be upheld. Any other approach will lead to undesirable uncertainty in commercial contact”.

In Jeancharm Ltd v Barnet Football Club Limited [2003] EWCA Civ 58, where a provision that non-payment of an invoice should incur interest at the rate of 5% per week was treated as a penalty, Jacob J did not read the decision in Philips as departing from the law laid down by Lord Dunedin. He declined to interpret the case as supporting the contention that the test now was to ask whether the clause was an inappropriate clause having regard to the risk undertaken by the party relying upon it. He accepted that since Dunlop the courts have continued to apply its “rule” but have held that one should be careful before deciding whether or not the clause is a penalty when the parties are of equal bargaining power. But there was “no abandonment of the rule that the clause must be a genuine pre-estimate of damage”.

In a case where there is a range of possible loss attributable to the breach or breaches upon which the liquidated sum becomes payable the authorities so far considered suggest the following guidelines for the purposes of determining whether the clause is a genuine pre-estimate:

There is, however, a limitation on the utility and application of these guidelines. First, everything depends on the circumstances of the individual case. Second, the propositions are for the most part expressed with reference to presumptions or the absence of them. Even if there is a presumption it is not irrebuttable. Even if there is no presumption the clause may be penal. Third, there is, as it seems to me, something of a tension between the speech of Lord Dunedin and that of Lord Parker.

Fourth, some of the propositions merit further comment. The first proposition has sometimes been regarded as having an overarching effect, so that if it does not apply, the clause will not be penal. However, the fact that the sum provided for does not exceed the maximum conceivable loss does not mean that the clause cannot be penal. If the maximum possible loss is £ 1 million and the “liquidated damages” are £ 2 million it is not difficult to conclude that the provision is not a true estimate of loss and is penal in intent. It does not, I think, follow that, if the maximum possible loss is £ 1 million, but the likely range of loss is between £ 250,000 and £ 750,000, a figure of £ 1 million, is necessarily acceptable, particularly if the £ 1 million loss would be quite exceptional. To hold that it must be a genuine pre-estimate would, as it seems to me, take no account of (i) Lord Dunedin’s Proposition 4 (c) or his reference to gauging the strength of the chain by reference to its weakest link; (ii) Lord Parker’s observations cited in paragraph 65 above; and (iii) Lord Woolf’s reference to a case in which the liquidated damages are totally out of proportion to certain of the losses in part of the range.

Fifth, the fact that the same sum is payable on different breaches of either the same stipulation, or different stipulations where the damage likely to arise from the breaches is of the same kind, may not save the clause. The fact that there is only one stipulation involved or that loss of the same kind may result from breaches of different stipulations may not tell you very much, because the breach and the loss resulting therefrom may vary very widely in character and extent. A similarity in kind may mask a myriad of differences. If the range of possible breaches and of likely losses from such breaches is large but, on a scale from 0 to 100, most cases would fall within the 30-70 range, a figure of 90 could be hard to justify, particularly if one set of breaches could only fall in the 30-50 bracket. In the latter case a figure of 70 might be unjustifiable. Lord Parker’s denial of any presumption save in cases where there are different stipulations and the damages are likely to vary in kind according to the stipulation in question does not mean that in such cases Lord Dunedin’s reference to taking the strength of the chain at its weakest link can never apply.

There are further matters to be taken in to account in deciding whether a clause is penal:

A question arose in the course of the hearing as to whether (a) the estimate that is contemplated is an estimate of the loss which the innocent party will probably suffer or which he might suffer; and (b) whether the loss must be a loss recoverable at common law.

As to (a), different expressions have been used in different cases. In Clydebank Engineering and Shipbuilding Co Ltd v Don Jose Ramos Yzquierdo y Castaneda [1905] AC 6 Lord Davey spoke of a clause “so extravagant that it could not possibly have been regarded as damages for any possible breach which was in the contemplation of the parties” . In Commissioner of Public Works v Hills [1906] AC 368 Lord Dunedin spoke of “a genuine pre-estimate of the creditor’s probable or possibleinterest in the due performance of the principal obligation”.

In Dunlop Lord Dunedin’s first proposition related to a sum which was extravagant and unconscionable in amount in comparison with the greatest loss that could conceivably be proved to have followed from the breach. Lord Atkinson referred to a sum which was “unconscionable and extravagant having regard to any possible amount of damage likely to have been in the contemplation of the parties when they made the contract”. Lord Parmoor used almost identical language. Lord Parker said that “If ...the sum agreed to be paid is in excess of any actual damage which can possibly, or even probably arise from the breach, the possibility of …a bona fide pre-estimate of damage has always been excluded”.

