The Supreme Court has reviewed the basis on which courts will uphold as non-penal a contractual term providing for a sum to be paid, or other sanction to apply, on breach of contract.  The Supreme Court has recast the rule on penalties, resulting in significant drafting and other practical lessons for those negotiating commercial contracts or seeking to impose charges for defaults in consumer arrangements.


For the first time in a century the highest court has reviewed the operation of the rule on penalties. The Supreme Court decided two very different cases: Cavendish Square Holdings BV v. Makdessi andParkingEye Ltd v. Beavis, reported together at [2015] UKSC 67. 

All the judges rejected a submission to either abolish the rule on penalties (at least in relation to commercial cases) or expand its operation, in particular to clauses other than those which are triggered by breach.  All also agreed that the current state of the authorities was unsatisfactory.  There were some differences of detail between the approach of the seven-judge Supreme Court panel to the reformulation of the rule (see below).  Lord Neuberger and Lord Sumption, in a joint judgment, held the true test for a penalty is whether the clause is a secondary obligation which imposes an obligation on the contract-breaker out of all proportion to any legitimate interest of the innocent party in the enforcement of the primary obligation.  A clause is not penal merely because it is not a genuine pre-estimate of loss, even if it operates as a deterrent.

Not all the other judges adopted the distinction between secondary and primary obligations.  But there was general consensus on the significance of the innocent party's interest in performance of the contract.  Emphasising such interest in the drafting of relevant clauses is now of increased importance.  This is especially so where the interest in performance, and the sanction for breach, goes beyond straightforward damages for (in effect) direct losses.  It will also help to build in any information showing that, even where there is a legitimate interest in performance, the remedy stipulated is not, in the circumstances, extravagant or unconscionable.

Alternatively, where feasible, it may be possible draft your term so the rules on penalties do not apply at all.  This could be by formulating the clause as a primary obligation in itself.  Thus, in the Cavendish decision, Lord Neuberger and Lord Sumption thought that the "bad leaver" provisions in the complex share sale agreement (disentitling the defaulting seller from deferred consideration for shares already transferred and requiring a sale of remaining shares at a significant discount to their original value) were, respectively, a price adjustment mechanism and a de-coupling provision, and, as such, primary obligations.  On this view, the rule against penalties was not engaged at all.  More traditionally, the effect of the rule could be circumvented by avoiding the relevant sanction being triggered by a breach.  But the Supreme Court confirmed that substance continues to trump form or labels, so the courts will be alert to identify "disguised" penalties.

A detailed analysis of the cases and the Supreme Court's decision follows.  Alternatively, use the links below to navigate to different sections or to go straight to a summary of the practical implications. 


The background facts – Cavendish Square v. Makdessi
The decisions below – Cavendish Square v. Makdessi
The background facts – ParkingEye Ltd v. Beavis
The decisions below – ParkingEye Ltd v. Beavis
The Supreme Court's decision in the conjoined appeals
How the rule on penalties has developed
The Supreme Court's restatement of the rule
Supreme Court's decision in Cavendish Square
The Supreme Court's decision in ParkingEye
Practical and drafting implications

The background facts – Cavendish Square v. Makdessi

Mr Makdessi was the founder of a group of companies which became the largest advertising and marketing enterprise in the Middle East.  He and his fellow shareholder agreed to sell a 60 per cent holding in the relevant holding company (target) to Cavendish.

Cavendish paid the initial consideration and later instalments, amounting to some US$75.5 million, to Mr Makdessi and his associate.  Further payments were due at later dates, calculated in part by reference to the audited consolidated profit after tax for specified periods.  Both the paid and deferred payments reflected considerable elements for goodwill.  Each seller also got a put option to elect to later sell their residual shareholding in the target to Cavendish at a price reflecting consolidated profit after tax over a specified period.

Mr Makdessi remained a director of the target after the transaction.  The agreement subjected him and his co-seller to various restrictive covenants.  These included not carrying on any competing business and non-solicitation of business or staff.

Clause 5 dealt with default.  It provided that on breaching any restrictive covenant the seller became a "Defaulting Shareholder" and would:

  • no longer have a right to the deferred consideration payments (clause 5.1); and
  • trigger Cavendish's right to exercise options to buy that seller's remaining shareholding in the target at a price calculated by reference to net asset value.  This was likely to be significantly less than the price the sellers would have expected to receive if they exercised their put options, subject always to the target's ongoing performance (clause 5.6).

