- Damages and contract — Liquidated or agreed damages — Penalty clauses and damages
4.01 This chapter will examine whether the current law with regard to stipulated damages clauses can be justified by reference to a single, or multiple, policy objective(s). Lord Diplock was not perhaps a judge who lacked intellectual confidence1 yet, as Diplock (p. 92) LJ in Robophone Facilities v Blank,2 he said that he would ‘make no attempt where so many others have failed, to rationalise this common law rule’. Similarly, in the Cavendish case Lords Neuberger and Sumption referred to Lord Jessel MR as a judge who was not noted for expressing ignorance but who nonetheless could not express the purpose and origin of the power to set aside penalty clauses.3 The judicial challenge of articulating a clear justification for the penalty jurisdiction is not new, with Lord Eldon in 1801 confessing his embarrassment at being unable to ‘[a]scertain the principle on which those cases [applying the doctrine] were founded’.4 Similarly, in an American case a judge complained that the ‘contradictory decisions’ that comprise this doctrine of law yield can only ‘obscure’, rather than reveal its true underlying purpose.5 Academic writers have expressed similar sentiments using deprecatory adjectives to describe this area of the law of contract that range in strength from ‘curious’,6 through ‘anomalous’,7 to ‘contradictory’8 and ‘paradoxical’.9 Richard Posner is a prolific academic, now a judge,10 who is associated with an eponymous thesis that much of the common law ‘can best be understood as a system of rules designed to produce economically efficient outcomes’.11 A large proportion of his prolific output of twenty books and over 180 articles is devoted to demonstrating the economic logic of judge-made law. However, even he has, perhaps despairingly, concluded that the penalty jurisdiction defies ‘the Posner thesis’ and so remains ‘a major unexplained puzzle in the economic theory of the common law’.12
4.02 In the Cavendish case, counsel for Cavendish urged the Supreme Court to abolish the penalty rule.13 When an established doctrine is challenged in such a fundamental way before the highest appellate court it might be expected that its underlying (p. 93) rationale would be exposed and subject to scrutiny. Indeed, this would seem to follow a fortiori when that court is exercising the first opportunity to examine a key doctrine of commercial law in over 100 years; the last occasion being the decision of its predecessor, the House of Lords, in the Dunlop case in 1915.14 In Cavendish, the Supreme Court mainly examined the penalty jurisdiction from a historical and comparative perspective but more briefly canvassed other rationales. This chapter will examine all the ‘justificatory’ arguments discussed in Cavendish and other cases, as well as those advanced in academic writings. However, first it is necessary to examine the context of the penalty jurisdiction in the broader law of contract. In other words, the relationship between the penalty jurisdiction and the principle of self-determination, often summarised as the freedom to contract, will be discussed.15
4.03 Freedom of contract is a phrase that is perhaps more often used than defined.16 It is a concept that might fulfil Arthur Leff’s well-turned description of an ‘emotionally satisfying incantation’.17 It makes a powerful appeal to an intuitively agreeable power of contractors to fashion their own, including remedial, obligations. In the words of Jessel MR: ‘You have this paramount public policy to consider—that you are not lightly to interfere with this freedom of contract’.18
4.04 More recent cases repeat this approach of regarding the power to set aside a penalty as an atypical derogation from the usual power to contract on such terms as contractors see fit. In Phillip Bernstein (Successors) Ltd v Lydiate Textiles Ltd Diplock LJ said that:
the ordinary rule is that contracts should be enforced, pacta sunt servanda, unless they can be brought within that limited category of cases in which, for reasons of public policy, the court refuses to give effect to the agreement of the parties … One limited and well-known class is the … penalty.19
(p. 94) 4.05 In Murray v Leisureplay Plc20 the members of the Court of Appeal repeated the orthodox view that the jurisdiction to set aside a stipulated damages clause was regarded as anomalous. Arden LJ described the penalty jurisdiction as a disapplication of the usual laissez-faire approach, which permits parties the maximum degree of contractual self-determination.21 Indeed, according to Beatson J, it was this anomalous character that justifies the imposition of the burden of proving it is a penalty upon the party challenging the clause.22 The Court of Appeal repeated the view taken in Murray the following year: ‘The rule against penalties is an exception to the general principle of English law that a contract should be enforced in accordance with its terms’.23 In McAlpine v Tilebox, Colman J described the jurisdiction as a principle whose origins lay in an earlier chapter of contractual history but which atypically survived the rise of freedom of contract.24 In the Cavendish case itself, Christopher Clarke LJ in the Court of Appeal noted that the law on penalties was ‘a blatant interference with freedom of contract’,25 while in the Supreme Court Lord Hodge said:26
The rule against penalties is an exception to the general approach of the common law that parties are free to contract as they please and that the courts will enforce their agreements – pacta sunt servanda.
The same sentiment about respect for what the parties have agreed is sometimes expressed differently absent any reference to the asserted hierarchy of principle (freedom of contract) and exception (the penalty rule) described above. Rather, it takes the form of a warning that courts should not lightly interfere with concluded agreements, and that, as a contemporary commentator puts it, ‘[t]hey should not mend bargains’.27
4.06 The concept of freedom of contract presents a number of difficulties when it is referred to in the context of the policy that supports the penalty jurisdiction. First, despite the many judicial references to the concept, it lacks definition. This was exposed in a neglected28 early article by Ian Macneil,29 who argues that there were ‘two sides of freedom of contract’. The first is freedom from restraint: effectively, an immunity from legal sanctions for making or receiving (p. 95) promises.30 The second is the power of one transactor to secure legal sanctions against another when that other is in breach of its contractual commitments; Macneil believes that this second sense is better described as a ‘power to contract’. Viewed in this way, Macneil argues that the common law refusal to enforce contractual penalties did not involve any qualification of the parties’ legitimate right of self-determination. The apparent contradiction between the comforting incantation of freedom of contract and the non-enforcement of penalties is, according to Macneil, a false opposition. In his view, it would only have force against the backdrop of a system under which ‘any promise seriously meant is a contract’.31 Yet the common law’s insistence on consideration and other prerequisites to the enforcement of a contract and, perhaps more importantly, its commitment to substitutional remedies such as damages, reveal a very different legal landscape. MacNeil asserts that promises are enforced to protect reliance and prevent unjust enrichment,32 and that the formal protection of expectations is a means whereby hidden or unprovable reliance is protected. Having adopted this premise,33 MacNeil argues that the penalty doctrine is consistent with the power of contract.34 This argument builds upon an analysis of a contract as comprising two parts: a base promise, ‘the substantive parts of the contract, ie all material agreements and terms of condition, other than the agreed remedy’,35 and a stipulated damages clause. Macneil argues that the protection of the identified interests is satisfied by an award of unliquidated damages sufficient to protect the base promise alone; it is not necessary to protect reliance upon a stipulated damages clause that under the present law is considered to be an unenforceable penalty. The law’s refusal to protect the promise to pay the stipulated sum does not offend the interest analysis because the contract demonstrates the promisee’s willingness to predicate his own performance upon the base promise and the promisee, at the time of contracting, knows that stipulated damages clauses that offend the penalty rule will be unenforceable.36
4.07 Whether the account of the basis of contract ‘adopted’ by Macneil is correct cannot be considered here. However, this detailed account of Macneil’s 1962 article has demonstrated that the concept of freedom of contract may be defined (p. 96) in different ways and these different approaches will yield different conclusions about the proper limits of a power to review stipulated damages clauses. A further but different illustration of this point is provided by a more recent academic article by Scott and Triantis,37 which categorises contractual termination rights, including the right to receive stipulated damages, as ‘embedded options’ that perform important risk management functions. Although the substance of this argument cannot be examined here, one key point about this characterisation should be emphasised, which is that the stipulated damages clause is regarded as an option that, like any other commodity, is purchased by the party it is designed to benefit. In other words, the clause should be enforced in the same way as any other contractual term specifying the contractual subject matter. The comparison to the approach taken by Macneil is instructive. When the stipulated damages clause is categorised as a remedial provision (Macneil), its control is more easily justified; when it is regarded as defining the contractual subject matter (Scott and Triantis), its control is difficult to justify.
4.08 Appeal to the concept of freedom of contract is nonetheless a powerful rhetoric mechanism, ‘the force [of which] should not be underestimated’.38 It is not, however, a universal legal value39 in the sense that it never yields to any competing policy. In the High Court of Australia, Keane J has suggested that the existence of statutes that relieve against ‘unconscionable conduct’,40 ‘unjust’ transactions,41 and ‘unfair’ contractual terms42 are evidence of this. Professor Waddams has argued that appeals to the concept of freedom of contract to support arguments for the enforcement of stipulated damages clauses ‘can be met only by a general principle of contract law that gives an excuse for non-performance. The relevant principle … is unconscionability’43 (emphasis added). In contrast, Leggatt J in MSC Mediterranean v Cottonex Anstalt44 urged that a different policy goal, the pursuit of economic efficiency, should be considered as a qualifying principle in this context. The possible relevance of unconscionability and economic (p. 97) efficiency as policy factors underlying the law in this area will be explored in separate sections later in this chapter.
4.09 ‘Freedom of contract’ in the sense of individuals who are free to trade together on mutually acceptable terms is closely related to one of the standard assumptions of economic analysis.45 Economic analysis of law assumes that benefits and costs can be measured in monetary values. Rational contractors will seek to maximise the difference between their benefits and costs, which is termed utility maximisation. Such individuals when left to trade will reach agreements that maximise the benefits for each. Such an outcome is regarded as efficient because it maximises the value, represented by the amount individuals are prepared to pay, for a given set of resources or entitlements. While an outcome is efficient in the sense described, it does not address any issue of distribution, ie in what proportion the benefit from trade should be enjoyed.46 Any such efficient allocation of goods is an allocation on the basis of the initial endowments of the economic actors and the fact that goods are allocated efficiently on that basis does not imply any conclusion about the justice of those initial endowments. In this sense, standard economic analysis is silent upon such distributional questions.47 Standard economic analysis nonetheless implies a commitment to contractual freedom, even though any allocation resulting from a process of free exchange cannot itself be said to be just in the absence of agreement that the initial allocations are themselves fair.48
4.10 The economic analysis just described aligns closely with the general approach of the English common law to vitiating factors in contract. A contrast is sometimes drawn between so-called procedural and substantive unconscionability.49 Procedural unconscionability refers to defects in the process of agreement, such as that it was affected by duress, misrepresentation, mistake, etc; substantive unconscionability refers to dissatisfaction with the outcome of the contract, ie it was too one-sided.50 English common law51 is often characterised by its preparedness to intervene to set aside a contract on the basis of some factor that affects (p. 98) the process of agreement and its disinclination to do the same on the basis of a contract that was in result more beneficial to one party than the other. This is apparent from the expansive approach to so-called contractual vitiating factors in comparison to the rejection of Lord Denning’s call in Lloyd’s Bank v Bundy52 for a unified doctrine of unconscionability.53 It is also apparent within individual contract doctrines. For example, the broad proposition that courts will not enquire into the adequacy of consideration (so that an agreement to rent property for £1 per annum would be enforceable as being made of sufficient, if not adequate, consideration54) is qualified by a narrowly drawn exception when the contract is one of salvage.55 The economic model described earlier is dependent upon the securities promoted by the common law model that contractors will transact freely based upon accurate information and that their resulting contract will not be set aside merely because one party was more successful in negotiations than the other.
4.11 A different critique of the pervasive influence of the idea of freedom of contract emphasises an asymmetry in the way that this concept applies to exemption and limitation, as opposed to stipulated damages clauses. It has been suggested that such exculpatory clauses are scrutinised in a more lenient manner than contractual provisions for the payment of liquidated damages. In other words, greater importance is attached to the principle of contractual self-determination expressed in the aphorism ‘freedom to contract’ when contractual provisions seek to reduce, rather than generally to enlarge, the remedial consequences of breach.56 The argument for consistency arises from the fact that exemption clauses may, as Diplock LJ pithily explained in Robophone Facilities Ltd v Blank,57 simply be regarded as ‘penalty clauses in reverse’. The two types of clause may be regarded as opposite ends of a ‘liquidation spectrum’. As Waddams explains: ‘[b]y a natural progression of thought one turns from unduly high liquidations (penalties) to unduly low liquidations (limitation and exclusion of liability)’.58 The supposed contrast drawn between the treatment of the two different types of clause is difficult to verify. Indeed, the commentator who noted this trend in Canada immediately qualified his view: ‘the assertion that penalties are invariably harder to enforce than exculpatory clauses is by no means (p. 99) unassailable’.59 The question is really an empirical one: are more exculpatory than stipulated damages clauses upheld by the courts? However, in relation to this, as with other60 questions about stipulated damages clauses, empirical data is simply scarce.61 Nonetheless, it is suggested that the analyses referred to above underestimate the general trend towards the enforceability of stipulated damages provisions both before and, a fortiori after, the Cavendish case. In 2005, in McAlpine v Tilebox,62 Colman J noted that he was only referred to four cases63 where the relevant clause was struck down as a penalty and it would seem that one, or possibly two64 of these cases did not in fact directly involve the exercise of the penalty jurisdiction. In the period 2006–9 immediately following the McAlpine case, in three of four major reported cases containing an extended (p. 100) discussion of the penalty jurisdiction the clause was found to be enforceable.65 When all of the cases since 1950 cited in the Cavendish case where a contractual provision was held to be subject to the penalty jurisdiction were collated, it was found that the clause was enforced on almost twice as many occasions as it was held to be an unenforceable penalty. After the Cavendish decision, it is to be expected that fewer clauses will be held to be unenforceable penalties than when the older test, requiring the sum stipulated not to exceed a genuine pre-estimate of loss, from the Dunlop case applied generally. A contractual provision that provides for the payment of a sum in excess of a genuine pre-estimate of loss may now be recovered if it is nonetheless proportionate to the protection of a legitimate interest of the party seeking to enforce it. This was neatly illustrated in the Makdessi appeal, where the Supreme Court applying the ‘new’ test upheld the disputed provisions and so reversed the decision of the Court of Appeal, which, applying the ‘old’ Dunlop test, had held the clauses to be unenforceable.66 On the basis of the trends67 identified above towards a greater willingness to enforce stipulated damages provisions, it is suggested that no clear contrast can be drawn between the treatment of exculpatory clauses and contract terms providing for the payment of liquidated damages.
4.13 Judicial discussions of the penalty doctrine sometimes refer to cost advantages or savings to justify the penalty doctrine. In Robophone Facilities Ltd v (p. 101) Blank,68 Diplock LJ noted the ‘good business sense’ of contractors who included a liquidated damages clause so that ‘they should know what will be the financial consequences to them of breach on their part’69 because ‘circumstances may arise when further performance of the contract may involve them in loss’. A particular application of this observation is that such a clause enables a contractor to decide if it would be profitable to sell his performance elsewhere (to commit what is sometimes referred to as an efficient breach). The concept of ‘efficient breach’, whereby the contract breaker is enabled to ‘redirect’ his contractual performance to a third party who is prepared to pay more for it than the original contractor is an important construct in the so-called economic analysis of law and is discussed in greater depth below.70 The idea of reducing the cost of contracting and so encouraging mutually beneficial trade has a powerful intuitive appeal. Consequently, judicial discussions about the so-called transaction cost advantages of liquidated damages clauses are often expressed forcefully but at a high level of generality. A typical example is that of Lord Hodge in the Cavendish case: ‘There is beyond doubt real benefit to the parties being able to agree to the consequences of a breach of contract … Parties save on transaction costs where they can avoid expensive litigation on the consequences of breach of contract’.71 Before statements like that above can be endorsed, it is necessary to look individually at the different transaction cost advantages which these statements aggregate. An examination of these different factors and their likely consequences follows below.