In Robophone Diplock LJ described the rule as being that the court:

“will not enforce a stipulation against a party in breach if it is satisfied that the stipulated sum was not a genuine estimate of the loss likely to be sustained by the party not in breach but a sum in excess of such anticipated loss “

That was subject to the qualification that a clause might be enforceable if the sum for which it provided was justifiable in the light of damage that would be recoverable under the so called “second rule” in Hadley v Baxendale. That rule entitles a party to recover loss not arising in the ordinary course of events where knowledge of the special circumstances likely to give rise to it has been made known to the guilty party by the innocent party (or in circumstances where the contract breaker knew that the innocent party knew that the contract breaker had such knowledge), in such a way as to give rise to the inference the guilty party impliedly undertook to bear any special loss referable to a breach in those special circumstances. Thus:

“If at the time of the contract the plaintiff informs the defendant that his loss in the event of a particular breach is likely to be £ x by describing this sum as liquidated damages, and the defendant expressly undertakes to pay £ x to the plaintiff in the event of such breach the clause is not a “penalty clause” unless £ x is not a genuine and reasonable estimate by the plaintiff of the loss which he will fact be likely to obtain”.

In Murrayv Leisureplay [2005] EWCA Civ 963 Arden LJ described how in the normal situation the test will be whether or not the parties genuinely pre-estimated the loss “that would occur on breach. This is a relatively low level of review”. In LordsvaleFinance plc v Zambia [1996] QB 752 Colman J referred to a comparison between the amount that would be payable on breach with the loss that might be sustained if the breach occurred. In Alfred McAlpine Jackson J referred to the need for a substantial discrepancy between the level of damages stipulated in the contract and the level of damages likely to be suffered.

This seems to me something of a barren controversy. Insofar as the court has to decide whether the sum provided for is a genuine pre-estimate of loss the question, prima facie, is whether it is a reasonable estimate of the likely recoverable loss of the innocent party. If the breach is of a covenant against competition that loss will be whatever loss is suffered as a result of the illegitimate competitive acts. But at the date of the agreement the nature, extent and consequence of those acts will be unknown and may be difficult or impossible to predict.

Whether or not the figure is a reasonable estimate of that loss is likely to be affected by the range of possibilities. The fact that the sum exceeds the maximum possible loss may be an indicator that the provision is penal. If the sum is within the range it may be reasonable, especially since a broad brush approach may well be justified. But that may not be the case if the range is very wide, the figure is at one end of it and there are likely to be breaches which will produce trifling loss; or a range of breaches where the losses are likely to be much less than the sum provided for. The wider the range and the higher the figure within the range or, a fortiori, without, the more likely it is to amount to a penalty.

As to (b), in Robophone Diplock LJ suggested (obiter) that a clause would be penal if it permitted recovery of a sum “greater than the measure of damages to which he would be entitled at common law”. Since the basis of the doctrine is to relieve against provisions which require payment of more than could be recovered in damages, the fact that a clause provides for such recovery will indicate that it is not an estimate of recoverable loss. That is relevant in determining whether the clause is penal; but, in the light of more recent authority, it is not determinative: see Murray at para 92 below.

I have so far approached the case on the footing of a dichotomy between a genuine pre-estimate of loss on the one hand and a penalty on the other. More recent authority indicates that this is too rigid an approach. In Lordsvale Colman J had to consider a provision in two facility agreements which provided that in the event of a default in payment the borrower was to pay interest at the aggregate rate of (a) the cost of obtaining dollar deposits to fund the lender banks’ participation; (b) a margin of 1.5%; and (c) an additional, but unexplained 1%, which was claimed to be a penalty. Colman J held that the provision was not penal.

In the course of his judgment he said this:

“The speeches in Dunlop …show that whether a provision is to be treated as a penalty is a matter of construction to be resolved by asking whether at the time the contract was entered into the predominant contractual function of the provision was to deter a party from breaking the contract or to compensate the innocent party for breach. That the contractual function is deterrent rather than compensatory can be deduced by comparing the amount that would be payable on breach with the loss that might be sustained if breach occurred. Thus the presumption of penalty arises where “a single lump sum is made payable by way of compensation on the occurrence of one or more or all of several events some of which may occasion serious and others but trifling damage...”

He accepted that a small rateable increase in the interest rate payable would not amount to a penalty:

“…the borrower in default is not the same credit risk as the prospective borrower with whom the loan was first negotiated . Merely for the pre-existing rate of interest to continue to accrue on the outstanding amount of the debt would not reflect the fact that the borrower no longer has a clean record. Given that money is more expensive for a less good credit risk than for a good credit risk there would in principle seem to be no reason to deduce that a small rateable increase in interest charged prospectively upon default would have the dominant purpose of deterring default. This is not because there is in any real sense a genuine pre-estimate of loss , but because there is a good commercial reason for deducing that deterrence is not the dominant contractual purpose of the term”.