Cavendish later alleged Mr Makdessi had breached his duty to the target by his involvement with a competing business and soliciting clients away from it.  Mr Makdessi admitted his breach of fiduciary duty to the target and settled that claim by paying US$500,000.  But his activities also amounted to a breach of the restrictive covenants under the share sale agreement.  Cavendish exercised its right under clause 5.1 to withhold further payment and exercised it call option under clause 5.6.  Mr Makdessi argued that Cavendish could not enforce those terms because they were penal.

The decisions below – Cavendish Square v. Makdessi

At first instance, the judge held the clauses were not in principle penal, either alone or in combination.  The purpose of each clause was to adjust the consideration if Mr Makdessi breached the non-compete covenants.  Clause 5.6 additionally enabled the parties to speedily de-couple their dealings in that event.  Both provisions were commercially justifiable to protect goodwill in the target.

However, the target had recovered compensation from Mr Makdessi for his breaches of duty.  This resulted in double-counting and, taken with clause 5.1, provided Cavendish with an extravagant return.  The court refused to enforce clause 5.1 unless the target agreed to repay the US$500,000.

Both sides appealed.  The Court of Appeal, applying the traditional test for penalty clauses, said neither provision was a genuine pre-estimate of loss.

Each clause was likely to come into play when Mr Makdessi was a Defaulting Shareholder and a director.  The target would thus have a claim against him for breach of duty.  Cavendish's claim for breach of the restrictive covenants was then likely to be zero.  This was because it would be based on the decrease in value of its shareholding and that loss was fully reflective of the target's own loss.  The sums that Mr Makdessi might forego if clauses 5.1 and 5.6 were validly invoked were uncertain (as dependent in part on the target's performance).  However, they were likely to run into many millions of pounds.  That was extravagant when the actual loss was zero.

Moreover, those significant sums would be withheld on Mr Makdessi's first breach of the covenants, even though the impact of that breach might have been trifling.  An example would be an unsuccessful attempt to solicit business.  Exactly the same sanction would apply to the most seriously detrimental of breaches.

The Court of Appeal also considered several more recent authorities suggesting that, even if a clause did not represent a genuine pre-estimate of loss, it might not be an unenforceable penalty if there was a commercial justification for it and its primary purpose was not to deter.  Cavendish argued in this regard that the provisions were part of the parties' commercial bargain.  The restrictive covenants were necessary to protect the target's goodwill, which formed a major plank of the valuation.  If Mr Makdessi breached the covenants, he would not receive the full price.  In those circumstances, it was also justifiable to force him to sell his shares (as it would no longer be appropriate for him to remain a shareholder) at a price that did not include value for the goodwill he had failed to protect.  The court should not interfere on matters of price and price adjustment, which were for the parties to agree.

The Court of Appeal disagreed.  The court should intervene where the amount to be paid or lost on breach is out of all proportion to the damage the breach results in.  The effect of the clauses – price adjustment and severing the parties' relationship – might well be justifiable.  But the court had to consider the terms on which those results were achieved.  There was no justification for the disparity between the breach and the sums in question.  The clauses were unenforceable penalties.

The background facts – ParkingEye Ltd v. Beavis

In contrast to the sums at stake in Makdessi, the ParkingEye case involved a parking charge of £85 levied against a motorist, Mr Beavis.  He had overstayed (by about one hour) the two-hour period of free parking allowed in a car park.  The scheme was operated by ParkingEye on behalf of the landowners, a pension fund which owned the nearby retail park. ParkingEye retained the charges generated by the scheme.  There were around 20 prominent notices at the car park indicating the terms on which cars could be parked there.  Mr Beavis refused to pay the fine imposed on him, arguing that it was penal.  Alternatively, he contended it was unfair under the Unfair Terms in Consumer Contracts Regulations 1999 and unenforceable for that reason.