(1) Saving of the parties’ and court’s time
4.14 The advantage that liquidated damages clauses bring through a saving of court time has been canvassed since Roman times; Justinian emphasised the fact that the judge was freed from the task of having to assess the evidence regarding damage.72 The power accorded to judges in some civil law systems to reduce the amount stipulated in the clause has been criticised in the past because it undermines the saving in court time that enforcement as stipulated would secure.73 Indeed, the author of a comparative study of stipulated damages clauses described this saving as ‘the sole major asset that these clauses have’.74
(p. 102) 4.15 Common law writers and judges have similarly applauded this effect. In the landmark English case of Kemble v Farren,75 an actor was employed to be the principal comedian at the Covent Garden Theatre for four seasons. He was to receive payment of £3 6s 8d for every night the theatre was open during that period. The contract further provided that if either party failed to perform its obligations, that party became liable to pay the other £1,000. The court held that the provision was unenforceable but nonetheless noted that: ‘In many cases such an agreement fixes that which it is almost impossible to be accurately ascertained; and in all cases, it saves the expense and difficulty of bringing witnesses to that point’.76
4.16 After a breach has occurred it has been said that these clauses save legal costs because the court will not be required to assess as unliquidated damages the losses of the promisee. In Diestal v Stevenson,77 a contract for the sale of coal was concluded between a coal exporter in Newcastle and a customer in Germany in volatile market conditions. The contract contained a clause providing for the payment by either party responsible for the failure to execute the agreement of 1s per ton on the unexecuted part. Upholding the clause (notwithstanding its self-description as a penalty) Kennedy J said:78
In this case I have come to the conclusion that what the parties really meant was not to leave the matter at large but in order to avoid the difficulty, as between a shipper at Newcastle and buyer at Lubeck, of proving the value of the goods in a market which is constantly fluctuating, to assess the damages beforehand.
This aim of avoiding ‘difficulty, uncertainty, delay and expense’ has been endorsed by cases79 and commentators in other common law jurisdictions. A typical commentary80 is that of Allan Farnsworth:81 ‘The advantages of stipulating in advance a sum (p. 103) payable as damages are manifold … [f]or society as a whole, it may save the time of judges, juries, and witnesses, as well as the parties and may cut the cost of litigation’. Similarly, Stephen Waddams notes that:82 there are ‘strong arguments for enabling parties to set their own value on performance’ including that: ‘it reduces the cost to the parties and to the state of settling a dispute after breach’.
4.17 Economic writers83 have also commented on the cost savings that such clauses may effect:84 ‘Liquidated damage clauses promote efficiency in contractual relationships by reducing the litigation and judicial costs that accompany breach’.85 It is suggested that the statements above exaggerate this supposed advantage in a way that is not immediately apparent. Most economic commentaries86 cannot (p. 104) explain why penalties are not enforced. In a magisterial survey of the literature Eric Posner concludes that ‘economics fails to explain contract law. It does not explain … why liquidated damages are not always enforced’.87 There is therefore ‘a growing academic consensus against the penalty rule’.88 Consequently comments like that above referring to savings in party and court costs are premised upon a jurisdiction that enforces all stipulated damages clauses. To the extent that any jurisdiction, as the common law systems of contract do,89 draws a distinction between valid liquidated damages clauses and unenforceable penalty clauses parties are required to devote greater resources90 to avoiding the unenforceability of their stipulated damages provision. Any indeterminacy in the rule,91 and we have seen in Chapter 3 that there is a great deal, will make the application of the law difficult and costly, both for the parties and for the courts. Ugo Mattei has characterised the process as a tripartite, rather than a bipartite, one: ‘The parties do not merely seek agreement amongst themselves; they must also seek the agreement of the court. This agreement, however, because of the nature of the judicial process, can only come ex post; and, moreover, it is highly uncertain’.92 The cost of this process is well illustrated by the details of a single piece of litigation, the case of Office of Fair Trading v Abbey National plc,93 which concerned the enforceability of fees levied by banks for so-called unauthorised overdrafts. The first instance judge was required to analyse the standard form banking contracts of all the major high street banks and building societies before producing a judgment that ran to no less than 450 paragraphs. The costs of the litigation were huge, with eleven Queen’s Counsel and a number of the ‘magic circle’ firms of City solicitors involved. This expenditure was presumably justified by the economic implications of the question at issue when the smallest institution involved, the Clydesdale (p. 105) Bank, had over 2.4 million personal customers in the UK! The challenge is not confined to the UK, as both the United States Supreme Court94 and the High Court of Australia95 have addressed an analogous issue, with the latter resulting in a judgment that was a ‘mere’ 376 paragraphs long!
4.18 It is suggested that the views of the legal, as well as the economic, commentators understate the costs imposed upon contractors by the current legal regime. At a general level it can be said that the prospective assessment of likely loss, which is a necessary step before the parties can draft an appropriate stipulated damages provision, will always be more time-consuming than the retrospective assessment of actual loss, which is undertaken by a court prior to making an award of unliquidated damages. This is because for the former, ‘all states of the world have to be taken into account’, whereas the latter ‘requires only information about the one situation that has really occurred’.96 However, comments such as that of Farnsworth seem to underplay the initial cost of negotiating and drafting97 the liquidated damages provision and then fail to take account of the likely response of the party identified as the payer in that clause. This can be readily demonstrated by reference to the facts of a well-known case. The Suisse Atlantique case98 involved a charter ‘for a total of two years consecutive voyages’. Voyage charters typically provide fixed periods of so-called laytime for the loading and discharge of the cargo. If these operations exceed the time allowed, or the ship is otherwise detained, liquidated damages99 termed ‘demurrage’ becomes payable at the rate of £X,000 per day. The negotiation of the applicable rate of demurrage will be contentious because of the diametrically opposed interests of the shipowner, who wants a high rate and the charterer, desirous of a lower rate. Confronted by the prospect of paying any demurrage, the rational charterer will respond with contractual clauses designed to avoid the payment of demurrage when the delay has been caused by factors beyond the charterer’s control. Typically, such clauses will provide for the suspension of ‘laytime’ on the occurrence of commonly anticipated delays to loading or discharge caused by port congestion,100 strikes bad weather, or civil commotion,101 or the more extensive exculpation of ‘any other cause beyond (p. 106) the control of the charterers’.102 The eventual choice of laytime clause will reflect the bargaining positions of the parties and the charterer’s perceived need for protection based upon, inter alia, the rate of demurrage stipulated by the shipowner. In all cases, the negotiation of the demurrage clause and any ‘counter measures’ demanded by the charterer will be time-consuming and costly.
(2) Recovery of damages for otherwise unrecoverable loss
4.19 The principal judicial remedy for breach of contract103 is an award of unliquidated damages.104 Unliquidated damages are damages that are assessed by the court as appropriate to be paid to the victim of a breach of contract in respect of his or, exceptionally, others’ losses caused by that breach. Unliquidated damages for breach of contract generally seek to compensate the victim for his loss rather than to punish the wrongdoer for his conduct. In Farley v Skinner,105 Lord Clyde stated that: ‘damages should not be awarded, unless perhaps nominally, for the fact of a breach of contract distinct from the consequences of the breach’. This general compensatory aim106 has been said to be ‘a bedrock of our law’107 and means that, in the absence of provable loss, only nominal damages will be awarded.108 The assessment of compensatory damages may be regarded as a two-stage process; first, the prima facie measure is ascertained109 and, secondly, the potential effect of (p. 107) a number of limiting factors are considered, such as remoteness, mitigation, and contributory negligence.
4.20 From the above it can be seen that there are two distinct circumstances when unliquidated damages may be unavailable to a contractor who has in fact suffered loss as a result of a breach of contract. The first is where, for some reason, the party who has incurred a loss is unable to prove that fact to the required standard of proof.110 In order to succeed in an action for breach of contract, a claimant must establish liability upon the normal civil burden of proof, ie the balance of probabilities. By extension, it might be thought that the claimant’s right to recover substantial damages (where loss must be shown) was also dependent upon proof of loss to the same standard. However, as Lord Nicholls observes in Gregg v Scott:111 ‘the common law has drawn a distinction between proof of past facts and proof of future prospects’. A past event either did, or did not, occur and falls to be established in civil proceedings upon the usual balance of probabilities. Claims for wasted expenditure will fall within this category.112 However, where the loss being claimed requires an investigation of future events, a different approach is taken;113 ‘[p]roof of future possibilities is approached differently’.114 Damages are only recoverable if the item of loss claimed for can be proven with reasonable certainty.115 The policy behind the application of different burdens of proof in respect of past facts and future events is unclear. It is possible that the risk of exaggeration and falsification in relation to the latter is thought to justify the imposition of a higher standard of proof.116 An empirical study of the tourism industry produced evidence that contractors were often unable to pre-estimate in advance the likely impact upon them of the cancellation of their holiday.117
4.21 The second situation is where the claimant is able to prove the fact of loss to the required standard but the loss claimed is nonetheless irrecoverable. This might occur in a number of circumstances. When the loss is not a financial (p. 108) one it might not be recoverable under the current law. Although the former prohibition, associated with the Addis case,118 upon the recovery of non-pecuniary losses as damages for breach of contract no longer prevails generally,119 it is clear that the recovery of such damages is still more restricted than is the case with pecuniary losses. At present it seems that contractual damages for non-pecuniary loss will be recoverable in a number of defined circumstances.120 Goetz and Scott121 have argued that the enforcement of stipulated damages clauses is justified by the need to empower contractors to protect themselves against so-styled ‘idiosyncratic loss’, which would not be compensable in an action for unliquidated damages. They give an example of an alumnus who is a fanatical supporter of the college baseball team, which has just advanced to the final of an important baseball competition. The ‘anxious alumnus’122 attaches a special importance to attending the game with a group of his friends. When he contracts for the provision of a bus to take them to the game he includes in the contract a substantial sum of money to be paid in the event that the transport they have booked to take them to the game arrives late. In the example, the sum stipulated represents no more than their own idiosyncratic subjective losses which would not be compensable under the current law.123 The arguments developed here in respect of the need to permit contractors to protect themselves against non-compensable non-pecuniary losses will also apply to some claims for pecuniary losses. This possibility arises because of the limits based upon the recovery of damages for (unliquidated) pecuniary losses by the second stage of damage assessment described above, especially under the principles of remoteness and mitigation. In (p. 109) Robophone Facilities v Blank,124 Diplock LJ suggested that a stipulated damages clause that provided for compensation of losses that would be too remote to be recoverable as unliquidated damages might nonetheless be enforceable.125 The situation may be different where the doctrine of mitigation, rather than remoteness, is involved. It has been proposed that although limitations imposed by the doctrine of remoteness should be ignored, in the interests of avoiding waste, those imposed by the doctrine of mitigation should be considered relevant in this context.126 In any event, it seems that this discussion will only be relevant to the greatly reduced ‘straightforward’ cases, where the Dunlop case will continue to apply. In all other cases, the emphasis in the Cavendish case upon the legitimate interest of the party seeking to enforce the stipulated damages clause does not require a discussion of this divergence between a claimant’s actual loss and the amount that would be recoverable by way of unliquidated damages.
4.22 In the first edition of his seminal work, The Economic Analysis of Law, Richard Posner suggested that stipulated damages clauses127 can play a valuable role by allowing the promisor to ‘signal’128 or communicate the sincerity of his (p. 110) undertaking to perform by agreeing to pay a substantial sum of money if he defaults. Later, after his appointment as a judge, he again referred to this facility in a case where a party had previously acquired a reputation for defaulting on promises of payment.129 This ability to ‘signal’ sincerity may be especially valued by a particular class of parties contemplating entering into a contract. Contractors who have no established reputation in a market will probably experience difficulty in convincing others of their reliability. For such new entrants to an industry or market, a large stipulated damages clause might operate as a substitute for having an established reputation. Indeed, it has been argued that such a clause will probably be preferred by a new ‘player’ in a market as ‘being less costly … than offering a lower price.130 This possible function has been promoted in the US131 and Canada.132
4.23 The narrow justification for enforcing contractual provisions that would otherwise be struck down as penalties advanced above would involve the courts in a difficult enquiry if the enforcement was to be limited to the justifying circumstances (whether or not these extended beyond new entrants to a market). The argument has not yet formed the basis of any decision in the UK or the US. In Lake River Corp v Carborundum Co,133 Judge Posner had to decide whether a clause in a packaging contract whereby one party undertook to ship a minimum quantity of goods to the other for packing and if in default to pay for the difference between the quantity of goods supplied and the quantity guaranteed was a penalty. Even Judge Posner, who in an academic capacity had made the argument for greater enforcement of such clauses, was characterised as ‘playing against type’ because, although he did not believe that the penalty rule ‘made sense’ with respect to ‘agreements negotiated by sophisticated business parties’, he nonetheless applied Illinois law and held the clause to be unenforceable.134
4.24 A similar argument to that used above to justify the enforcement of stipulated damages clauses has been applied to so-called specific performance clauses, ie contractual provisions that amount to an express and explicit guarantee of contractual performance.135 The enforcement of specific performance clauses is sometimes objected to on the ground that to allow their enforcement would usurp the (p. 111) unique role of the court in deciding upon the appropriate remedy for breach of contract. The non-delegable nature of this remedial choice is said to be a long-established characteristic of common, as opposed to civil, law systems:
[C]ommon law … courts have never been weak institutional actors. They have a tradition of fighting for jurisdiction that dates back to the early period of the Common Law. Still today the courts are reluctant to surrender jurisdiction.136
4.25 This objection is, of course, as applicable to stipulated damage provisions as it is to contractual clauses ‘guaranteeing’ specific performance. However, the objection appears to be applied selectively ie more often to the latter, than the former, category of terms.137 The selective application of this objection and the associated asymmetry between the legal treatment of terms that seek to enlarge (stipulated damage and specific performance clauses) as opposed to reduce (exemption and limitation clauses) contractual remedies138 would seem to suggest that it may conceal, rather than itself constitute, a valid objection to the enforcement of contractual provisions.