However, in Jeancharm Jacob J characterised the uplift of 1% as having been “held to be a genuine pre-estimate of loss on the basis that it indicated that the borrower was a risky borrower”.

Colman J said this:

“It is perfectly true that for upwards of a century the courts have been at pains to define penalties by means of distinguishing them for (sic) liquidated damages clauses. The question that had always had to be addressed is therefore whether the alleged penalty clause can pass muster as a genuine pre-estimate of loss. That is because the payment of liquidated damages is the most prevalent purpose for which an additional payment on breach might be required under the contract. However, the jurisdiction in relation to penalty clauses is concerned not primarily with the enforcement of inoffensive liquidated damages clauses but rather with protection against the effect of penalty clauses. There would therefore seem to be no reason in principle why a contractual provision the effect of which was to increase the consideration payable under an executory contract upon the happening of a default should be struck down as a penalty if the increase could in the circumstances be explained as commercially justifiable, provided always that the dominant purpose was not to deter the other party from the breach.”

The situations in which a clause is commercially justifiable but its dominant purpose is to deter are not readily discernible. Colman J may have had in mind an outrageous increase in the default rate.

It is apparent from this case that the fact that a payment on breach may not really be a pre-estimate of loss does not mean that it must be penal. If there is a good commercial justification for the provision that may be a ground for deducing that deterrence of breach was not the dominant purpose of the term. However, the clause in that case was sufficiently close to a pre-estimate as to cause it to be so described in Jeancharm; and it represented a modest additional compensation for the fact that the risk had turned out to be worse than anticipated and was a payment for continued use of the bank’s money on terms that properly reflected the new risk.

Colman J’s approach was approved by Mance LJ in Cine Bes Filmcilik ve Yapimilil & Anor v United International Pictures & Ors [2004] 1 CLC 401. Mance LJ regarded as valuable the observation that a dichotomy between a genuine pre-estimate of damages and a penalty clause did not necessarily cover all the possibilities:

“There are clauses which may operate on breach, but which fall into neither category and they may be commercially justifiable”.

In that case the Court of Appeal had to consider the provisions of a licence granted by UIP, a joint venture company, to Cine Bes, a Turkish cable television company, to show various UIP titles on its pay movie channel. The licence had been granted as part of the settlement of earlier litigation about an earlier licence agreement. The agreement provided that in the event that UIP as Licensor terminated the licence agreement for breach Cine Bes should pay (i) all future Licence Fees which would have fallen due up to the Expiration Date; (ii) certain specified amounts contained in an account funded by the Licencee for use by UIP and the members of the joint venture in advertising the films; (iii) all damage sustained or incurred as a result of such termination; and (iv) all actual and reasonably incurred enforcement costs including those connected with the prior litigation between the parties. In addition all rights in connection with the currently licensed films were to revert to UIP. The Court held that there was a triable issue as to whether amounts (i) and (ii) amounted to penalties.

In relation to item (iv) Mance LJ held that, insofar as the clause covered the cost of the present enforcement proceedings, they had a legitimate commercial purpose which took them outside the law of penalties. In respect of the costs of the previous litigation he did not consider that the requirement to pay such costs should be regarded as a penalty attached to Cine 5’s breach of the Agreement:

“Rather it is one of the terms on which the respondents were prepared to forego further pursuit of the prior litigation against Cine 5 and Avrupa. There could have been no objection if the respondents had stipulated for payment of such costs outright , or if they had stipulated for them but agreed to forgo payment as long as the licence agreement was entered into and duly performed. In relation to the actual costs of prior litigation I consider that the parties were free to make any agreement they wished. The mere fact that they chose to make such costs depend on the (non) performance of their fresh licence agreement, foregoing such costs in the meanwhile, does not mean that the obligation to pay such costs, which has now matured, should be regarded as a penalty imposed for breach of that agreement. Colman J’s analysis in Lordsvale of the boundaries of the doctrine relating to penalties is here in my view of relevance. The agreement regarding past litigation costs was understandable in the overall context of the settlement of the prior litigation. It would be wrong to treat it as if it was there to deter Cine 5 from or to penalise and punish Cine 5 for any default. It was an understandable and reasonable commercial condition upon which UIP was prepared to dispose of the prior litigation and to enter into the fresh licence”.

In Murray v Leisureplay [2005] IRLR 946 the claimant had a service agreement under which, on a wrongful termination, he was forthwith entitled to “1 years gross salary, pension contributions and other benefits in kind assuming that the salary pension contributions and benefits in kind had continued to be paid at the same rate as immediately prior to the day of wrongful termination”. Mr Murray’s employment was wrongly terminated with 7 ½ weeks’ notice. Burnton J held that the contractual provision was a penalty because it took no account of Mr Murray’s duty to mitigate his damages. The Court of Appeal disagreed. It held that the fact that the clause could result in a greater recovery than the actual loss did not automatically mean that without further justification the clause was penal. The clause would only be held to be a penalty if the sum stipulated was extravagant and unconscionable, which the terms in question were not.