The decisions below – ParkingEye Ltd v. Beavis

The judge at first instance held that anyone parking in the car park did so on the basis of the terms set out in the relevant notices.  They entered a contract with ParkingEye, including the term requiring payment of the charge for overstaying.  That charge was clearly intended to deter a motorist from overstaying.  It exceeded the loss ParkingEye was likely to suffer in that scenario (because had Mr Beavis returned on time his space would either have remained empty or been taken up for free for the next two hours by another vehicle).  However, the provision was commercially justifiable as it was not improper in purpose or manifestly excessive in amount.  Nor was it unfair under the 1999 Regulations.
The Court of Appeal upheld this decision.  Moore-Bick LJ, giving the leading judgment, pointed out that most of the reported cases on alleged penalties concern commercial contracts, where the relationship between the parties and consequences of the clause in question are wholly or principally financial in nature.  In the current context, it was inappropriate simply to compare the amount of the fine and ParkingEye's financial loss.  The clause was justified by a mix of commercial and social considerations, going beyond a simple assessment of whether the amount in question was extravagant or unconscionable.  Those considerations included the practicalities of providing parking facilities of this nature, benefiting both local consumers and retailers.  This required a mechanism to ensure the provision of free short-term parking was not abused by motorists.  Therefore, any charge had to be sufficiently high to act as a deterrent.  So long as that charge was not extravagant or unconscionable (which this one was not) it should not be struck down as penal.

The court was bolstered in this view by the fact that Parliament had considered it in the public interest to allow parking charges of this type, so long as they were clearly brought to the attention of those using the car park (see section 56/Schedule 4 of the Protection of Freedoms Act 2012).

The Court of Appeal also agreed that the provision was not unenforceable under the 1999 Regulations.

The Supreme Court's decision in the conjoined appeals

How the rule on penalties has developed

The Supreme Court deprecated the way in which the rule has developed over time.  For example, Lords Neuberger and Sumption called it "an ancient, haphazardly constructed edifice which has not weathered well".  Lord Hodge said the rule needed to escape the "straightjacket [of] … an over-rigorous emphasis on a dichotomy between a genuine pre-estimate of damages on the one hand and a penalty on the other".

Whilst reviewing the many authorities in this area, the Supreme Court gave particular attention to the prior House of Lords decision in Dunlop Pneumatic Tyre Co Ltd v. New Garage & Motor Co Ltd [1915] AC 79, generally regarded a the seminal case in the development of the principles.  The House of Lords drew a distinction between a penalty and a clause which provides for a genuine pre-estimate of loss.  A penalty was also described as a clause "in terrorem" the contract-breaker – in other words, intended to deter breach.  Lord Dunedin formulated four tests:

  1. the clause would be penal if the sum stipulated for was extravagant and unconscionable in amount compared with the greatest loss that could flow from the breach;
  2. the clause would be penal if the breach only consisted in non-payment of a sum due and the sanction exceeded the figure that should have been paid;
  3. there was a presumption a clause was penal if it specified payment of a single lump sum but the loss that could be suffered from a breach could vary from the serious to the trifling;
  4. the fact the loss flowing from the breach was difficult or impossible to calculate in advance did not mean the sum specified was not a genuine pre-estimate.

Lord Dunedin had stated that these tests might be "helpful or even conclusive … if applicable to the case under consideration" (emphasis added).  But Lords Neuberger and Sumption thought that, over the years, the tests had achieved the status of a "quasi-statutory code".  This had created a number of difficulties:

  • it ignored what lord Dunedin himself described as the essential question – whether the clause was unconscionable or extravagant;
  • whilst easier to apply to a straightforward clause specifying the damages to be paid on a breach, the four tests were not readily applicable to more complex cases; and
  • inadequate weight had been given to the other judgments in Dunlop (none of which referred to the four tests).  In particular, Lord Atkinson had adopted a broader approach, focusing on the innocent party's interest in performance of the relevant obligation.  That interest could extend beyond merely seeking compensation for the direct loss caused by the breach.

The difficulties of applying Lord Dunedin's four tests to more complex cases had resulted in the development of the "commercial justification" gloss.  Thus if, applying the traditional tests, the party seeking to uphold the clause failed to establish it constituted a genuine pre-estimate of loss, it could nonetheless still argue that there was some other justification for the clause.  Effectively, if such evidence existed, it could show that the purpose of the clause was not wholly or mainly to deter breach.