4.26 The origins of the current penalty doctrine in the principles applied by the Lord Chancellor to ameliorate the harshness of the common law and which ultimately became assimilated into the common law were described in Chapter 1. Early commentators such as McCormick in the first 1935 edition of his authoritative book on the law of damages in the US noted these origins and so classified the penalty doctrine as one illustration of a situation where: ‘the freedom usually allowed to parties to contract as they please has been curtailed because of the unusual danger of oppressive and extortionate bargains’.139 In the same work, however, the author noted the degree to which the penalty doctrine was now removed from those origins to the extent that: ‘to-day the pendulum is swinging in the other direction, and the tendency is rather towards resolving the doubt in favour of the enforceability of such clauses as agreements for liquidated damages’.140 In the year after McCormick’s book was published (but with no reference to it), Lord Wright MR in Imperial Tobacco Co v Parsley141 reaffirmed that the test to be (p. 112) applied to identify a penalty clause did not depend upon the proof of actual, or even potential, advantage taking:
A millionaire may enter into a contract in which he is to pay liquidated damages, or a poor man may enter into a similar contract with a millionaire, but in each case the question is exactly the same, namely, whether the sum stipulated as damages for the breach was exorbitant or extravagant …142
4.27 An important distinction needs to be made between the extent to which the language, as opposed to the substantive investigation, of unconscionability survives in the formulation of the penalty rule. In the Dunlop case, Lord Dunedin stated that a sum would be a penalty if it was ‘extravagant or unconscionable in amount’ (emphasis added). However, the reference to unconscionability does not describe an independent test or justification for the categorisation of a contractual provision as a penalty because the quotation continues: ‘if the sum stipulated for is extravagant and unconscionable in amount in comparison with the greatest loss that could conceivably be proved to have followed from the breach’. Thus, the descriptor ‘unconscionable’ is merely a label for a conclusion of law that follows from the application of a test that is described in the second part of the quotation. This non-substantive survival of the language of unconscionability can be seen in older143 and more recent cases144 in the UK. It can also be seen in the ‘reformulation’ of the test for a penalty proposed by some members of the Supreme Court in the Cavendish case, with the difference that the point of comparison is not, as in Dunlop, a genuine pre-estimate of likely loss, but rather the legitimate interest which the clause seeks to protect. Lord Mance says that: ‘it is necessary in each case to consider, first … whether any legitimate business interest is served and protected by the clause, and, second, whether the provision made for the interest is nevertheless in the circumstances extravagant, exorbitant or unconscionable’.145 Similarly, Lord Hodge under a heading of ‘The true test for a penalty’ says that the ‘essence of the test’ was whether ‘the secondary obligation was exorbitant and unconscionable’.146 Interestingly, in Lord Mance’s formulation, exorbitance and unconscionability (together with extravagance) are proffered as alternatives, whereas Lord Hodge appeared to regard them as cumulative requirements. In contrast to Lords Mance and Hodge, the ‘true test’ formulated (p. 113) by Lords Neuberger and Sumption does not refer to unconscionability at all.147 Recent cases in Australia also use the language of unconscionability.148
4.28 The statement of Lord Wright MR above was cited with approval by Lords Neuberger and Sumption, as well as by Lord Hodge in the Cavendish case.149 The judgments acknowledge the origins of the rule in the ‘concern of the courts to prevent exploitation’ at a time when credit was scarce and borrowers consequently ‘particularly vulnerable’.150 However, although originally motivated by a desire to prevent the oppression of the weaker, by the more powerful, party to a transaction151 its current application does not depend at all upon any disparity of power between the parties. Lord Mance forcefully rejected the argument of counsel that the penalty doctrine should be confined to demonstrated instances of ‘procedural misconduct, involving duress, undue influence, misrepresentation, or something similar’ as having ‘no basis in authority or principle’.152
4.29 The judgments in Cavendish are of course authoritative statements of UK law. A different approach has been urged by some academic commentators. Waddams has argued that because ‘unconscionability or a similar principle of fairness’ is ‘widely recognized’, there is ‘strong reason to assimilate the law relating to penalty clauses’.153 This approach has been urged by other writers154 and endorsed judicially in other jurisdictions. In Elsley v JG Collins Insurance Agencies, Dickson J (p. 114) in the Canadian Supreme Court said that: ‘[t]he power to strike down a penalty clause … is designed for the sole purpose of providing relief against oppression for the party having to pay the stipulated sum. It has no place where there is no oppression’.155
4.30 In AMEV-UDC Finance Ltd v Austin,156 Mason and Wilson JJ suggested that the appropriate test for a penalty should depend upon a number of circumstances, including: ‘the oppressiveness of the term … and … the nature of the relationship between the contracting parties’.
4.31 Indeed, there were early signs that the UK’s domestic law might follow this approach. Referring to the comments of Dickson J, Tomlinson J in Indian Airlines Ltd v GIA International Ltd157 expressed the view that there has been ‘something of a sea change in the approach of the courts to penalty clauses’. However, the following year, in another decision at first instance, Jacob J expressed the contrary view that the principles laid down in Dunlop were not altered by the consideration of Australian and Canadian decisions (including the comments of Dickson J in Elsley considered above).158 Although Elsley was not expressly considered by the Court of Appeal in Murray v Leisureplay plc, Arden LJ expressed her clear view that she did ‘not consider that oppression … is of itself a criterion in determining whether a contractual sum is a penalty’.159
4.32 Even before the decision of the Supreme Court in the Cavendish case it was clear in this country160 and to some extent elsewhere161 that restraint of unconscionable or oppressive conduct should not be regarded as the underlying rationale of the penalty doctrine. This was confirmed unequivocally in the statements by Lords Neuberger and Sumption, Lord Mance and Lord Hodge discussed above. However, in the Cavendish case it was said that the circumstances in which the contract was made are not regarded as ‘entirely irrelevant’.162 From the passage that follows it would appear that there are two aspects to this consideration. The (p. 115) first aspect, which we can characterise as the ‘positive’ aspect, is pro enforcement, Lords Neuberger and Sumption saying that:163
In a negotiated contract between properly advised parties of comparable bargaining power, the strong initial presumption must be that the parties themselves are the best judges of what is legitimate in a provision dealing with the consequences of breach.
This positive aspect is no more than an endorsement of the oft-repeated statement164 that courts should not be overzealous when applying the penalty rule to clauses negotiated between parties of equal bargaining power. The second aspect, which we can characterise as ‘negative’, gives some indication as to when the parties’ individual circumstances might support the court’s holding of non-enforceability. In the Cavendish case in the same paragraph referred to above, Lords Neuberger and Sumption appeared to endorse the reference by Lord Woolf in Philips Hong Kong Ltd v Attorney General of Hong Kong165 to the possibility of taking into account the fact that ‘one of the parties to the contract is able to dominate the other as to the choice of terms of a contract’.166 The purpose of this endorsement is unclear. From its presentation as the endorsement of a statement made elsewhere without further principled support it cannot be reasonably interpreted as support for the principle of inequality of bargaining power as the underlying rationale of the penalty jurisdiction. Rather, like the positive aspect described above, it might provide guidance in a general way as to how hard the courts should scrutinise a clause to ascertain the existence of the required, justifying, legitimate interest. If this factor is to be taken account of to any greater extent, it would be necessary for the courts to define clearly the threshold test for its application to avoid the uncertainty generated by over general appeals to fairness or unconscionability standards.167
4.33 It is therefore clear that, whatever its origins and despite some lingering vocabulary, the principles of unconscionability cannot yield a coherent rationale for the law relating to stipulated damages provisions, thereby endorsing the prescient (p. 116) comment of Ian Macneil in 1962 that ‘[t]he oft repeated reference to oppression in penalty cases somehow sounds curiously flat, as it does in the whole history of relief against forfeitures and penalties’.168 Indeed, this conclusion is supported by further comparisons. If the sole justification for the penalty doctrine was to prevent a weaker party from being oppressed by a stronger party it is difficult to understand why only stipulated damages provisions are rendered unenforceable as other types of clause may be similarly drafted to the detriment of the oppressed party.169 A similar argument from consistency has been made to the effect that in several other respects the law of contract does not seem to regard ‘overcompensation’ as unconscionable.170
4.34 In the discussion above the arguments of a number of economic writers have already been considered,171 eg that all stipulated damages should be enforced because the contingent right to damages is something which has been paid for,172 that the enforcement of stipulated damages provisions effects time and cost savings for both the parties and the courts,173 that stipulated damages permit the recovery of idiosyncratic loss such as that suffered by Goetz and Scott’s ‘anxious alumnus’174 or Mattei’s ‘nervous father’175 and that such clauses perform an important ‘signalling’ function.176 A number of other economic arguments emphasising economic efficiency and the idea of efficient breach which was introduced in section C will now be examined. It should be noted that the overwhelming majority of articles (p. 117) in this tradition support the abandonment of the current distinction between enforceable liquidated damages clauses and unenforceable penalties.177 For this reason that rare species, the economic argument in favour of the present regime, will be examined first.
4.35 In an important article Clarkson, Miller, and Muris178 note that when contractual clauses stipulate for the recovery of damages greater than the ‘victim’s’ actual loss, the victim, will prefer it if the contract is breached rather than performed. The authors then describe the ways in which such a victim may manoeuvre his contractual partner so that the latter is in breach of contract, avoiding any conduct on the part of the victim that would itself constitute a breach of contract and thereby give rise to a counterclaim. An early English case179 provides a ‘colourful’ example of breach inducement where the defendant undertook to secure a marriage between the plaintiff and Bridgette Palmer or to pay the plaintiff £80. When the plaintiff sued successfully to recover £80, the defendant argued that the plaintiff had approached Bridgette Palmer, called her a whore, and warned her that she would be tied to a post if the marriage went ahead. The potential for breach inducement in a commercial setting can be illustrated by the facts of a case180 where the parties contracted for the sale of 2,600 tons of rails. Rails were only available in suitable quantities at two locations. The buyer began negotiations to purchase a very large quantity of rails from one of the suppliers with whom the seller was negotiating for the purchase of the contract quantity to meet his commitment to the buyer. Clarkson, Miller, and Muris justify the retention of a, though not necessarily the current, distinction between liquidated damages and penalty clauses as necessary to deter parties from engaging in wasteful, breach-inducement activity.181
4.36 Whereas the argument of Clarkson, Miller, and Muris builds upon the parties’ behaviour towards each other, Rubin182 emphasises third-party effects called externalities. He argues that stipulated damages clauses are not ‘self-enforcing’; they (p. 118) can be enforced only with a court’s assistance and the costs of trials are not borne by the parties themselves. These costs that are borne mainly by the state can be justified because litigation results in precedents and rules derived from them, which will benefit others in the future. However, in the case of litigation concerning the enforcement of penalty clauses, no such general rules are generated for the benefit of others, according to Rubin.183 Whether the current non-enforcement of penalty clauses reduces litigation is, Rubin conceded, an unproven empirical assertion.184 Indeed, the opposite conclusion, ie that the uncertainty inherent in the current law will result in more, not less, litigation is intuitively credible.185
4.37 A weakness of Rubin and Clarkson, Miller, and Muris’s argument is that both depend upon empirical assertions about, respectively, the effect of uncertainty in the modern law of stipulated damages upon the volume of litigation and the potential for contractors to engage in wasteful breach inducement activity. It is a common and oft repeated criticism of the economic analysis of law that its policy prescriptions often depend upon unevidenced empirical assertions. A typical statement is that of Macneil:186
Neither the quantity nor the quality of existing empirical ‘proof’ of the efficiency model in the realm of contract remedies law is remotely close to proving its accuracy in that area. When empirical proof is lacking, the sole remaining test of the usefulness of a model is the quality of its assumption.
4.38 The concept of the ‘efficient breach’ of contract was introduced above.187 It refers to the situation where the breach, as opposed to the performance, of a contract benefits the promisor, and the promisee, usually because of an award of damages which fully compensates for his loss, suffers no detriment.188 Consider a simple example where A contracts to supply a machine to B for £100,000. A third party, C, who believes he can put the machine to profitable use even if he pays (p. 119) £200,000 for it may offer say, £150,000 for the machine. Economic theory urges that resources should be in the hands of those who value them the most because, in this way, benefits from trade are exploited and wealth maximised.189 This can be achieved in a way that makes no-one worse off if A is able to breach his contract with B and sell the machine to C. The substitution of a sale to C for that to B is said to be an efficient breach of the contract to sell to B when that breach results in full compensation for B. Upon the assumption of full compensation, B is indifferent between performance and breach, because the latter leaves him no worse off than the former and A is desirous of breach because he will be better off.
The only universal consequence of a legally binding promise is that the law makes the promisor pay damages if the promised event does not come to pass. In every case it leaves him free from interference … and therefore free to break his contract if he chooses.
Holmes thought of contracts as ‘options’, ie when you enter a contract and promise a performance you are buying an option either to perform or to pay damages. This option feature is ‘particularly pronounced’ when applied to a contract containing a stipulated damages clause; you are undertaking that you will either perform or tender the amount stipulated in the clause.192 The modern theory of efficient breach is a ‘variation and systematic extension of Holme’s outlook on contractual remedy’ which has been embraced enthusiastically by proponents of the economic analysis of law, ‘endowed … with economic apparel and terminology’193 and now forms a central part of the wider economic analysis of remedies. It has also been used in cases in Canada,194 the US195 and, more recently, in the UK. The more general promotion of remedial efficiency has been endorsed by Leggatt J in MSC Mediterranean v Cottonex Anstalt:
While the common law attaches very great importance, however, to enforcing contracts which parties have freely entered into, this is not the only value which the English law of contract upholds. Another important objective of the law is encouraging efficiency and the productive use of resources.196
It is argued by proponents of efficient breach theory that when damages for breach of contract are set at the innocent party’s full expectation loss197 the efficient breach of contract is encouraged and the inefficient breach of contract discouraged.198 As Shavell explains, damages in excess of the expectation measure ‘might induce performance when it is inefficient’ whereas lower awards ‘might induce breach when that is inefficient’.199 Writers have also objected to the more general availability of the remedy of specific performance, as opposed to an award of compensatory damages, on the ground that it would discourage efficient breach.200 Richard Posner’s statement is typically concise and direct when he says that a general entitlement to specific performance ‘would thwart some efficient breaches’.201 Parallel arguments are made with respect to the enforcement of stipulated damages clauses ie that the enforcement of such clauses may prevent efficient breach of contract. As Klass observed ‘[s]uper compensatory damages (penalties or punitive damages) threaten efficient breach’ for the following reason:202 ‘The bilateral (p. 121) monopoly every contract creates makes renegotiation difficult, meaning a promisor faced with a penalty or punitive damages is likely to perform in situations where there is more value in breach’. So-called super compensatory damages will also set up undesirable incentives at an earlier point in time.203 A promisor may be motivated by the prospect of an onerous payment in the event of his breach to overspend upon precautions to ensure he is able to perform while a promisee may underspend on protecting himself against the risk of non-performance. Further, as we have already seen the promisee may engage in wasteful and non-productive activity to attempt to induce the promisor to breach the contract.204 The overall effect of these incentives is to reduce the potential benefit of the contract.205
4.40 The concept of ‘efficient breach’ has attracted criticism of different kinds.206 The first stream of criticism comes from contract writers who in different ways assert a moral dimension to the law of contract which does sit easily with the freedom to breach associated with the efficient breach approach.207 A related critique points out that the underlying policy goal of moving resources to their most highly valued uses could be applied to property, as well as contractual, rights where it would lead logically, but unacceptably, to a theory of ‘efficient theft’ or ‘efficient conversion’.208 A second line of criticism states that the efficient breach model makes assumptions about contractor’s behaviour as wealth maximisers that are unwarranted and even ‘facile’.209 This criticism leads to a methodological debate about the role of assumptions in economic theory and a demand that the traditional assumptions are relaxed or modified where situational complexity impedes optimising behaviour.210 The end point of this development is a new movement within legal scholarship called behavioural decision theory or behavioural law and (p. 122) economics. This movement ‘builds upon the core insights of law-and-economics scholarship, but seriously scrutinizes the shortcomings of rational choice theory’ and is discussed further below.211 A third critique emphasises that court awards of damages for breach of contract only rarely fully compensate the ‘innocent’ party. The cumulative effect of the limiting factors of remoteness, mitigation, causation, and contributory negligence combined with restrictive approaches to some categories of harm such as non-pecuniary loss is that compensation on an indemnity basis is never achieved. When compensation is imperfect the promise will not be indifferent between performance and breach so that breach is no longer efficient.212 A fourth critique asserts that the proponents of efficient breach often underestimate the cost of the type of post breach negotiation which their model builds on. As Macneil points out: ‘[i]n the real world “breach” is not a very nice word; almost no-one ever admits to breaching, and to accuse someone of breaching a contract is to start a fight’.213 He therefore concludes that ‘[i]n the real world of commerce’ the ‘opportunities for gain through “efficient breach” of transaction in goods … are so rare as to be almost nonexistent’.214
4.41 It can be seen that there is no agreement between commentators as to the value and implications of the concept of efficient breach. However, these criticisms of the concept discussed in this section are to some extent contradicted by a small empirical study which offers experimental evidence that contractual partners are ‘more willing to exploit efficient-breach opportunities when the contract in question includes a liquidated damages clause.215
4.42 At the beginning of this chapter216 we noted the close association of the intuitively appealing idea of so-called freedom of contract and the standard assumption of (p. 123) the economic analysis of law that individuals will act rationally to maximise their gains from trade. Melvin Eisenberg has summarised the traditional position:217
Parties are normally the best judges of their own utility, and normally reveal their determinations of utility in their promises. Bargain promises are normally made in a deliberative manner for personal gain, and promises so made should normally be kept. Bargains normally create value, enable the parties to plan their future conduct reliably, allocate commodities to their highest-valued uses, and best distribute the factors of production, and the enforcements of bargain promises promotes these desirable ends. Ultimately, these propositions, and therefore the bargain principle itself, rest on the empirical premise that in making a bargain a contracting party will act with full cognition to rationally maximize his subjective expected utility.