The judgments reveal a difference in approach. Arden LJ said that “extravagant and unconscionable” had to be given a modern meaning. The real question was whether:

“the sums for which the parties have provided to be paid on breach differ substantially from the sums that would be recoverable at common law and whether there is shown to be no justification for that”.

To similar effect, in Alfred McAlpine Jackson J reconciled two strands of authority. One asked whether there was an unconscionable or extravagant disproportion between the damages stipulated in the contract and the true amount of damage likely to be suffered; the other, whether the level of damages was reasonable. He reconciled the two by saying that “there must be a substantial discrepancy between the level of damage stipulated in the contact and the level of damages which is likely to be suffered before it can be said that the agreed pre-estimate is unreasonable".

Arden LJ set out the questions which the court should ask itself in a case such as that. They were:

She held that the sum payable under the clause was likely to greatly exceed the sum which Mr Murray could have claimed if he had brought an action for breach of contract because it took no account either of mitigation or the fact that the first £ 30,000 of the compensation would be free of tax. As to the last question she found that the term agreed was not out of line with market expectations, and that the original proposal for a 3 year period had been disclosed in a prospectus and reduced to 1 year when there was objection that 3 years was too long. No one had objected to 1 year. For a year after his appointment Mr Murray could not engage in a competing business, as defined. He was reasonably entitled to demand some recompense for that and also for the fact that, if the company was taken over and he was offered comparable terms by the acquirer he would, under the agreement, have no claim for damages if he decided that his ideas were not consistent with the ethos of the new management. She took into account that it was advantageous to the company to effect a clean break and to know the measure of its exposure without lengthy litigation. She concluded that it had not been shown that the clause was not a genuine pre-estimate of damages or that it was not otherwise justifiable. The basis of the former conclusion appears to rest on the fact that the relevant clause provided that if the company was in arrears of contributions to his pension these would not be compensated for in addition to the payment. The basis of the latter appears to have been the matters referred to above.

Lord Justice Buxton took a broader approach. He cited with approval Colman J’s analysis in Lordsvale. He regarded the two alternatives of a deterrent penalty or a genuine pre-estimate of loss as alternatives with no middle ground between them: compare the different view of Mance LJ in Cine Bes.

He disagreed with an approach that always required a comparison between the liquidated and the common law damages to see if the comparison disclosed a discrepancy and then required that discrepancy to be justified as a genuine pre-estimate or by some other justification. He did so because he regarded that as introducing a rigid and inflexible element into what should be a broad and general question. It was inconsistent with the warnings by judges of high authority of the great caution needed before striking down a clause as penal. Exclusive concentration on the difference between the liquidated and the contractual damages overlooked Lord Dunedin’s principal test in Proposition 1.

Looking at the matter broadly he regarded the terms as generous but not unconscionable. It was difficult to say with confidence what might happen to Mr Murray on dismissal. Mr Murray might be unemployable in the short term. A clause which took account of mitigation would be likely to give rise to disputes which it was advantageous to avoid. A clause that made reference to the duty to mitigate would postpone payment beyond the termination date which reinforced the impression that “the traditional learning as to penalty clauses is very unlikely to fit into the dynamics of an employment contract at least when the penalty is said to be imposed on the employer”. The company would be likely to place a high value on Mr Murray’s services to the extent of including in his package generous reassurance against the possibility of dismissal. That such reassurance exceeded the likely amount of contractual damages did not render the terms penal unless the party seeking to avoid the terms could show that they met “the test of extravagance" posited by Lord Dunedin and Lord Woolf. He regarded that as a comparatively broad and simple question which would not normally call for detailed analysis of the contractual background. I do not take that to mean that the only question for consideration in any case is whether Proposition 4 (a) is engaged. That would be to ignore the remainder of the speech of Lord Dunedin as well as the opinion of Lord Woolf in Philips.

Lord Justice Clarke preferred Buxton LJ’s broader approach, insofar as there was a difference between it and the approach taken by Arden LJ. The modern approach to Lord Dunedin’s test in Dunlop was to be found in Lordsvale and Cine Bes. The comparison between the amount payable and the loss on breach was relevant but no more than a guide. Each case depended on its circumstances, and, in particular, the importance of the parties knowing what would be the financial consequences and of avoiding disputes. The onus of proof was always on the party asserting that the clause was a penalty. No or no sufficient evidence had been adduced to show that the clause was unconscionable and he did not regard it as either extravagant or unconscionable. A company was entitled to include in an employment package generous reassurance against the eventuality of dismissal.