But, as Lords Neuberger and Sumption pointed out, a clause can still have a deterrent effect even if it has a commercial justification.  The rule on penalties had therefore become "the prisoner of artificial categorisation", due to the unsatisfactory distinction between, first, a penalty and a genuine pre-estimate of loss and, second, between a genuine pre-estimate of loss and a deterrent.  Similarly, Lord Hodge suggested that to focus on the disproportionality between the sum specified in the clause and damage capable of pre-estimation made sense in the context of a damages clause.  But it was an artificial concept when applied to clauses with another commercial justification.

The Supreme Court's restatement of the rule

Given these shortcomings, the Supreme Court elected to reformulate the rule against penalties.

Lords Neuberger and Sumption held that the "true test is whether the impugned provision is a secondary obligation which imposes a detriment on the contract-breaker out of all proportion to any legitimate interest of the innocent party in the enforcement of the primary obligation".  However, they also thought the four tests propounded by Lord Dunedin in Dunlop would still be perfectly adequate to decide a straightforward damages clause.

Lord Mance broadly agreed, but without the focus on the distinction between primary and secondary obligations.  He said the first step is to identify what, if any, legitimate business interest is protected by the clause.  The second step is then to decide whether the sanction stipulated is extravagant, exorbitant or unconscionable.

Lord Hodge (with whom Lord Toulson agreed) formulated the test as "whether the sum or remedy stipulated as a consequence of a breach of contract is exorbitant or unconscionable when regard is had to the innocent party's interest in the performance of the contract".

Lords Clarke and Carnwarth agreed with the other judgments on the formulation of the test.

The Supreme Court was in agreement that the rule against penalties does not only apply to clauses requiring a party that is in breach to pay his counterparty a sum of money.  It also applies to clauses disentitling the contract-breaker from receiving a sum otherwise due or calling for it to transfer property to the innocent party for nothing or at an undervalue.

There was some disparity in how the judges dealt with the interrelationship between the rule on penalties and the doctrine of relief from forfeiture.  Lords Neuberger and Sumption did not decide this issue.  But Lord Hodge thought it was possible in principle for a contractual provision, allowing for forfeiture of sums otherwise due, also to be subject to the rule against penalties.  On that analysis, if the forfeiture is wholly disproportionate to either the loss suffered by the innocent party or another justifiable commercial interest that party was seeking to protect by the clause, it would be penal.  But if it was enforceable under the rule against penalties, the court could go on to decide whether to grant relief from forfeiture on "normal" equitable grounds. 

Supreme Court's decision in Cavendish Square

The Supreme Court unanimously decided that neither clause 5.1 nor 5.6 fell foul of the rule against penalties, as reformulated.  They allowed the appeal and restored the first instance decision.

Lords Neuberger and Sumption assumed, without deciding the point, that a contractual provision might in some circumstances be penal if it disentitled the contract-breaker from receiving a sum of money which would otherwise have been due to him.  But, in this case, in their view the rule on penalties was not engaged by either clause 5.1 or 5.6.

Clause 5.1 was a price adjustment clause and as such a primary obligation.  Mr Makdessi's consideration for the shares already sold was earned in part by observing the restrictive covenants.  The clause bore no relationship, even approximate, to the loss Cavendish would actually suffer from the breach.  But it had a legitimate interest in observance of the restrictive covenants, to protect the target's goodwill.  Goodwill was critical to the value to Cavendish and Mr Makdessi's continued loyalty to the company was critical to the goodwill. Even a minor breach of the covenants would cause a significant loss of trust in Mr Makdessi: "loyalty is indivisible".  It was not for the court to calculate how much less Cavendish would have paid for the target without the assurance of Mr Makdessi's loyalty.  The parties were the best judges of how to reflect the impact of that loss of loyalty in their agreement.

On this analysis, the question of reflective loss did not arise.  Clause 5.1 was not concerned with the measure of compensation for the breach.  The correct question was whether any damages had been suffered from the breach in circumstances where the price had been adjusted downwards due to that same breach.

Lord Hodge thought there were strong reasons for agreeing that clause 5.1 was a primary rather than secondary obligation.  He did not decide this point, instead accepting that, even if it was a secondary provision operating on breach of Mr Makdessi's primary obligations, there were good reasons to uphold it based on Cavendish's interests and the fact the consequences of the clause were not disproportionate to the impact of a breach once looked at in the wider context of the parties' bargain.