These propositions are the basis of the economic analysis of law218 which had ‘become the dominant academic style of contract theory’219 (although this is more evident in the US220 than the UK221). In the immediately preceding section222 we noted, but did not expand upon, the challenge to these foundational concepts, particularly the problem of rationality.223 Put simply, behavioural scientists have discovered that ‘such [economic] models are often wrong in the simple sense that (p. 124) they yield inaccurate prediction’.224 The body of literature they have produced, called inter alia225 behavioural decision theory226 (BDT), replaces the simplifying assumptions of economists with a richer and more accurate account of human behaviour. The best way to understand and model rational behaviour should not be, as economists do, ‘to assume away … pervasive human frailties’ instead it should be ‘to learn how people of limited capacities learn to cope’.227 The starting point for this task is the explicit but generalised idea that that the rationality of an individual is limited.228 Behaviour is said to be ‘intendedly rational, but only limitedly so’.229 Psychologists and experimental economists have produced an array of evidence that demonstrates that the factors affecting individuals when making a choice are multiple and complex but that their behaviour is nonetheless neither unpredictable nor random. Rather, it can be explained by reference to ‘heuristics, effects and biases’.
4.43 A heuristic is a ‘mental shortcut’230 that is used by individuals when processing information. An example is the so-called availability heuristic, which refers to the tendency to simplify the processing of information by giving greater emphasis to (p. 125) current and recent events which are easiest to recall.231 A different phenomena is the dispositional effect of over-optimism,232 the essence of which is captured by an instinct expressed elegantly by McCormick in 1935 in the first edition of his Handbook on the Law of Damages:
It is characteristic of men, however, that they are likely to be guided by the ‘illusions of hope’, and to feel so certain of their ability to carry out their engagements in future, that their confidence leads them to be willing to make extravagant promises and commitments as to what they are willing to suffer if they fail.233
Many experiments have demonstrated the truth of this intuition generalised to many new contexts, showing that individuals are systematically over-optimistic about planned actions. Worryingly, 90 per cent of drivers believe they are more skilful than the average234 and a sampling of individuals who were about to get married estimated their own prospects of divorce as zero but those of others, accurately, as about 50 per cent.235 An example of a bias is the so-called status quo bias, whereby people systematically prefer the preservation of a current state of affairs than switching to an alternative if one is available.236 A further bias is based upon hindsight and has been called ‘the probabilistic version of “I told you so”’.237 In hindsight, there is a tendency for people consistently to exaggerate what could (p. 126) have been anticipated. They think of what has happened as inevitable and retrospectively think that, to them and others, it must in the past have seemed certain to occur.238 A different factor affecting decision-making has been termed ‘ambiguity aversion’; certainty is preferred to uncertainty.239 This has been demonstrated to be the case, even where the uncertain option is more valuable. On this basis, someone would prefer to receive a guaranteed £100 than a 15 per cent chance of £1,000 (which should, by the law of probability, be valued at £150).
4.44 A small number of commentators have sought to apply these insights to the law relating to stipulated damages provisions. Melvin Eisenberg argues that:240
To begin with, bounded rationality and rational ignorance have a special bearing on liquidated damages provisions. The justification for the special scrutiny is not that liquidated damages provisions are specially amenable to advantage taking and oppression, but that such provisions are systematically more likely to be the products of the limits of cognition than performance terms, that is, terms that specify the performance each party is to render.
4.45 It is argued that contractors will easily focus upon the terms of the contract that define performance such as those that specify the contractual subject matter, the quantity to be delivered, the timing of delivery and the price to be paid. In contrast, at the time of contracting it is often impossible or at least impractical to anticipate all the different ways in which the contract might be breached and a fortiori to plan an appropriate stipulated damages provision to cater for all these contingencies.241 It is clear that Eisenberg’s argument to this point builds upon a general sense of contractors’ cognitive impoverishment rather than any identified heuristic, effect or bias. However, this assertion is then bolstered by reference to the effect of over-optimism and the availability heuristic. Over-optimism is likely to cause the promisor to think that performance is more, and his breach less, likely than is in fact the case.242 Further, the availability heuristic might cause a promisor to give greater weight to his present intention to perform, which is ‘vivid and concrete’ in comparison to the ‘abstract possibility’ that future eventualities will cause him to breach.243
4.46 In contrast Robert Hillman concludes that despite its ‘obvious contribution to legal analysis’ BDT simply ‘cannot resolve the mystery of liquidated (p. 127) damages’.244 He acknowledges that the effect of over-optimism may be a partial explanation for the judicial control of penalty clauses. However, he points out that judges themselves are affected by the same phenomena as contractors.245 Judges may be affected by the hindsight bias and so overestimate the parties’ ability, at the moment of contracting, to predict the actual damages that would result from the contemplated breach of contract.246 Further, Hillman points out that there is a ‘problem of conflicting lessons’ when BDT is applied to stipulated damages provisions. If it is conceded that reference to over-optimism and the availability heuristic support the courts’ zealous scrutiny of stipulated damages clauses other cognitive phenomena can be said to enforce the more general enforcement of these clauses. For example, if contractors regard the default award of unliquidated damages assessed by reference to the promisee’s lost expectation as part of the status quo, their decision to substitute a personalised measure of recovery should be acknowledged and respected as the clearest indication of their intention.247 Further, if cognitively challenged contractors dislike ambiguity they may prefer the security of liquidated, over the uncertainty of unliquidated, damages.248
4.47 Rachlinski disputes this conflict between behavioural phenomena and argues instead that ‘a reasonably clear prescription emerges’ that is consistent with the current law:249
Courts generally should be sceptical of liquidated damages clauses, but should enforce them in circumstances in which ambiguity is so large as to impede contract formation. In fact, the current judicial approach to enforcing liquidated damages clauses follows this formulation. Courts generally disfavour such clauses, but enforce them if the actual damages are difficult to calculate, and the liquidated damages clause reflects a reasonable effort to approximate actual damages.250
The policy prescriptions of the commentators canvassed above vary considerably between on the one hand those who assert that BDT supports a version of the present legal regime, which would enforce some, but strike down other, stipulated damages provisions and on the other those who believe BDT does not support the current position.251
(p. 128) 4.48 At this point some general points about BDT should be mentioned. First, even if it is conceded that BDT can explain some aspects of the current regulation of stipulated damages provisions it cannot explain why the heuristics, effects, and biases that justify the scrutiny of these terms should not also be applied mutatis mutandis to other contractual clauses. This is a powerful but neglected critique of many articles written from the perspective of BDT analysing private law remedial doctrines.252 Secondly, a methodological question arises as to what extent the experiments used to verify the phenomena described accurately replicate the experience of contractors making real life choices.253 Finally, it should be noted that BDT does not support a distinction in the legal regime that should be applied to more and less sophisticated contractors. The reason for this is because, ‘contrary to intuition empirical studies tell us that experience does not eliminate some biases particularly … overoptimism’.254
4.49 It has been argued that stipulated damages provisions may have a role to play in assisting parties to protect what has been termed their ‘secrecy interest’.255 This approach builds upon the intuition that parties to a contract may prefer not to disclose sensitive business information, which would otherwise be necessary to substantiate a claim for loss of profits, information such as ‘materials and labor costs, inventory size, availability of alternative suppliers, the identity of … customers … and in the case of newer businesses her business plan’.256 Revealing such information it is suggested may weaken the promisee’s bargaining position in relation to the negotiation of future contracts for the supply of the same, and ancillary, goods and services.257 Indeed, it is asserted that the suppression of such information may cause the promisee to prefer ‘even a highly undercompensatory provision’.258
4.50 There are a number of difficulties with this argument. First, a considerable amount of private information of the type the authors suggest the parties may wish to (p. 129) withhold will be revealed in the pre-contract process of negotiating a stipulated damages provision259 and post-contract, in negotiating a settlement based upon any enforceable liquidated damages provision.260 Secondly, the intuition upon which the argument builds, that parties will want to keep certain information secret, is at base an unevidenced hypothesis.
4.51 The general, compensatory aim of unliquidated damages has been stated many times and is said to be ‘a bedrock of our law’.261 In Farley v Skinner, Lord Clyde stated that: ‘damages should not be awarded, unless perhaps nominally, for the fact of a breach of contract distinct from the consequences of the breach’.262 More recently in A-G v Blake Lord Nicholls reiterated that: ‘with breaches of contract … [t]he general principle regarding assessment of damages is that they are compensatory for loss or injury’263 and Lord Hobhouse in the same case, dissenting, noted that: ‘The principle of compensation is both intellectually sound as the remedy for breach and provides the just answer’.264
4.52 The concept of just compensation in relation to liquidated, as opposed to unliquidated, damages is emphasised more in the US than in the UK. This distinction arises from the different tests for a penalty that prevail in the two jurisdictions. The provisions of the Uniform Commercial Code265 and the Restatement 2d266 allow the reasonableness of the clause to be established at either of two moments in time: when the contract was entered or when it was breached. Or, put another way, the test for a penalty takes account of either(p. 130) anticipated or actual loss.267 The alternative reference to actual harm in the US is an explicit incorporation of the just compensation principle because contractual clauses will not be enforced when they stipulate the recovery of a sum that is disproportionate to the actual damage inflicted upon the promisee.
4.53 In contrast, in the UK the appropriate test is to be applied only at the time of contracting where the ‘new’ test formulated by the Supreme Court in Cavendish268 governs and also in any cases269 where the ‘old’ Dunlop case continues to apply.270 Nonetheless, Andrew Burrows argues the provision of just compensation underlies the prevailing refusal to enforce so-called penalty clauses: ‘Penalties run counter to the central principle that contractual damages are for compensating the claimant and not punishing the defendant’.271 He later states that: ‘the strongest argument for knocking down penalty clauses is that this would contradict the traditional approach to damages according to which punitive damages are not available for breach of contract’.272 To similar effect, Popplewell J has said: ‘The policy of the rule is aimed at a party stipulating to be paid more than his actual loss in circumstances in which he can fairly be compensated for such loss by being left to his remedy in damages’.273 It is suggested that the provision of compensation does not yield a convincing justification for the current law on stipulated damage provisions for the following reasons.
4.54 First, recent developments in the law, chiefly that effected in the Blake case,274 qualify the generality of the statements of the compensation principle above.275 In Blake, the House of Lords acknowledged that there was ‘a light sprinkling of cases’ where courts had, under different rubrics, granted financial remedies that went beyond compensating the claimant for his loss and required the defendant to account for the profit he derived from his breach of contract.276 In (p. 131) exceptional cases,277 there was therefore ‘no reason, in principle, to rule out an account of profits as a remedy for breach of contract’.278 Secondly, the emphasis in the Cavendish case upon a proportionality between the sum stipulated and the legitimate interest it is protecting is less directly routed in compensation that the previous test requiring a comparison between the sum stipulated and a genuine pre-estimate of likely loss.279 Thirdly, one commentator simply denies the existence of ‘any inevitable link’ between the purpose of damages assessed by the courts and stipulated damages clauses on the basis that if parties are to enjoy the right of contractual self-determination, this should extend to a stipulated damages provision.280 In some cases, this provision will overcompensate the promisee but, as Muir notes:281
it is obvious that the courts have no catholic objection to over-compensation, as there are many contractual rules282 which allow one party to secure a greater benefit than he has paid for.
4.55 Another category of recovery which provides for compensation in excess of loss is an award of exemplary or punitive damages. It has been established in Canada that such damages may be awarded for a breach of contract,283 even where there is no concurrent tortious duty. It does not, however, seem likely that the English courts284 will follow this lead, despite285 the acknowledgement by the Law Commission that such awards need to be put on a more principled basis.286 (p. 132) In contrast, there are also of course contractual clauses which seek to reduce the promisee’s recovery to less than actual loss (limitation clauses or ‘underliquidated’ damages clauses287) or to zero (so-called exemption clauses). It has been suggested that such clauses are treated with greater leniency than stipulated damage provisions.288 While this suggestion is difficult to evaluate,289 it is clear that there is no consistent approach to contractual provisions of different types which, respectively, over- or underestimate a promisee’s loss. Such inconsistency at least calls into question the law’s general commitment to just compensation. Finally, comments such as those of Burrows above present a simple binary opposition between unenforceable penalties which seek to punish the promisee and enforceable liquidated damages provisions, which provide compensation that may no longer be appropriate. In the Cavendish case, Lord Mance290 suggested that this was an oversimplification because recent authority ‘[s]uggests that this dichotomy may not be exclusive and that there may be clauses which operate on breach and which are commercially justifiable although they fall into neither category’.291 The recognition by the Supreme Court in the Cavendish case that a deterrent penalty can be enforceable is simply at odds with the assertion that the underlying rationale of the current penalty rule is the provision of just compensation.