In Euro London Appointments Ltd v Claessens International Ltd [2006] 2 Lloyd’s Rep 436; [2006] EWCA Civ 385 the claimants were an employment agency engaged by the defendants to find suitable applicants. The agency’s standard term provided for a fee to be paid in the event of a successful introduction within 7 days from the invoice. By clause 4 a proportion of the fee was refundable if the applicant’s employment terminated within 3 weeks. The proportion was on a sliding scale depending on how long the applicant stayed. If he left in week 11 or 12 only 10% would be refunded and thereafter none. Clause 4.1 provided that in order to qualify for the refund the defendants had to pay the claimants’ fee within 7 days. The latter provision was held not to engage the doctrine of penalties.

The Court of Appeal held, obiter, that even if the provision was within the rule the dominant contractual purpose was not deterrence. There was thus no reason why the condition precedent as to prompt payment should not take effect, even if the doctrine was engaged. Chadwick LJ made a careful examination of the function of the relevant clause and its importance for the agency. He regarded the need for prompt compliance with the payment obligation as going directly to the credit risk which the agency was willing to accept. In the absence of the condition precedent as to prompt payment the client would be likely not to make any payment until the end of the 12 week period and, if it could set off the potential refund against the introduction fee, would have a powerful incentive to withhold payment of that fee until it knew whether or not there would be a refund due. Meanwhile theagency would bear the credit risk. The inclusion of the condition precedent would counter the incentive to withhold payment, and provide a powerful incentive to pay the introduction fee before it was known whether there was a refund. Thus the client would bear the credit risk. The allocation of the risk to the client rather than the agency was the dominant contractual purpose of the clause.

In General Trading Company Holdings Ltd v Richmond Corporation [2008] 2 Lloyd’s Rep 475 the claimant agreed to buy GTC Mayfair which owned the General Trading Company retail business. The purchase price was £ 60,000 in cash and the provision by the buyer to the seller of £ 540,000 in loan notes. Richmond, the seller, undertook to procure a loan guarantee to enable GTC Mayfair to obtain a loan of £ 200,000 from a major high street bank for two years. The seller failed to do so and the buyer cancelled the notes. The agreement provided that if Richmond failed to procure the provision of a guarantee as required the loan notes would be cancelled. Beatson J held that the non provision of the guarantee by the date when the loan notes were cancelled was the result of the Buyer’s failure to serve the appropriate contractual notice and that the doctrine of penalties was not engaged.

He also held, obiter, that the clause was not penal. The evidence before him was that the loss attributable to an inability to purchase stock for the Christmas season (for which the loan to be guaranteed was required) would be in the region of £ 240,000. That was substantially less than the loss of payment of
£ 540,000 under the loan notes. Contrary to his initial inclination he held, applying the broader approach in Murray, that the clause was commercially justifiable, did not amount to oppression and was negotiated and freely entered into by parties of comparable bargaining power. Most significantly its predominant purpose was not to deter breach of contract. The purpose of the buyers was to protect their investment by putting in place a mechanism which would enable the company to survive. At the time the agreement was made the amount of loss that would be sustained if no guarantee and no loan was forthcoming was difficult to assess. The cancellation of the loan notes if the guarantee was not provided would improve the company’s balance sheet position and make it easier to obtain funding elsewhere.

These cases show the court adopting the broader test of whether the clause was extravagant and unconscionable with a predominant function of deterrence; and robustly declining to do so in circumstances where there was a commercial justification for the clause. That this is a reversion to the foundation of the doctrine is apparent from the observations of Lord Halsbury in Clydebank Engineering when he asked of the relevant clause:

“whether it is, what I think gave the jurisdiction to the Courts in both countries to interfere at all in an agreement between the parties, unconscionable and extravagant, and one which no Court ought to allow to be enforced”.

Against that background I turn to consider whether clauses 5.1 and 5.6 were extravagant and unreasonable (in the sense employed by Jackson J: see paragraph 92 above). I do not do so on the basis that the answer is determinative as to whether the clauses are penal, but because the issue is undoubtedly a relevant one. If the clauses are genuine pre-estimates they can scarcely be penal.

When the Agreement was made Cavendish’s damages for breach of the restrictive contracts were likely to be zero because they would be damages for the loss of the value of their shareholding and thus reflective of a loss to the Company itself. True it is that if Mr Makdessi ceased to be a director of the Company the Company could not recover damages for breach of fiduciary duty. But the Agreement contemplated that he should be a director for at least 18 months or more, as in the event he was, and he was entitled to remain a director so long as he was a shareholder unless Cavendish considered that his outside business interests were likely to result in a material ongoing conflict with his duties as a director: see clauses 14.2 and 13.2 (b). There was, thus, a strong likelihood that clauses 5.1 and 5.6 would fall to be relied on (as in the event they were) in circumstances where Mr Makdessi had become a Defaulting Shareholder whilst a director. Further clause 11 is not expressed as an undertaking in favour only of Cavendish, with the result that, even if he was not a director, the Company would have a claim against him under clause 11.