Lord Mance appears to have considered that the rule on penalties was potentially engaged by clause 5.1.  However, although "not without initial hesitation", he ultimately decided it was a justifiable price adjustment provision intended to protect goodwill.  It should not be struck down as it was not extravagant, exorbitant or unconscionable.  He gave weight to the parties' equality of bargaining power in coming to that view.

As for clause 5.6, Lords Neuberger and Sumption held that the logic for the price formula for the forced sale of the retained shares was similar to the price adjustment mechanism for the transferred shares under clause 5.1.  Again, it reflected the reduced price Cavendish was prepared to pay in circumstances where it could not count on Mr Makdessi's loyalty.  Cavendish again had a legitimate interest in achieving its commercial objective: to sever relations with Mr Makdessi in such circumstances, meaning no future goodwill would be attributable to his role in the business.  This clause also operated as a primary obligation and did not engage the rule on penalties.

Lord Mance thought that a clause such as 5.6 providing for a forced transfer for no consideration or at a consideration which did not reflect the clause of the asset could be a penalty.  But he considered that, though triggered by breach, like clause 5.1 it amounted to a reshaping of the parties' primary relationship. 

Lord Hodge, however, thought clause 5.6 was a secondary obligation.  Its purpose was to deter breach.  But it fell to be interpreted against the agreement as a whole. Cavendish and Mr Makdessi had agreed that consideration would reflect Mr Makdessi's ongoing involvement with and loyalty to the company.  It was foreseeable when the parties contracted that, if he left the company having defaulted on his obligations, this would cause significant damage to goodwill and materially reduce its value.  The price formula for the forced share sale was on this basis not an unconscionable or exorbitant undervalue.

The Supreme Court's decision in ParkingEye

The Supreme Court agreed unanimously that the parking charge did not infringe the rule against penalties.  By a 6:1 majority, they held it additionally did not contravene the 1999 Regulations.

The purpose of the charge was twofold.  First, it facilitated efficient use of the car park in the interests of retailers and their customers by deterring long-stay visitors (including commuters).  Second, it generated income for ParkingEye in return for running the scheme.

The term clearly operated as a deterrent and did engage the rule against penalties.  But it did not contravene it.  ParkingEye (and the landowners) had a legitimate interest in imposing the charge on overstaying motorists.  That interest went beyond recovering any losses caused by the motorist's breach.  For the landowners, the interest lay in ensuring an efficient parking system and encouraging regular turnover of visitors for the retail park.  For ParkingEye, its interest lay in generating income to cover the running charges of the car park and to make a profit.  The charge was not extravagant or unconscionable, having regard to the terms of similar schemes across the country.  The terms upon which a motorist could use the car park were also made abundantly clear.

The majority held that the charge was also not unfair under the 1999 Regulations.  It could fall within the description of potentially unfair terms but was outside the basic test for unfairness in Regulations 5(1) and 6(1), as interpreted by the EU Court of Justice.  Even if the charge created a significant imbalance in the parties' rights and obligations arising under the contract to the consumer's detriment, that imbalance had to arise contrary to the requirements of good faith.  That test was not met here because ParkingEye and the landowners had a legitimate interest in deterring motorists from overstaying.  The charge was no higher than necessary to meet that aim.  Assessed objectively, the reasonable motorist would have agreed – and often did agree – to the charge.

Dissenting on this point, Lord Toulson thought that the charge did infringe the 1999 Regulations.  These implemented the relevant EU Directive and afforded consumers special protection.  The burden is on the supplier to show that the consumer would have agreed to the term in individual negotiations on equal terms.  It was not reasonable to assume that Mr Beavis would have so agreed.