4.56 The idea that the prohibition upon the enforcement of penalty clauses is motivated by a desire to avoid the punishment of the promisee was recently endorsed by Kiefel J in the High Court of Australia’s decision in Paciocco v Australia and New Zealand Banking Group Ltd:
. . . the basal purpose of the larger principle, or policy, of the law is not stated. That policy has not changed over time. It is that a sum may not be stipulated for on payment or default if it is stipulated as a threat over the person obliged to perform; it may not be stipulated where the purpose and effect of requiring payment is to punish the defaulting party.292
This idea of the prevention of punishment as the rationale for the penalty rule was articulated most clearly in a line of Scottish cases from the mid-nineteenth century the effect of which were summarised by Lord Hodge in the Cavendish case.293 In Craig v McBeath,294 Lord Justice Clerk Inglis cites Home v Hepburn295 to support the proposition that: ‘Parties cannot lawfully enter into an agreement that the one party shall be punished at the suit of the other’. A similar view was expressed by Lord Young in Robertson v Driver’s Trustees.296
4.57 The concept of punishment may have functioned differently in the past to the way it does now. In the past, the intention to punish may have been a ground sufficient alone to render a stipulated damages provision unenforceable. However, in later cases the intention to punish seems to be an inference which is made after the application of a different test of enforceability, to which it adds nothing. It is in this sense only that we should understand Lords Neuberger and Sumption’s comment in Cavendish that: ‘[t]he innocent party can have no proper interest in simply punishing the defaulter’.297 This is made clear by two factors. First, this sentence immediately follows the single sentence in which Lords Neuberger and Sumption state what they term the ‘true test’, which 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’.298 Secondly, Lords Neuberger and Sumption endorsed299 the insightful, but until recently300 neglected, comment of Lord Radcliffe in Campbell Discount Co v Bridge:
I do not myself think that it helps to identify a penalty, to describe it as in the nature of a threat ‘to be enforced in terrorem’ (to use Lord Halsbury’s phrase in Elphinstone v Monkland Iron & Coal Co Ltd (1886) 11 App Cas 332, 348). I do not find that (p. 134) the description adds anything of substance to the idea conveyed by the word ‘penalty’ itself …301
4.58 There is, however, a different constitutional aspect to some statements about the proper limits upon the ability of contractors to ‘punish’ with excessive stipulated damages provisions parties who default on their contractual obligations. This arises from the fact that ‘common law … courts have never been weak institutional actors’. Rather they have a ‘long tradition of fighting for jurisdiction’ which to this day they are ‘reluctant to surrender’.302 The jurisdiction here referred to is the function, more usually associated with criminal law, to discipline wrongdoers. Frankfurter J in the US case of Priebe & Sons Inc v United States303 said that the ‘basic reason’ for the doctrine was the idea ‘[t]hat the infliction of punishment through courts is a function of society and should not inure to the benefit of individuals’. Such appeals to the view that punishment is something that should be the responsibility of the state are both emotive and overgeneral. Application of the current law will often result in the enforcement of stipulated damages clauses that in fact overcompensate the promisee. Indeed, the simple acknowledgement of the parties’ inability to predict future losses combined with judging the validity of a stipulated damages clause (in the UK) at the moment of contracting means that this was always so. In such cases, to describe the award of liquidated damages in excess of the promisee’s provable loss as inappropriate punishment would, if punishment was the litmus test for a penalty, render ineffective many currently enforceable clauses. This is contrary to the prevailing tendency to enforce more clauses than previously. As long ago as 1935 the author of the first edition of the leading work on damages in the US noted that ‘today the pendulum is swinging in the other direction, and the tendency is rather towards resolving the doubt in favour of the enforceability of such clauses’304 and the decision of the UK Supreme Court in the Cavendish case may be regarded as a significant stop, if not the endpoint, of that journey.
4.59 The argument that contractors should not be empowered to oust the jurisdiction of the court is sometimes made in a weaker form, absent any mention of punishment. In this form the objection is perhaps most often encountered when a contractual clause providing for specific performance as a remedy is challenged. In this context, it may be argued that: ‘[b]y incorporating into their agreement a clause that provides for specific performance, contracting parties may side-step (p. 135) the strong inclination of English law towards compensatory damages’.305 This argument cannot of course be made with the same force in relation to the enforcement of stipulated damages clauses because they involve only the substitution of one financial award for another, liquidated for unliquidated damages. Further, in other ways the law of contract shows itself prepared to enforce contractual provisions which affect the remedy awarding capacity which the court would otherwise enjoy. Exemption clauses may prevent the promisee from making any recovery as may force majeure clauses. Further termination clauses may provide a contractor with the right to resile from a contract when the promisor has done something that would otherwise not amount to a repudiatory breach.
4.60 In the Cavendish case, Lords Neuberger and Sumption described the penalty rule as: ‘[a]n ancient, haphazardly constructed edifice which has not weathered well’.306 Notwithstanding this and having noted the frustration of many able judges in the past who have tried to uncover the origins of the rule Lords Neuberger and Sumption analyse the equitable origins of the modern rule from the beginning of the sixteenth century. It would seem that the Supreme Court was in agreement with the High Court of Australia, which stated in Andrews v Australia and New Zealand Banking Group Ltd that307 ‘[a]n understanding of the penalty doctrine requires more than a brief backward glance’. The agreement between the highest appellate courts of the United Kingdom and Australia unfortunately ended there. The complete history and antecedents of the jurisdiction to control stipulated damages provisions was explored in Chapter 1. At this point only the more abbreviated historical analyses proffered in the Cavendish and Andrews cases will be examined. Both the jurisdiction to control penalty clauses and forfeiture have a common origin in the relief offered by courts of equity.308 However, the subsequent development of this jurisdiction has been complex.309 The necessarily (p. 136) abbreviated account offered in the Cavendish case describes how the penalty jurisdiction grew from the practice in equity of offering relief from penal bonds. A penal bond consisted of a promise to pay a stated sum, subject to a condition that if the main obligation is fulfilled by a particular time, the promise to pay the stated sum would be void. The essential preconditions for the exercise of this relief were: that the penal provision was in reality offered as security for the recovery of a debt or damages and that this objective could be achieved by restraining proceedings on the bond in the common law courts subject to the defendant paying damages.310
4.61 The common law followed this lead in statutes of 1696 and 1705.311 The penal bond then metamorphosed into a closer approximation to the modern form when it appeared as an agreement to pay a sum on breach of contract. This turned around the old penal bond; now the ‘penalty’ was stated as the subsidiary obligation. By 1801, it was established that the claimant had a technical election: he could bring an action in debt to recover the penalty, for which recovery would be limited by the statute to the damage sustained, or he could sue in assumpsit for his actual losses.312 Therefore, in either case the result was the same: recovery for actual loss only.313 Today, the action to enforce the penalty has been largely forgotten.314
4.62 In Andrews, the High Court of Australia built upon a subtly different and highly abbreviated315 historical account, which asserts the existence of a distinct and subsisting equitable jurisdiction to relieve against penalties which is not co-extensive with the common law jurisdiction but wider than it and so extends to relief from obligations to pay money contingent upon events other than a breach of contract. In Cavendish, this conclusion was disapproved of as a matter of principle and of authority. The objection of principle is centred upon the difficulty of defining the limits of such a power of review and the consequent uncertainty.316 As to authority, Lords Neuberger and Sumption point out the asserted broad equitable jurisdiction to relieve ‘appears to have left no trace in the authorities since the fusion of law and equity in 1873’.317 The only recent (p. 137) authority318 would appear to be a case in the Court of Appeal319 that may have been wrongly decided.320 This conclusion is supported by a thorough review of the Andrews decision by a group of the leading Australian contract scholars who conclude that:
[t]he High Court does not discuss any case which clearly supports its perspective on the history of the subject. Imperfectly reported cases, decided in contexts vastly different from those which arise under the modern law, to say nothing of fragments from a handbook of equity written in the 17th century have no bearing on the law of today.321
From the above discussion of the Cavendish and Andrews cases it can be seen that at a single point in time retrospective reflection upon the purpose and ambit of the penalty rule can yield distinct answers.322 It should not therefore be a surprise that at varying points of time the purpose of stipulated damages and its precursors have been perceived differently. Roman law used the stipulation poenae at first to ensure the performance of obligations and later to assess damages323 in order to effect a saving in court time assessing damages.324 Subsequently penal clauses were used as a means of outflanking prohibitions on usury.325 The relevance of such distant considerations in the modern environment was doubted by Keane J in Paciocco v Australia and New Zealand Banking Group Ltd:
Medieval religious scruples against usury associated with a primitive agrarian economy do not provide a satisfactory basis on which the penalty rule might now be sustained.326
[t]he historical perspective suggests that the [penalty] rules … reflect an assortment of notions; a jumble of historical curiosities which out of context provide no unitary rationale for invalidating stipulations for the payment of an agreed sum.327
4.63 Richard Posner has written that:328
Global consensus (to exaggerate a bit) is further evidence – of course not conclusive – for the optimality of our existing law . . .
This comment could function well as a summary of the comparative law approach of the Supreme Court in the Cavendish case, where counsel for Cavendish were asked by the court to prepare an appendix to its case which included ‘a valuable examination of the law of, and relevant academic commentary from, other common law countries’.329 Although only directly referred to by Lord Mance the appendix is assumed to be the basis for the comparative reflections in the judgment of Lords Neuberger and Sumption concluding that: ‘[t]he penalty rule … is common to almost all the major systems of law, at any rate in the western world’.330 This over-general conclusion is supported by a single paragraph of discussion which notes ‘a corresponding rule … found in the Civil Codes’ and also ‘in influential attempts to codify the law of contracts internationally’.331
Despite the highly abbreviated (and as we will see in a minute misleading)332 nature of this survey of world legal systems it nonetheless plays a key role in the approach of the majority333 judgment in the Cavendish case where it is used to (p. 139) rebut the ‘primary case’334 of counsel for Cavendish that the penalty rule should be abolished.335
4.64 An impression is conveyed of a general acceptance of a penalty rule across all major western legal systems with Belgium, France, Germany, Italy, Scotland, Switzerland and the US mentioned.336 However, these legal systems are all of course very different, the group containing representatives of the common law, civil law, and mixed ‘legal families’. It has been said that ‘Nowhere in the law of contracts has the clash of common law and civil law seemed as irreconcilable as in the treatment of penalty clauses’337 yet these differences are ignored or at least downplayed in this aggregation. The different juridical starting point and distinct historical contexts of the civil and common law systems are not mentioned. The starting point for the civilian commitment to the literal enforcement of penalties derives from the respect for the intention or will of the parties.338 It was a rule of classical Roman law that an aggrieved party was entitled to recover an agreed sum without restriction.339 This was received into the Napoleonic Code in Article 1152 which formed the basis for the laws of neighbouring countries: Belgium, Italy, Portugal, and Spain. Thereafter there was a progressive retreat from this extreme position340 with most European countries giving to a judge the power to moderate an agreed sum341 which was adjudged as grossly excessive. There is, however, some irregularity in discrete legal systems.342 Particularly, under German (p. 140) law, the power to reduce the agreed sum is exercisable only where the clause in question is intended to pressurise a party to perform a legal obligation (rather than to fix in advance the amount of damages)343 and where that party is not a merchant (BGB 343 and 348).344
4.65 Lords Neuberger and Sumption refer expressly to two international codifications of contract law,345 that include prohibitions of penalty clauses. However, they did not refer to the most successful international convention for the unification of commercial law, the United Nations Convention on Contracts for the International Sale of Goods (1980) (known as the Vienna Convention or CISG),346 which contains no specific347 provisions dealing with stipulated damages clauses. Indeed, the reason for this omission is instructive in this context. Allan Farnsworth,348 who played a significant role in the development of the CISG, has said that it was because of the ‘wide gulf’ between common and civil law systems that ‘the Vienna Convention contains no provision on the important subject of stipulated damages’.349
4.66 The survey of different common law systems which was also relied upon in Cavendish is similarly truncated. Other than the account of Australian law where the High Court’s ‘removal’ in Andrews350 of the ‘breach’ requirement was noted and criticised, an unwarranted impression of uniformity throughout the common law jurisdictions is conveyed. In Canada it has been accepted judicially that the (p. 141) proper basis of the rule against penalties is unconscionability351 and further that the jurisdiction should more explicitly reflect that ‘[t]he power to strike down a penalty clause … is designed for the sole purpose of providing relief against oppression’.352 The law in the US is different to that in the UK in several ways.353 The UK approach places emphasis upon ex ante ‘reasonableness’; in contrast the US approach requires either ex ante or ex post reasonableness. There is therefore a considerable difference between the two major common law jurisdictions. Indeed, one common law jurisdiction even adopts an approach to agreed damages clauses that is closer to that of the unmodified Code Napoleon. India has a partly codified law of contract and the India Contract Act 1872 rejects the distinction between valid liquidated damages clauses and invalid penalties and section 74 states that the victim of the breach is entitled to ‘reasonable compensation not exceeding the amount so named or … the penalty stipulated for’.354
4.68 In this long chapter we have critically examined the different, but often inter-related, rationales that have been proposed to justify the current position, or a different approach to the enforcement of stipulated damage clauses. In summary, all have been found wanting in some way and no single policy rationale provides either an explanation for the law we have or an incontrovertible case for its abolition or modification. The explanations most frequently articulated by the courts themselves, transaction cost savings, ignore the cost of the indeterminacy inherent in the current law and the burden of the consequential self-protection which the (p. 142) promisor must undertake.355 The major contribution of academics to the debate around the rationale of the penalty rule has been by advocates of the economic analysis of law356 and behavioural decision theory.357 However, the accounts in the tradition of economic analysis and behavioural decision theory are often contradictory358 and in the case of the former also challenging to read for non-specialists.359 As a result, they have received scant attention from the courts.360 The different approaches and constraints upon on the one hand an academic and on the other a judge were well illustrated by the US decision in Lake River Corp v Carborundum Co.361 Judge Posner, who decided the case, is also the author of one of the seminal books in this field, Economic Analysis of Law, in 1973. His judgment in the Lake River case seemed to have two parts: it rehearsed the usual economic arguments for the enforcement of clauses such as that under consideration and then applied the law of Illinois to deny enforcement. Other explanations that emphasise the protection of the secrecy interest and the signalling functions (at least in relation to new entrants to a market) are narrow in their focus.
4.69 Certainty has long been recognised as a key value in the law of contract.362 In 1774, Lord Mansfield said: ‘In all mercantile transactions the great object should be certainty: and therefore, it is of more consequence that a rule be certain, than whether the rule is established one way or the other’. In the present day, the demand for certainty is, if anything, more pressing. In a public lecture delivered in 2016, Lord Thomas, the most senior judge in the UK, stated that: ‘Clarity and predictability in the law, as well as its ability to develop in a principled manner, is the bedrock upon which businesses, just as much as individuals order their affairs and enter into binding agreements’.363
4.70 It was noted earlier364 that, despite some lingering vocabulary and consistency with earlier decisions of inferior courts, in the Cavendish case the Supreme Court of the UK emphatically rejected the Supreme Court of Canada’s365 endorsement of unconscionability as the underlying rationale for the penalty jurisdiction. (p. 143) A further powerful, although indirect, endorsement of the value of certainty in Cavendish can be seen in the Supreme Court’s rejection366 of the approach taken by the High Court of Australia in the Andrews case,367 which extended the jurisdiction to set aside contractual clauses as penalties to provisions contingent upon events other than the contemplated payer’s breach of contract.