At the time of the Agreement the sums that might be withheld from Mr Makdessi under clause 5.1 were undetermined. But they could be anything from zero to over $ 44 million, depending on the OPAT valuation as at the respective Due Dates for the Interim and the Completion Payments. It is possible that a breach which caused Mr Makdessi to become a Defaulting Shareholder would so impact on the goodwill of the Company as to reduce the two final payments to nothing. But it was likely, or at any rate readily foreseeable, that the sums forfeited would be in millions or tens of millions of dollars. If the recoverable figure is zero, any estimate other than zero is excessive. A fortiori a figure in millions is extravagant in amount in comparison with the greatest loss. In consequence proposition 4 (a) of Dunlop (“it will be held to be [a] penalty”) is directly engaged.

If one leaves that consideration aside, the fact that the amount which might be withheld was indeterminate militates against the clause being a reasonable pre-estimate. Liquidated damage clauses may well incorporate formulae to determine what is payable which make them entirely reasonable. But in the present case the effect of the formula is that upon the first and any breach by Mr Makdessi of clause 1, whether material or not, (“the operative breach”), Mr Makdessi becomes a Defaulting Shareholder and no further payment of consideration in respect of the goodwill attaching to the shares purchased is to be made. The whole of what would otherwise be payable is forfeit. There is no proportionate relationship, even a rough and ready one, between the breach which triggers the operation of the clause and the amount withheld. The breach may be of any one of four stipulations. The losses likely to follow from breaches of clauses 11.2 (a) – (c) may be regarded as similar in kind, namely a loss of custom. Nevertheless the range of losses that may follow from such breaches was likely to be very large and dependent on (i) the nature and extent of the competitive activity, which could differ widely; (ii) the number of countries in which it takes place, in circumstances where the Group operated in over 15; (iii) the time for which it lasts; and (iv) the degree of its success.

I fully accept (a) that the return of Mr Makdessi to competitive activity would signal that, as the judge put it, the wolf was among the folds i.e. in the market again; (b) that that might have a very significant adverse effect, at least potentially, on the Company’s goodwill, in view of his links with clients and employees and that the extent of it might be difficult to discover and to prove; (c) that the court should not be astute to postulate hypothetical breaches of minimal significance in order to assess the chain by its weakest link, when the purpose of the clause was to address a situation where the extent of any breach was necessarily unknown, but was likely to involve the resumption of competitive activity which might well not be on a limited scale; and (d) that assessment of the possible losses from one or more breaches of clause 11 would be very difficult.

But a provision that Mr Makdessi should forfeit the totality of the outstanding price so soon as he becomes a Defaulting Shareholder seems to me an extravagant one. A trifling breach (such as an unsuccessful attempt to solicit business) or one that is very short-lived has the same effect as a breach of very substantial gravity and, insofar as there can be said to be a sliding scale, it is in the wrong direction.

The losses attributable to a breach of the covenant against employing or soliciting senior employees are likely to be different in kind to those that are likely to follow from breaches of the covenants against competition. The effect of any breach of that covenant is likely, in many circumstances, to be markedly less than a breach of the covenants against competitive activity. The losses attributable to breaches of the covenant against solicitation are, also, themselves likely to vary widely according to the nature, extent, duration and success of the solicitation. Nevertheless clause 5.1 takes effect regardless of the covenant broken or the extent of the consequences of breach.

I do not regard this analysis as resting on hypothetical but unlikely illustrations of breach, or as one that seeks, by artificially dividing the range of losses into parts, to create a cohort of losses in respect of which the effect of the clauses can be depicted as totally disproportionate. The width of the covenants and the wide range of possible losses that might result from the operative breach are such that the amount likely to be withheld will in respect of part of the range be totally out of proportion. It is, as I have said, possible that the operative breach would have such an effect that there was no Interim or Final payment to be made. But that would seem a remote contingency and wholly implausible for a range of breaches including unsuccessful attempts at soliciting business or senior employees and competitive activity limited in either scale, location or duration.