Practical and drafting implications

  1. The Supreme Court's decision is generally positive for those attempting to uphold alleged penalty clauses in commercial contracts.  Whilst declining to abolish the rule against penalties in the context of such cases, several of the judgments refer to the fact that the clauses in Cavendish were part of "a carefully constructed contract which had been the subject of detailed negotiations over many months between two sophisticated commercial parties, dealing with each other on an equal basis with specialist, experienced and expert legal advice".  The clear suggestion is that in such cases the courts will prefer to adopt an approach that respects the parties' freedom of contract whenever possible.
  2. The Supreme Court's recasting of the rule against penalties leads to an emphasis on the innocent party's legitimate interest in performance of the contractual obligations. Deterrence is no longer an essential component, nor is the assessment of whether the clause constitutes a genuine pre-estimate of loss. Nevertheless, it is clear that in cases where the provision takes the form of a "straightforward" liquidated damages clause – in other words simply quantifying the damages payable on specified breach – application of the four tests of Lord Dunedin in Dunlop will continue to be important and often conclusive.  So in such a case consider building into your clause any information that helps show the sum stipulated is a genuine pre-estimate of loss.
  3. The Supreme Court confirmed that the rule against penalties can apply to other, more complex, types of clauses.  These can include provisions that withhold payment on breach or which require the contract-breaker to transfer property for no consideration or at an undervalue following a breach.  When drafting such clauses, ensure so far as feasible that they either do not engage or, failing that, do not contravene the rule against penalties as now reformulated.
  4. One way to achieve the former would be to draft such provisions as primary obligations rather than as secondary obligations.  But the distinction is not wholly apparent from the Supreme Court's judgments – for example there was disagreement as to whether clause 5.6 in the Cavendish contract was a primary or secondary obligation.  But such an argument will be assisted if the clause can be said to form part of the parties' substantive obligations (or, as Lord Mance described it, "reshapes" their primary obligations) rather than just spell out a remedy for breach of those primary obligations.
  5. Another way to avoid the rule on penalties from applying is to draft the clause so the remedy specified is not triggered by breach.  The Supreme Court resolutely rejected any suggestion that the rule should be extended (as, for example, the Australian courts have done) to such clauses if they operate unconscionably.  The Court of Appeal in Cavendish specifically pointed out that the agreement could have provided that the deferred consideration and Mr Makdessi's right to exercise his put option were conditional on complying with the restrictive covenants.  Then, in the Court of Appeal's view, no question of penalties would have arisen.  The Supreme Court did not expressly comment on this but several of the judges accepted that by "clever drafting" it might be possible to avoid application of the rule.  However, the court also stressed that the proper interpretation of a clause is a matter of substance over form: so "disguised" penalties can still face attack.
  6. If it is not possible to draft your clause so the rule against penalties is inapplicable, you need to maximise the chance of the wording being found not to infringe the rule.  The test put forward by Lord Hodge is perhaps the most straightforward, requiring the drafter to consider "whether the sum or remedy stipulated as a consequence of breach of contract is exorbitant or unconscionable when regard is had to the innocent party's interest in performance of the contract".   
    To maximise the chance of a clause not falling foul of this test, consider including details (for example, by way of a preamble) of what the legitimate interest in performance is and why the remedy is proportionate, unless these matters are obvious and beyond argument.  Thus, in a contract such as the one in Cavendish, reference could be made to the critical importance of the ongoing loyalty and commitment of a seller who is continuing as a director and employee and the significance of goodwill to the ultimate price to be paid.  Another example might be a supplier seeking damages for breach of a "take or pay" commitment.  Such a clause could point out any capital or other outlay incurred to meet the supply commitment which will be wasted if the customer fails to buy.  In the case of a clause increasing interest on a loan during a period of default, ensure it is possible to show the increase is commercially justifiable and proportionate, for example to the increased credit risk.
  7. Where the clause in question involves withholding sums which would otherwise be due, be aware that the rules on forfeiture may apply in addition to the rule against penalties.  To satisfy the latter, ensure that the remedy is not out of all proportion to the loss suffered by the innocent party either in money terms or in respect of damage to other interests which that party justifiably needs to protect.  Again, explicit drafting setting out those legitimate interests in preambles may help.  If the court is satisfied on this basis that the provision is not a penalty it can go on to decide whether to grant equitable relief from forfeiture.  This involves looking at the position of the parties after the breach and the circumstances in which the contract was broken.
  8. Also bear in mind that where a clause is impugned as a possible penalty the result is likely to be "all or nothing".  The Supreme Court rejected a submission that there is power to modify the terms on which the clause operates.
  9. In a consumer context it will probably be difficult if not impossible to avoid the application of the rule against penalties through drafting.  The ParkingEye decision is useful, however, in clarifying how the rule will apply including in the type of situation there where there is no economic relationship as such between supplier and consumer.  The innocent party's legitimate interest in performance of the contract will still be relevant.  The draftsman should ensure that such interest exists and that the nature and extent of the remedy stipulated for is not extravagant or unconscionable when compared with it.