4.71 It is perhaps unsurprising that in the past, when notions of consumer protection were less embedded in the law, there was a reluctance to treat private contractors more ‘sympathetically’. Writing in 1915, a commentator noted that: ‘[i]f the law bore heavily upon the individual, at least there was a known law, the certainty of which had become the “safety of all” ’.368 It is perhaps more surprising that all proposals to cleave apart the common law doctrine of penalties into two parts, respectively applicable to consumer contracts (so-called B2C contracts) and those where both contractors are businesses (so-called B2B contracts) have failed. The conjoined appeals in the Cavendish case set up the possibility of such a division as the two appeals represented very different contexts: the Makdessi appeal a substantial commercial contract and the ParkingEye appeal a low value consumer contract. In the former the upholding of the clause would cause the party subject to it to ‘lose’ US$44 million, in the latter it would require the payment of an £85 parking ‘fee’. However, the Supreme Court declined any implied invitation to distinguish the two contexts.369 Lord Hodge explained this rejection:
Creating such a gateway to the application of the rule would risk adding to the expense of commercial disputes by requiring the court to rule on issues of fact about the bargaining power of the parties and the calibre of their respective legal advisors.370
4.72 Consumers of course now enjoy considerable statutory protection in relation to any contracts they enter. With respect to stipulated damages provisions they enjoy a parallel protection to the common law rule under the Consumer Rights Act (p. 144) 2015 Part 2.371 If the Supreme Court had decided to abolish the common law rule against penalties the statutory protections afforded consumers would of course have remained. This fact adds force to the argument for the abolition of the penalty rule on the basis that the statutory regime adequately protects the interests of the most vulnerable class. In the Andrews case, the opposite argument was put that the protection afforded to consumers in Australia justified the expansion of the common law rule to cover non-breach events,372 although this approach has been disputed373 and not acted upon in other decisions by the same court.374
4.73 In the Cavendish case, as in the Australian Andrews case, the court relied upon a historical perspective375 to justify its very different decisions upon where the boundaries of the jurisdiction to review stipulated damages clauses should lie. That such radically different conclusions can be reached from the same evidence refutes history’s claim to provide a complete rationale for the doctrine. At the same time, a doctrine of law that predates the emergence of the modern law of contract itself will never be lightly overturned. It is perhaps this deference to longevity that results in a case where the leading judgment of Lords Neuberger and Sumption admits that ‘[w]e rather doubt that the courts would have invented the rule today if their predecessors had not done so three centuries ago’ but nonetheless declines the invitation to abolish the rule.376
4.74 The comparative perspective377 similarly is unable to provide a complete explanation for the existence of the penalty rule. It was shown that the comparative analysis relied upon in Cavendish was incomplete and used only to rebut the argument that the penalty rule should be abolished. A more nuanced account of other jurisdictions would have described a spectrum of different approaches to stipulated damage clauses with contrasts between, and within, the civil and common law traditions.
(p. 145) 4.75 It seems that no single rationale provides a complete account of either the law we have378 or a completely convincing argument for any other approach. Rather it is clear that the courts take a ‘multifactorial’ approach relying upon a number of articulated but, considered singly, objectively unconvincing rationales (eg transaction cost savings). These factors include: an overstated reverence for history, an abbreviated account of the law in other jurisdictions, suffused with a consistent emphasis upon the need for certainty in commercial dealings which nonetheless falls short of a formal commitment to enforce all stipulated damages provisions in commercial contracts.(p. 146)
1 See B Dickson, ‘The Contribution of Lord Diplock to the General Law of Contract’ (1989) 9 Oxford Journal of Legal Studies 441 at 443: ‘By most accounts Kenneth Diplock was not a modest man. He did not suffer fools gladly. He was regarded by his contemporaries as someone who always spoke his mind, at times employing irony in his put-downs’ and R Stevens, Law and Politics: The House of Lords as a Judicial Body 1800–1976 (Weidenfeld and Nicolson 1979), 362 who considered him as analytically outstanding, iconoclastic, and outspoken. In Final Appeal (Clarendon Press 1972) Louis Blom Cooper and Gavin Drewry relate how in an unreported case Diplock LJ dissented from the judgments delivered by Lord Denning MR and Harman LJ, who allowed the appeal by saying simply: ‘For the reasons given by my brother Harman I would dismiss the appeal’ quoted in Dickson above at n 10. Lord Diplock was aware of his intellectual reputation and in one of his most famous cases, United Dominions Trust (Commercial) Ltd v Eagle Aircraft Services Ltd  1 WLR 74 at 82G–H, mocked himself for his own ‘gratuitous philological exhibitionism’.
5 Giesecke v Cullerton 280 Ill 510, 513, 117 NE 777, 778 (1917), referred to by J B Coopersmith, ‘Refocusing Liquidated Damages Law for Real Estate Contracts: Returning to the Historical Roots of the Penalty Doctrine’ (1990) 39 Emory Law Journal 267.
7 C J Goetz and R E Scott, ‘Liquidated Damages, Penalties and the Just Compensation Principle: Some Notes on an Enforcement Model and a Theory of Efficient Breach’ (1977) 77 Columbia Law Review 554 at n 12.
12 R A Posner, ‘Some Uses and Abuses of Economics in Law’ (1979) 46 U Chi LR 281 at 290. See the similar language of Eric Posner, ‘Economic Analysis of Contract Law after Three Decades: Success or Failure?’ (2003) 112 Yale Law Journal 829 at 860.
13 Cavendish (n 3) , or restrict it to the extent that it should not apply to commercial transactions between legally advised parties with equal bargaining power: see ibid .
15 Such a discussion typically precedes, often as part of the introduction, many discussions of the penalty jurisdiction. For two recent critiques of the High Court of Australia’s decision in Andrews v Australia and New Zealand Banking Group Ltd  HCA 30 and the UK Supreme Court’s decision in Cavendish that follow this pattern see, respectively, A Gray, ‘Contractual Penalties in Australian Law after Andrews: An Opportunity Missed’ (2013) 18 Deakin Law Rev 1 and Lord Hope, ‘The Law on Penalties: A Wasted Opportunity’ (2016) 33 Journal of Contract Law 1.
16 In the Patrick Atiyah’s magisterial work on the intellectual history of the law of contract, P S Atiyah, The Rise and Fall of the Freedom of Contract (Oxford University Press 1985) the first indexed reference to the eponymous concept appears at p 270.
18 Printing and Numerical Registering Co v Sampson (1875) LR 19 Eq 462 at 465. For an academic writing in a similar vein see S Rowan, Remedies for Breach of Contract: A Comparative Analysis of the Protection of Performance (Oxford University Press 2012) 220: ‘The most persuasive argument in support of penalty clauses … is the paramount importance of freedom of contract’.
25 Cavendish (n 3) , quoted by Leggatt J in MSC Mediterranean v Cottonex Anstalt  EWHC 283 (Comm) affirmed  EWCA 789.
26 Cavendish (n 3) at  and  (Lords Neuberger and Sumption).
30 Sometimes referred to by other commentators as the freedom to contract: see R E Barnett, ‘Conflicting Visions: A Critique of Ian Macneil’s Relational Theory of Contract’ (1992) 78 Virginia Law Review 1175, 1181.
34 ibid 500.
36 ibid 499–500.
37 R E Scott and G G Triantis, ‘Embedded Options and the Case against Compensation in Contract Law’ (2004) 104 Columbia Law Review 1428. For a more general overview of their approach see R E Scott and G G Triantis, ‘Incomplete Contracts and the Theory of Contract Design’ (2005) 56 Case Western L Rev 187.
39 Andrews v Australia and New Zealand Banking Group Ltd (n 15)  quoted by Keane J in Paciocco v Australia and New Zealand Banking Group Ltd  HCA 28 at .
45 See also the discussion of ‘Pareto-optimality’ in n 188 below.
46 See generally A Ham, ‘The Rule against Penalties in Contract: An Economic Perspective’ (1990) 17 Melbourne University Law Review 649, 654. Such ‘distributive’ questions are sometimes contrastingly described as raising ‘political’ questions upon which standard economic analysis is silent.
47 For the atypical economic argument that sometimes private law should be used to promote distributional objectives see A T Kronman, ‘Contract Law and Distributive Justice’  89 Yale LJ 472 and S A Smith, ‘In Defence of Substantive Fairness’ (1996) 112 LQR 138 and S A Smith, Contract Theory (Oxford University Press 2004) 352–7.
51 Other common law jurisdictions have more readily acknowledged a general concept of unconscionability. See ibid 341.
56 Rowan, Remedies for Breach of Contract (n 18) 224, referring to the Unfair Contract Terms Act 1977 but for so-called B2C (business to consumer) contracts see now the equivalent provisions in the Consumer Right Act 2015 and C Bildfell, ‘Exculpatory Clauses and Liquidated Damages Clauses: Two Sides of the Same Coin?’ (2015) 78 Saskatchewan Law Review 347 at 358 examining Canadian law.
59 See Bildfell, ‘Exculpatory Clauses and Liquidated Damages Clauses’ (n 56) 358, referring to the following statement in Ontario Ltd v Huang (2013) 27 CLR (4th) 292 at para 13: ‘The law on enforceability of so-called penalty clauses over the past 30 years has moved in the direction of upholding freedom of contract rather than setting aside too readily clauses negotiated by the parties having penal consequences as unenforceable, absent unconscionability’.
60 Eg the question of how frequently contractors include stipulated damages provisions in their contract? Different authors have used different adjectives to describe the frequency with which such clauses are used absent any reference to empirical data: the Law Commission of England and Wales began a working paper with the observation that ‘The use of penalty clauses is extensive’ but the Scottish Law Commission in a report suggest more modestly that contracts only ‘often’ provide for payment of a stated sum in the event of breach. See, respectively, Law Commission, Penalty Clauses and Forfeiture of Monies Paid (Working Paper No 61 HMSO 1975) para 10 and Law Commission, Report on Penalty Clauses (Working Paper No 171 HMSO 1999) at para 1.2. Individual academic’s assessments also vary: one UK author’s first sentence states that their use ‘is a widespread practice’ whereas an American writer states that contracting parties only ‘occasionally’ stipulate the dollar amount that the breaching party must pay the non-breacher. See respectively: the first sentence in L Gullifer, ‘Agreed Remedies’ in A Burrows and E Peel (eds), Commercial Remedies: Current Issues and Problems (Oxford University Press 2003) ch 16 and T Muris, ‘Opportunistic Behavior and the Law of Contracts’ (1981) 65 Minnesota Law Review 521, 581. Other writers adopt an intermediate position, eg DiMatteo, who states that: ‘[f]rom a practical perspective, such liquidated damages clauses are commonly found in many types of contract’. See L A DiMatteo, ‘A Theory of Efficient Penalty: Eliminating the Law of Liquidated Damages’ (2001) 38 American Business Law Journal 633.
61 The closest attempt to date which examines the extent to which stipulated damages provisions are used might be the Report of the Secretary-General: Liquidated Damages and Penalty Clauses in the Yearbook of the United Nations Commission on International Trade Law 1979 Vol X section VI, where: ‘[i]n order to determine the nature and extent of the use of liquidated damages and penalty clauses in international trade contracts’, a representative selection of general conditions was analysed. Of the 167 general conditions and contracts analysed 79 contained liquidated damages clauses and 88 did not. The biggest category of contracts were international sales with delays in delivery of goods and corresponding payment of the price the main obligations to which the provisions attach. Of course, the focus of this ‘survey’ is international not domestic sales contracts.
62 McAlpine v Tilebox (n 24) .
63 Commissioner for Public Works v Hills  AC 368; Bridge v Campbell Discount Co Ltd  AC 600; Workers Trust & Merchant Bank Ltd v Dojap Investments Ltd  AC 573; and Ariston SRL v Charly Records Ltd (1990) Financial Times (21 March 1990).
64 See the third, and possibly the first, of the cases in the previous footnote. To his list Colman J could have added the (then) recent case of Jeancharm Ltd (t/a Beaver International) v Barnet Football Club Ltd  All ER (D) 69 (Jan), where the Court of Appeal held that a clause in a contract for the supply of football kits which sought to apply 5% per week interest (ie 260% pa) to unpaid sums was penal.
65 Enforceable—Euro London Appointments Ltd v Claessens International Ltd (n 23); M & J Polymers Ltd v Imerys Minerals Ltd  EWHC 344 (Comm),  1 Lloyd’s Rep 541; and General Trading Co (Holdings) Ltd v Richmond Corp Ltd  EWHC 1479 (Comm),  2 Lloyd’s Rep 475; unenforceable—Lansat Shipping Co Ltd v Glencore Grain BV  EWHC 551 (Comm),  1 CLC 379.
66 See ch 2(A).
67 A similar trend was evident in Australia where, in AMEV-UDC Finance Ltd v Austin (1986) 162 CLR 170 at 190, Mason and Wilson JJ urged that contractors should enjoy ‘greater latitude in determining what their rights and liabilities will be’. This invitation was later taken up in Ringrow Pty Ltd v BP Australia Pty Ltd  HCA 71 at , where it was accepted that contract law ‘normally upholds the freedom of the parties, with no relevant disability, to agree upon the terms of their future relationship’. This trend came to an abrupt halt with the High Court’s decision in Andrews v Australia and New Zealand Banking Group Ltd (n 15), which justified the extension of the penalty jurisdiction to non-breach events by reference to ‘the need for caution in dealing with the unwritten law as if laissez faire notions of an untrammelled “freedom of contract” provided a universal value’. It is suggested that the powerful and full (the article is longer than the court judgment!) critique of the Andrews decision in J W Carter, W Courtney, E Peden, A Stewart, and G J Tolhurst, ‘Contractual Penalties: Resurrecting the Equitable Jurisdiction’ (2013) 30 Journal of Contract Law 99 makes a convincing argument for following the previous trend.
68 Robophone Facilities Ltd v Blank (n 57) 1447.
71 Cavendish (n 3) .
73 This was the view of Panormitanus Com. In decret, ad. C. 4 (dilecti), X De arbitus (I 43) No 14 1586 edn. Cf the opinion of Hostiensis as discussed by P Benjamin, ‘Penalties, Liquidated Damages and Penal Clauses in Commercial Contracts: A Comparative Study of English and Continental Law’ (1960) 9 International and Comparative Law Quarterly 600, 610.
74 Benjamin, ‘Penalties, Liquidated Damages and Penal Clauses in Commercial Contracts’ (n 73) 627. See also P Hachem, Agreed Sums Payable upon Breach of an Obligation (Eleven International Publishing 2011) 49.
80 For others see Ham, ‘The Rule against Penalties in Contract’ (n 46): ‘In the course of drawing up an agreement, the parties may determine in advance the damages that are payable should one party breach the contract. Thus, the expense and uncertainty of litigation are avoided and the parties can be sure that their interests are fully protected’; J P George, ‘Reimposable Discounts and Medieval Contract Penalties’ (2007) 20 Loyola Consumer Law Review 50 at 52: ‘[t]hey [ie stipulated damages clauses] tend to reduce trial costs by sidestepping proof of quantified damages’, and at 78: ‘Contracts with stipulated damage clauses perform valid functions … [t]he result is a greater likelihood of quick settlement, or if a trial is necessary, greatly facilitated proof of damages’; M Pressman, ‘The Two-Contract Approach to Liquidated Damages: A New Framework for Exploring the Penalty Clause Debate’ (2013) 7 Virginia Law & Business Review 651 at 656 : ‘Another common justification for the judicial enforcement of these clauses is their perceived efficiency. If there is no need to try the issue of damages, both the judicial system and the parties will avoid the extra time and expense associated with more litigation’ (footnotes omitted); and Waddams, The Law of Damages (n 38) para 8-330: ‘it [ie the enforcement of a stipulated damages clause] reduces the cost to the parties and to the state of settling a dispute after breach’.