Clause 5.1 also has to be looked at in the context of the Agreement as a whole. The Agreement contains no provision which precludes the Company from bringing a claim for damages if Cavendish withholds the outstanding payments under Clause 5.1 or exercises the Call Option under clause 5.6, or vice versa. Moreover clause 6.11 entitles Cavendish to withhold from any Consideration otherwise payable its reasonable estimate of the liability for “any claim pursuant to any breach of warranty” which had not been settled. Thus, if Clause 5.1 is not a penalty, Mr Makdessi will (a) have suffered a forfeiture of the remaining instalments; and (b) incurred a liability to pay the Company damages for conduct which is said to make him a Defaulting Shareholder. The judge held that this rendered clause 5.1 penal, whereas, in my judgment, the fact that this result was very likely to occur, meant that, prima facie, it was penal.

In respect of clause 5.6 similar considerations apply. The result of the clause is to require Mr Makdessi to sell his remaining shares at a value which takes no account of goodwill and deprives him of any opportunity to exercise the Put Option. The effect is (a) that, insofar as his remaining shares have, at the date when he becomes a Defaulting Shareholder, a value which is attributable to goodwill he will not receive it; and (b) that he will be unable to exercise an option, which, when exercised, would provide a consideration which took account of goodwill.

When the Agreement was made the difference between a NAV valuation and a goodwill valuation at the date when Mr Makdessi becomes a Defaulting Shareholder or at the date of any exercise of the Put Option was necessarily unknown. But, in the nature of things it was likely to be significant. The potential loss arising from an inability to exercise the Put Option is up to
$ 75,000,000 less the proportion attributable to his percentage of the NAV at the relevant time, which he will receive under clause 5.6 which Cavendish puts at about $ 9 million.

Mr Makdessi’s inability to exercise the Put Option will arise because, if the Call Option has been validly exercised, he is not in a position to transfer the Option Shares. His inability to do so does not, as it seems to me, of itself engage the law of penalties. But the fact that one of the consequences of clause 5.6 is that he loses the benefit of the Put Option is relevant to any determination as to whether the clause is penal.

For these reasons I am satisfied that the clauses, taken in the context of the Agreement as a whole, are not genuine pre-estimates of loss. On the contrary they are extravagant and unreasonable. That is not necessarily conclusive. A commercial justification may mean that a clause which is not a genuine pre-estimate is not penal.

Cavendish submits that that the clauses are commercially justified. They were part of a commercial bargain, reached after extensive negotiation, as to the price at which shares in the Company were to change hands. Cavendish would pay the full price for the shares sold on the footing that, if Mr Makdessi did not abide by the restrictive covenants (held by the judge to be reasonable), which were there to protect the goodwill whose existence formed a major part of the price, he would not receive the full amount of it. In that event, which would indicate that it was no longer appropriate for him to remain a shareholder, he could be required to sell his remaining shareholding, valued on a recognised basis, at a price which did not reflect the goodwill, which he would have shown himself unwilling to protect in the manner that he had promised. The price at which the shares sold would be transferred and the shares retained purchased was peculiarly a matter for commercial bargaining between the parties. Clause 5.1 should not be regarded as a pre-estimate of loss but as expressing what Cavendish was prepared to pay. The fact that the Agreement could have been constructed – by making full payment conditional on Mr Makdessi not being a Defaulting Shareholder shows that the provisions had a genuine purpose. Unless the terms were absurd the Court should not intervene.

Powerfully though these submissions were developed by Ms Smith, and predisposed as I am to hold Mr Makdessi to his bargain, I do not feel able to accept them.

The underlying rationale of the doctrine of penalties is that the Court will grant relief against the enforcement of provisions for payment (or the loss of rights or the compulsory transfer of property at nil or an undervalue) in the event of breach, where the amount to be paid or lost is out of all proportion to the loss attributable to the breach. If that is so, the provisions are likely to be regarded as penal because their function is to act as a deterrent.

That seems to me the position here. The payment terms of clauses 5.1 and 5.6 do not serve to fulfil some justifiable commercial or economic function such as is exemplified in the cases – a modest extra interest in respect of a defaulting loan; a provision for the payment of the costs of earlier litigation; a generous measure of damages for wrongful dismissal; an allocation of credit risk; or the provision of capital which would be needed if a promised guarantee of a loan was not forthcoming. Their effect is that Mr Makdessi stands likely to forfeit sums in the tens of millions in circumstances where, because of the unacceptability of double recovery the law, for reasons of public policy, precludes any recovery by Cavendish at all. The Agreement prescribes a form of double jeopardy because Cavendish has the remedies provided for by the clauses and Mr Makdessi remains liable to the Company. These consequences occur upon the first, not necessarily material, breach of any one of 4 different provisions, when the range of activities which may amount to breach and of their possible consequences is likely to be very wide and to fall into different categories of seriousness many of which could not attract compensation anywhere near the value of what Mr Makdessi would forfeit or lose. This seems to me to go way beyond compensation and into the territory of deterrence. That is the predominant function of the particular terms upon which the adjustment and decoupling is to take place.