83 A broader issue that is relevant to the economic discussion here is the fact that to a larger extent than in the past court proceedings are paid for by the state rather than from litigation fees. It can be argued that because the full costs of litigation are not borne by the parties (in economic jargon the cost is ‘externalised’) parties will include penalty clauses more frequently than is optimal. However, as De Geest and Wuyts point out, even if this is the case then ‘the excessive subsidisation of the courts can be resolved more easily by increasing trial costs’ (than by changing the law of stipulated damages). See G De Geest and F Wuyts, ‘Penalty Clauses and Liquidated Damages’ in B Bouckaert and G De Geest (eds), Encyclopaedia of Law and Economics: The Regulation of Contracts, Vol 111 (Edward Elgar Publishing 2000).
84 L A Stole, ‘The Economics of Liquidated Damage Clauses in Contractual Environments with Private Information’ (1992) 8 Journal of Law, Economics and Organisation 582, 582. See also DiMatteo, ‘A Theory of Efficient Penalty’ (n 60) 634: ‘A number of reasons exist for the popularity of these clauses. First, because of the fears and costs associated with litigation, parties who draft such clauses do so in the hopes that they will preclude the need for litigation’.
85 See I Macneil, ‘Efficient Breach of Contract: Circles in the Sky’ (1982) 68 Virginia L Rev 947 at 968–9, who uses a pithy phrase to make the point that a contractual provision that avoids a breach of contract is less costly than post breach litigation when he says that ‘ “talking after a breach” may be one of the more expensive forms of conversation to be found, involving, as it so often does engaging high-price lawyers’. In a later article he warns against ‘casual empiricism’ referring to the retrospective assertion of unevidenced transaction costs to justify a particular policy prescription see I Macneil, ‘Contract Remedies: A Need for a Better Efficiency Analysis’ (1988) 144 Journal of Institutional and Theoretical Economics 6 at 21.
86 J H Barton, ‘The Economic Basis of Damages for Breach of Contract’ (1972) 1 Journal of Legal Studies 277; Goetz and Scott, ‘Liquidated Damages, Penalties and the Just Compensation Principle’ (n 7); P H Rubin, ‘Unenforceable Contracts: Penalty Clauses and Specific Performance’ (1981) 10 Journal of Legal Studies 237; R A Epstein, ‘Beyond Foreseeability: Consequential Damages in the Law of Contract’ (1989) 18 Journal of Legal Studies 105; Muris ‘Opportunistic Behaviour and the Law of Contracts’ (n 60) 580–8; S A Rea, ‘Efficiency Implications of Penalties and Liquidated Damages’ (1984) 13 Journal of Legal Studies 147; B E Hermalin and M Katz, ‘Judicial Modification of Contracts between Sophisticated Parties: A More Complete View of Incomplete Contracts and their Breach’ (1993) 9 Journal of Law, Economics & Organization 230; A S Edlin and S Reichelstein, ‘Holdups, Standard Breach Remedies, and Optimal Investment’ (1996) 86 America Economic Review 478; Edlin and Schwartz, ‘Optimal Penalties in Contracts’ (n 6); Scott and Triantis, ‘Embedded Options and the Case against Compensation in Contract Law’ (n 37); W Wang, ‘Comment: An Austrian Perspective on the Law of Liquidated Damages: The Case for an All-Enforcement Rule’ (2012) 84 Temple Law Review 795; and S Grant, J Kline, and J Quiggin, ‘A Matter of Interpretation: Ambiguous Contracts and Liquidated Damages’ (2014) 85 Games and Economic Behaviour 180. For rare defences of the penalty rule see K W Clarkson, R L Miller, and T J Muris, ‘Liquidated Damages v Penalties: Sense or Nonsense’ (1978) Wisconsin Law Review 351, who support the prevailing distinction between unenforceable penalty clauses and enforceable liquidated damage provisions; and P H Rubin, ‘Unenforceable Contracts: Penalty Clauses and Specific Performance’ (1981) 10 Journal of Legal Studies 237.
89 A rare exception to this statement is India, which has a partly codified law of contract. The Indian Contract Act 1872 rejects the distinction between valid liquidated damages clauses and invalid penalties and s 74 states that the victim of the breach is entitled to ‘reasonable compensation not exceeding the amount so named or … the penalty stipulated for’. The Cypress Law of Contract (1930) is modelled on the Indian Act.
90 The contrary suggestion of Muris in ‘Opportunistic Behaviour and the Law of Contracts’ (n 60) 585 is not consistent with the extent of litigation described in the remainder of this paragraph.
91 Benjamin, ‘Penalties, Liquidated Damages and Penal Clauses in Commercial Contracts’ (n 73) 625 suggests that the effect of uncertainty is more injurious in international, as opposed to domestic, contracts.
93  UKSC 6,  1 AC 696. For a further jurisdictional contrast see Smiley v Citibank (South Dakota) NA 517 US 735 (1996), where the US Supreme Court held that late payment fees levied by a bank on its credit card accounts were not unenforceable penalties because they could be characterised as ‘interest’, which the bank was entitled to charge under an old statutory provision.
94 Smiley v Citibank (South Dakota) NA (n 93).
95 Paciocco v Australia and New Zealand Banking Group Ltd (n 39).
96 See De Geest and Wuyts, ‘Penalty Clauses and Liquidated Damages’ (n 83) 146.
97 Steven Walt acknowledges this fact and outlines the unrealistic assumptions that would have to be satisfied before this cost could be avoided. See S Walt, ‘Penalty Clauses and Liquidated Damages’ in G De Geest (ed), Contract Law and Economics (Edward Elgar Publishing 2011) 178, 181.
99 Demurrage clauses were recognised authoritatively to be a form of liquidated damages in President of India v Lips Maritime Corpn  AC 395. See generally the discussion in ch 6C.
103 For a general discussion of ‘unrecoverable loss’ see D Harris, D Campbell, and R Halson, Remedies in Contract and Tort (2nd edn, Cambridge University Press 2002) ch 9 and ch 12, respectively, 140–1 and 167–75.
104 Perhaps most famously summarised by O W Holmes, ‘The Path of the Law’ (1897) Harvard Law Review 457, 462: ‘The duty to keep a contract at common law means a prediction that you must pay damages if you do not keep it—and nothing else’.
106 See also Viscount Haldane LC in British Westinghouse Electric and Manufacturing Co Ltd v Underground Electric Rlys Co of London Ltd  AC 673 at 689: ‘the fundamental basis [of an award of contractual damages] is thus compensation for pecuniary loss naturally flowing from the breach’; Lord Wilberforce in Johnson v Agnew  AC 367 at 400: ‘the general principle for the assessment of damages is compensatory’; and Lord Nicholls and Lord Hobhouse (dissenting) in A-G v Blake  1 AC 268,  4 All ER 385,  3 WLR 625, who at 632 and 652 respectively said: ‘with breaches of contract … [t]he general principle regarding assessment of damages is that they are compensatory for loss or injury’ and ‘The principle of compensation is both intellectually sound as the remedy for breach and provides the just answer’.
108 Farley v Skinner (n 105) .
109 An alternative starting point has been posited obiter dicta by Lord Hoffmann in Transfield Shipping Inc of Panama v Mercator Shipping Inc of Monrovia (The Achilleas)  UKHL 48 at ,  1 AC 61,  4 All ER 159: ‘[o]ne must first decide whether the loss for which compensation is sought is of a “kind” or “type” for which the contract breaker ought fairly to be taken to have accepted responsibility’. This is an extension of the principle established in South Australia Asset Management Corpn v York Montague Ltd  AC 191 (HL), also known as Banque Bruxelles Lambert v Eagle Star Insurance Co (the liability of a negligent surveyor towards a mortgagee was limited by the scope of the duty of care assumed by the surveyor to the extent of the surveyor’s initial overvaluation).
114 Gregg v Scott (n 111) .
116 The issue of burden of proof in relation to loss of profits on future contracts was given extensive consideration by the High Court of Australia in Commonwealth of Australia v Amann Aviation Pty Ltd (1992) 66 ALJR 123; G Treitel (1992) 108 Law Quarterly Review 226.
119 Although Addis has not been overruled, it has been interpreted narrowly in subsequent cases. See eg Malik v Bank of Credit and Commerce International SA (in liquidation)  3 All ER 1 (HL) at 9 (Lord Nicholls) and at 19–20 (Lord Steyn), Lords Goff, Mackay, and Mustill agreeing.
120 These are: contracts where a major, but not necessarily the only, purpose is the provision of enjoyment: Jarvis v Swans Tours  QB 233 (CA), interpreted in the light of Farley v Skinner  UKHL 49,  2 AC 732; contracts where a major, but not necessarily the only, purpose is the provision of peace of mind: Farley v Skinner  UKHL 49,  2 AC and Hamilton Jones v David & Snape (a firm)  EWHC 3147 (Ch) and awards made in respect of: a ‘loss of amenity’: Ruxley Electronics v Forsyth  3 All ER 268 (HL) and Newman v Framewood Manor Management Co  EWCA Civ 159; physical inconvenience: Hobbs v London and South Western Rly Co (1875) LR 10 QB 111 and Milner v Carnival plc  EWCA Civ 389,  1 Lloyd’s Rep 374; and mental distress consequent upon physical inconvenience: Watts v Morrow  1 WLR 1421 (CA) and O’Carroll v Ryanair (2009) SCLR 125.
122 Ugo Mattei uses a similar, but non-alliterative, example of a ‘nervous father’ planning his daughter’s wedding in the opening paragraph of ‘The Comparative Law and Economics of Penalty Clauses in Contracts’ (n 92).
123 Either as liquidated or unliquidated damages. Goetz and Scott (n 7) at fn 56 refer to Muldoon v Lynch 6 P 417 (1885) at 539–40, where a contract totalling US$18,788 for the erection of a marble monument on the defendant husband’s grave provided for liquidated damages of US$10 per day for delayed completion. The court rejected any claim for compensation on the basis that: ‘There is nothing in this case to indicate that the defendant has suffered any actual damage which can be measured or compensated by money’.
124  1 WLR 1428 at 1448 (CA). This approach may have received indirect support from Clarke and Buxton LJJ at  in Murray v Leisureplay Plc  EWCA Civ 963,  IRLR 946,  All ER (D) 428 (Jul) at , who expressly disagreed with Arden LJ’s approach, which took as its starting point for any investigation into the validity of a stipulated damages clause a comparison between the ‘amount that would be payable under the terms of the agreement and … what would have been payable if [the party] had had to bring its claim under the common law’. Cf the Law Commission’s Working Paper No 61 (n 60) para 44, which suggests that it is the amount that could be recovered as unliquidated damages that should be considered and the similar sentiment expressed in the Court of Appeal decision in El Makdessi v Cavendish Square Holding BV  EWCA Civ 1539 at , where it was said that the fact that a clause provides for a payment in excess of what would be recovered as unliquidated damages ‘will indicate’ that it is a penalty, although ‘it is not determinative’.
125 Diplock LJ’s argument equates the inclusion of a stipulated damages clause as equivalent to notification before contracting of an unusual loss that would result from breach, which would, as a result, be recoverable as unliquidated damages under the second limb of Hadley v Baxendale (1854) 9 Exch 341, whereby the victim may also recover damages in respect of losses ‘which may reasonably be supposed to have been in the contemplation of the parties at the time they made the contract as the probable result of the breach’.
126 H G Beale, Remedies for Breach of Contract (Sweet & Maxwell 1980) 57, which view is supported, without reference, by Leggatt J in MSC Mediterranean Shipping Co SA v Cottonex Anstalt  EWHC (Comm) at  affirmed by the Court of Appeal at  EWCA Civ 789.
127 In a ‘classic’ economic article, Nobel laureate Oliver Williamson described the medieval practice of exchanging hostages as a means of signalling a credible commitment to exchange, which he relates to modern forms of contracting: see Oliver Williamson, ‘Credible Commitments: Using Hostages to Support Exchange’ (1983) 73(4) American Economic Review 519. The author of this book was privileged and awed when as a postgraduate student at the University of Oxford he heard Oliver Williamson present an early version of this article and became incoherent when probed gently by Williamson about his own research.
134 See generally V P Goldberg, ‘Cleaning up Lake River’ (2008) 3 Virginia Law and Business Review 427, 436, reproduced in an updated form as ch 7 in V P Goldberg, Rethinking Contract Law and Contract Design (Edward Elgar Publishing 2015) 76.
135 E Yorio, Contract Enforcement: Specific Performance (Aspen Publishers Inc 1989) para 19.2.2 and Rowan, Remedies for Breach of Contract (n 18) 234. Such clauses may be objected to on the ground that their enforcement usurps the role of the court in deciding upon the appropriate remedy between specific performance and damages.
137 Rowan, Remedies for Breach of Contract (n 18) 230 argues that such arguments are ‘largely misguided’ because ‘the force of freedom of contract should extend to [all] remedial terms’. On the asymmetry between the legal treatment of terms that enlarge, as opposed to those that reduce, remedial recovery see Bildfell, ‘Exculpatory Clauses and Liquidated Damages Clauses’ (n 56) 361–2.
138 See the discussion in section B above.
140 ibid 603.
141  2 All ER 515 at 523, applying the ‘old’ test derived from Lord Dunedin’s judgment in Dunlop Pneumatic Tyre Co Ltd v New Garage and Motor Company Ltd (n 14).
142 For a similar statement in an American case see In re A J Lane & Co 113 BR 821 at 828 (D Mass 1990): ‘Relief is afforded even when the transaction is fully voluntary and the parties have equal bargaining power’.
144 Fahim Imam-Sadeque v Bluebay Asset Management (Services) Ltd  EWHC 3511 (QB) at  (Popplewell J): ‘As I understand it, the justification for the court interfering in the parties’ freedom of contract, in the limited class of case to which the doctrine applies, is that it regards as unconscionable a contract which enables a party to recover more than the amount of the loss which he has suffered by reason of the breach’.
145 Cavendish (n 3) .
147 ibid at : ‘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’.
148 O’Dea v Allstates Leasing System (WA) Pty Ltd (1983) 152 CLR 359 at 399–40; AMEV-UDC Finance Ltd v Austin (n 67) at 190, 193, 198, 201, and 218, Ringrow Pty Ltd v BP Australia Pty Ltd (n 67) at . Gray in ‘Contractual Penalties in Australian Law after Andrews’ (n 15) at 25 argues that the Australian High Court’s failure in Andrews v Australia and New Zealand Banking Group Ltd (n 15) to subsume the penalty jurisdiction ‘[i]nto the principles of unconscionability as an organising principle’ was ‘an opportunity missed’.
149 Cavendish Square Holding BV v Makdessi; ParkingEye Ltd v Beavis (Consumers’ Association Intervening) (n 3) at  and , respectively.