To describe the clauses as having a commercial justification because they adjust the consideration and effect a decoupling is not, in my judgment, sufficient. Clause 5.1 adjusts the consideration by withholding a large part of it. Clause 5.6 decouples Mr Makdessi from the Company by requiring him to transfer his shares at what is likely to be an undervalue. That the clauses have this effect compels the question whether they are penal, to which the fact that they adjust and decouple does not provide the answer. The important consideration is the terms upon which they do so. Thus, to take decoupling, it did not have to be on terms that nothing would be paid for goodwill, as is apparent from clause 5.2 which provided that in the case of a Seller who was dismissed and was not a Defaulting Shareholder Cavendish had the option to require the transfer to it of the Seller’s shares at sum equal to the Option Price.

Nor is clause 5.1 saved by reason of the fact that, if the Agreement had provided that payment of the Interim and Final Payments was conditional upon compliance with the provisions of clause 11, the doctrine of penalties might not have been engaged. There are many cases in which, if matters had been structured differently, the obligation to make payment or the entitlement to withhold it would have been valid because they would not have been dependent on any breach; that is one of the anomalies of the law relating to penalties. But that has not saved the provision from being held to be penal. Thus, in CMC Group PLC v Zhang [2006] EWCA Civ 408 a provision that upon any breach of a settlement
agreement Mr Zhang should pay back the $ 40,000 paid to him under the settlement was held to be penal even though, as the Court held, the position would have been different if it fell to be construed as providing for the payment of $ 40,000 conditional upon observance of the terms.

I am conscious that the approach I have adopted may be viewed as similar to that of Arden LJ in Murray with which Buxton LJ disagreed and which Clarke LJ did not prefer. It is however difficult to address any question of penalty without considering whether the provision is extravagant, and, if it is, whether there is a commercial justification. I venture to think that the difference in approach may not be as marked as it might appear provided that (a) undue significance is not given to the discrepancy between the amount payable under the clause and the loss that might be sustained on breach – the significance of the discrepancy may depend on how closely the justification relates to the nature or extent of the loss; (b) no presumptions are treated as irrebuttable; (c) proper account is taken of the desirability of upholding bargains freely entered into and of any commercial justification for allegedly penal clauses, before deciding that the predominant function of the clause is deterrent and that it is penal. That seems to me to be the case here, not least because the relevant clauses fall foul of Lord Dunedin’s Proposition 4 (a) as well as having several other indicia of their penal nature.

I am also conscious that there is a degree of ambiguity as to what is meant by the terms “extravagant” and “unconscionable” and how such descriptions fit with the concept of deterrence. “Extravagant” and “unconscionable” were terms originally used to characterise a provision which required far too high a payment in the event of breach. That it did so offended the conscience of equity, which treated it as penal – because its function was not to compensate but to deter breaches of obligations - and unenforceable (save as to the amount of the proved damage). Nowadays, when a term which provides for excessive payment on breach may be valid if it has a proper commercial justification, the term “unconscionable” would, perhaps more appropriately be used for a clause which provides for extravagant payment without sufficient commercial justification. Such a clause is likely to be regarded as penal and deterrence its predominate function, on the basis that if it requires excessive payment and lacks commercial justification for doing so, there is little room for any conclusion other than its function is to deter breach or, to put it positively, to secure performance.

I have considered whether the Court should attempt some form of scaling down of the clauses along the lines plotted in Jobson v Johnson. One possible method of doing so would be to permit the withholding under clause 5.1 to take place but only to the extent that it reflected the loss to the value of Cavendish’s shareholding attributable to the breach which led to Mr Makdessi becoming a Defaulting Shareholder; or to decree specific performance in respect of the Call Option but only on the footing that the price was either the Option Price or one that reflected the fair value of the shares taking into account the operative breach.

I do not regard either of these courses as appropriate. In circumstances where Cavendish’s loss at common law is nil scaling down is a misnomer and in any event inappropriate where what the law regards as the proper claimant – the Company – has settled its claim. To order specific performance on the terms postulated would be to rewrite the bargain the parties made, which did not specify payment for the shares the subject of the Call Option at anything other than NAV and did not provide for a fair value, let alone define it.

I would, therefore, allow the appeal and set aside (a) the declarations made by the learned judge in his order sealed on 3 January 2013, save for the declaration that clause 11.2 of the Agreement was not an unreasonable restraint of trade; and (b) the order for further directions as to specific performance made in sub-paragraph (5) of the order. The orders for costs contained in sub-paragraphs (6) – (10) will need to be set aside or varied in the light of the costs order made in consequence of this judgment.

I would also allow the cross appeal insofar as it relates to the $ 500,000. Cavendish was not liable to pay that sum.

I agree.

I agree.