152 Cavendish (n 3) .
154 See T Downes, ‘Rethinking Penalty Clauses’ in P Birks (ed), Wrongs and Remedies in the Twenty-First Century (Clarendon Press 1996) and M Chen-Wishart, ‘Controlling the Power to Agree Damages’ in P Birks (ed), Wrongs and Remedies in the Twenty-First Century (Clarendon Press 1996) ch 12 at 267 and 299, respectively: ‘[t]he way is now open for the rule against penalties to be based upon unconscionability and relief from oppression’; and ‘The necessary realignment of the penalty rule can then be informed … by our understanding of unconscionability’. A more subtle argument in favour of a role for unconscionability has been proposed by DiMatteo, ‘A Theory of Efficient Penalty’ (n 60) at 716–18. The crux of his argument is that stipulated damages clauses should not be enforced beyond the point where they begin to deter so-called efficient breach. However, the non-enforcement of such clauses results in the party for whose benefit the clause operates having paid by an increased consideration for a benefit that he will not receive. Consequently, DiMatteo (n 60) suggests replacing the currently applicable reasonableness in the US with one based on unconscionability which will only result in the non-enforcement of the clause where there is oppression or unfair surprise.
156 AMEV-UDC Finance Ltd v Austin (n 67) 193–4.
160 To the authorities above can be added Nicholls LJ’s statement in Jobson v Johnson  1 All ER 621 at 633 that, once a court classifies a term as a penalty, ‘[t]he legal consequence which follows, as day follows night’ is that the amount stipulated will not be recovered and ‘[t]hat consequence … is not dependent on … the circumstances’.
161 In Canada, later case law seemed to retreat from the approach of Elsley v JG Collins Insurance Agencies (1978) 83 DLR 1 at 15. See Peachtree II Associates – Dallas LP et al v 857486 Ontario Ltd (2005) 256 DLR (4th) 490 and the discussion in Bildfell, ‘Exculpatory Clauses and Liquidated Damages Clauses’ (n 56) 361–2.
162 Cavendish (n 3) at  (Lords Neuberger and Sumption).
164 See eg Lord Woolf in Philips Hong Kong Ltd v Attorney-General of Hong Kong (n 69) 59: ‘[t]he court has to be careful not to set too high a standard and bear in mind that what the parties have agreed should normally be upheld. Any other approach will lead to undesirable uncertainty especially in commercial contracts’. See also Robophone Facilities v Blank  3 All ER 128,  1 WLR 1428 (CA) at 1447 (Diplock LJ); Murray v Leisureplay Plc (n 124)  and  (Arden and Buxton LJJ); Meretz Investments NV v ACP Ltd  Ch 197 (Lewison J) (upheld on appeal on this point at  Ch 244) and Cadogan Petroleum Holdings Ltd v Global Process Systems  EWHC 214 (Comm) at  (Eder J), referring to the statements in Murray (ibid).
165 Philips Hong Kong Ltd v Attorney-General of Hong Kong (n 69) 57–9.
166 Referring also to Lord Browne-Wilkinson in the Privy Council in Workers Trust & Merchant Bank v Dojap Investments (n 63) 580 and Mason and Wilson JJ in the High Court of Australia in AMEV-UDC Finance Ltd v Austin (n 67) 190.
168 Ian Macneil, ‘The Power of Contract and Agreed Remedies’ (1962) 47 Cornell Law Review 495 at n 26.
170 See Muir, ‘Stipulations for the Payment of Agreed Sums’ (n 27) 519. This argument from consistency is discussed further below at nn 279 and 280.
171 For general overviews of the economic literature see De Geest and Wuyts, ‘Penalty Clauses and Liquidated Damages’ (n 83) 150; Edlin and Schwartz, ‘Optimal Penalties in Contracts’ (n 6); S Walt, ‘Penalty Clauses and Liquidated Damages’ ch 10 in G De Geest (ed), Contract Law and Economics (Edward Elgar Publishing 2011); and Pressman ‘The Two-Contract Approach to Liquidated Damages’ (n 80) 659–65. For a critical overview see Posner, ‘Economic Analysis of Contract Law after Three Decades’ (n 12) 859–63.
172 See section B above; Scott and Triantis, ‘Embedded Options and the Case against Compensation in Contract Law’ (n 37). For a more general overview of their approach see Scott and Triantis, ‘Incomplete Contracts and the Theory of Contract Design’ (n 37). See also Rea’s earlier article ‘Efficiency Implications of Penalties and Liquidated Damages’ (n 86).
173 See above section C(1); Stole, ‘The Economics of Liquidated Damage Clauses in Contractual Environments with Private Information’ (n 84) 582.
174 See above section C(2). See also Goetz and Scott, ‘Liquidated Damages, Penalties and the Just Compensation Principle’ (n 7).
175 ibid. See also Mattei, ‘The Comparative Law and Economics of Penalty Clauses in Contracts’ (n 92).
177 See the preponderance of ‘anti’ over ‘pro’ penalty rule articles cited above in n 84.
180 Iron Trade Products Co v Wilkoff Co 116 A 150 (1922). The buyers’ conduct in the Iron Trade case was apparently held to be permissible and so if there were an enforceable penalty clause in favour of the buyer, it would constitute the kind of breach-inducement activity that Clarkson, Miller, and Muris (n 86) were discussing.
181 A T Kronman and R A Posner, The Economics of Contract Law (Little, Brown 1979) 225 suggest that ‘[t]he refusal to enforce a penalty clause on this ground is difficult to reconcile with efficiency’, because if there was a prospect of breach inducement activity the party who it would impact adversely upon would reflect this fact in the price he was prepared to pay. The behaviour described in the original article rather than critique seems better to reflect intuitions about the behaviour of contractors.
183 ibid 244.
184 ibid 239.
185 This forms the basis of Rea’s critique of Rubin’s argument: Rea, ‘Efficiency Implications of Penalties and Liquidated Damages’ (n 86) 149.
188 The situation where a change from one situation to another will make at least one person better off and no-one worse off is said to be Pareto optimal. For a comparison with other approaches to efficiency, particularly so-called Kaldor Hicks efficiency, where a change from one situation to another is said to be efficient if the net gains from that move are greater than the net losses, irrespective of the distribution of those gains and losses, see M J Trebilcock, The Limits of Freedom of Contract (Harvard University Press 1997) 7–8. For a more technical discussion of Vilfredo Pareto’s contribution to economics and especially his eponymous conception of efficiency see A M Feldman in P Newman (ed), The New Palgrave Dictionary of Economics and the Law, Vol 3 (Palgrave Macmillan 1998) 5. For a ‘personal’ illustration of the concept see A I Ogus, Costs and Cautionary Tales: Economic Insights for the Law (Hart Publishing 2006) 206–10.
189 The efficiency analysis of law is sometimes described by the mixed metaphor ‘growing the size of the pie’ in order to make clear its limited distributional ambition. It is not concerned with how those gains are divided, ie how the pie is sliced, and has been criticised on this basis: see J L Harrison, ‘Trends and Traces: A Preliminary Evaluation of Economic Analysis in Contract Law’ (1988) Annual Survey of American Law 73, 99.
190 See R L Birmingham, ‘Damage Measures and Economic Rationality: the Geometry of Contract Law’  Duke Law Journal 49, 63. Cf Macneil, ‘Efficient Breach of Contract’ (n 85), who traces the concept to Goetz and Scott, ‘Liquidated Damages, Penalties and the Just Compensation Principle’ (n 7).
192 R Posner, ‘Let us Never Blame a Contract Breaker’ (2009) 107 Michigan Law Review 1349 at 1350, also published as ch 1 in O Ben-Shahar and A Porat (eds), Fault in American Contract Law (Cambridge University Press 2010) 5.
197 The classic statements of the expectation measure of recovery is that of Parke B in Robinson v Harman (1848) 1 Exch 850 at 855: ‘the rule of the common law is that where a party sustains a loss by reason of a breach of contract he is so far as money can do it to be placed in the same situation as if the contract had been performed’. See, to similar effect, Viscount Haldane LC in British Westinghouse Electric and Manufacturing Co Ltd v Underground Electric Rlys Co of London Ltd  AC 673 at 689 (HL): ‘[a]s far as possible, he who has proved a breach of a bargain to supply what he contracted to get is to be placed, as far as money can do it, in as good a situation as if the contract had been performed. The fundamental basis is thus compensation for pecuniary loss naturally flowing from the breach’.
200 A T Kronman, ‘Specific Performance’ (1978) 45 University of Chicago Law Review 351 argued that specific performance was justified where goods have ‘unique’ characteristics. When specific performance is awarded the promisee is in a strong position to demand a substantial payment to ‘release’ the promisor from the order. When unliquidated damages are awarded there is a substantial risk of under compensation. Kronman effectively seeks to balance these two risks. In the case of so-called unique goods he concludes that the risk of under compensation is greater and so is sufficient to favour an award of specific performance. Cf Schwartz, ‘The Case for Specific Performance’ (1979) 89 Yale Law Journal 271, who argues that the risk of under compensation is more pervasive than Kronman admits and sufficient to justify the adoption of specific performance as opposed to an award of damages as the default remedy for breach of contract.
201 R Posner, ‘Let us Never Blame a Contract Breaker’ (n 192) 5: ‘If A breaks his contract with B to sell to C because C will pay more than the harm (which equals damages) to B from the breach, the breach increases the social product: B is no worse off, and A and C are both better off. But if B is entitled to specific performance, A cannot sell to C without paying B to agree to terminate A’s contract with him, creating a bilateral monopoly situation’.
203 ibid. For a richer and necessarily more complex overview of this area see Edlin and Schwartz, ‘Optimal Penalties in Contracts’ (n 6).
204 Cf n 172.
205 Klass, ‘Contracting for Cooperation in Recovery’ (n 202) 55: ‘[t]hey decrease the size of the pie the parties have to divide between them, a result no one prefers’ referring to Rea, ‘Efficiency Implications of Penalties and Liquidated Damages’ (n 86) 156.
206 See Macneil, ‘Efficient Breach of Contract’ (n 85); Friedmann, ‘The Efficient Breach Fallacy’ (n 193) and A Schwartz and D Markovits, ‘The Myth of Efficient Breach’ (2010) Yale Law School Faculty Scholarship Series Paper 93.
207 Friedmann, ‘The Efficient Breach Fallacy’ (n 193); H Collins, Regulating Contracts (Oxford University Press 1999) at 119–20 and D Kimel, From Promise to Contract: Towards a Liberal Theory of Contract (Bloomsbury 2005) 76.
209 R A Hillman, The Richness of Contract Law: An Analysis and Critique of Contemporary Theories of Contract Law (Kluwer Academic Publishers 1998) 220–1. Hillman notes at 225 that: ‘As a consequence “law and economics is becoming more sophisticated about some contract-law issues” but that this simplification is achieved at a cost. By abandoning simplifying assumptions, these “second generation” theories may lose much of their clarity and simplicity’. For an example of this process in the context of liquidated damages see the contrast in Edlin and Schwartz, ‘Optimal Penalties in Contracts’ (n 6) between their analysis of ‘The Early Literature’ from 1977–93 at 37–42, with their account of ‘Modern Investment Models’ at 43–52.
212 As in ‘Pareto-optimal’. Cf n 188 above.
214 An interesting empirical study by T Wilkinson-Ryan, ‘Do Liquidated Damages Encourage Breach? A Psychological Experiment’ (2010) 108 Michigan Law Review 633 does not address the issue raised by Macneil, ie the frequency with which opportunities for gain through efficient breach will arise. Rather, at 633, it ‘offers experimental evidence that parties are more willing to exploit efficient-breach opportunities when the contract in question includes a liquidated damages clause’. The article suggests at 646 that the inclusion of a liquidated damages clause may ‘force parties to deliberate about the expected benefit of the contract in a way that would otherwise be cursory or confused’. Reporting at 652 on an experiment concerning day care centres, it was found that parents were more likely to be late picking up their children after a system of fines for late collection was introduced. While different explanations were possible the study does yield experimental evidence that if the opportunity for breach is recognised the presence of a clear penalty seems to encourage, not deter, that behaviour.
216 See section B above.
218 Modern law and economics scholarship is often said to commence with R Coase’s celebrated article ‘The Problem of Social Cost’ (1960) 3 Journal of Law and Economics 1 (‘[t]he foundation of the modern application of economics to all fields of law’, cited by the Nobel committee in awarding Coase the Nobel Prize in Economics in 1991. See also William Landes, ‘The Empirical Side of Law and Economics’ (2003) 70 University of Chicago Law Review 167; Calabressi was responsible for developing the application of economic theory to the law of torts: see G Calabressi, ‘Some Thoughts on Risk Distribution and the Law of Torts’ (1961) 70 Yale Law Journal 499.
219 The statement was made by Eric Posner in ‘Economic Analysis of Contract Law after Three Decades’ (n 12) 829. Eric Posner is a Professor at the University of Chicago, where his father, Richard Posner, now Judge Posner, was also a professor when he published the seminal work on the subject Economic Analysis of Law in1973. The traditional law and economics movement is associated with the University of Chicago Law School where Ronald Coase, Richard Posner, and William Landes were professors in the 1960s and where the leading journals the Journal of Legal Studies and the Journal of Law and Economics were founded. As a result, the hypothetical actor which this school of scholarship builds on has been termed ‘Chicago Man’. See R Prentice, ‘Chicago Man, K-T Man, and the Future of Behavioural Law and Economics’ (2003) 56 Vanderbilt Law Review 1663, 1666.
220 A critic of the economic analysis of law and former Dean of Yale Law School, Anthony Kronman, admits the impact of this scholarship in the US: ‘The law and economics movement was and continues to be an enormous enlivening force in American legal thought and, I would say, today continues and remains the single most influential jurisprudential school in this country’. See A Kronman, ‘The Second Driker Forum for Excellence in the Law’ (1995) 42 Wayne Law Review 115 at 160. See also A Dailey, ‘The Hidden Economy of the Unconscious’ (1999) 74 Chicago-Kent Law Review 1599, 1600: ‘Economic analysis has without question enjoyed a powerful and widespread influence within the legal academy’ and W Wendel, ‘Mixed Signals: Rational-Choice theories of Social Norms and the Pragmatics of Explanation’ (2002) 77 Indiana Law Journal 1, 8.
223 For a critique, but with no favoured solution see Posner, ‘Economic Analysis of Contract Law after Three Decades’ (n 12) 865–8 and 872–3.
225 For example ‘behavioural law and economics’: see T Ulen, ‘The Growing Pains of Behavioral Law and Economics’ (1998) 58 Vanderbilt Law Review 1747; C Jolls, C Sunstein, and R Thaler, ‘A Behavioural Approach to Law and Economics’ in C Sunstein (ed), Behavioral Law and Economics (n 224) 13–21. Other more general descriptors include the ‘“new” law and psychology’, eg J J Rachlinski ,‘The “New” Law and Psychology: A Reply to Critics, Sceptics, and Cautious Supporters’ (2000) 85 Cornell Law Review 739, especially the discussion at 739–41; ‘cognitive psychology’ eg R Scott, ‘Error and Rationality in Individual Decisionmaking: An Essay on the Relationship Between Cognitive Illusions and the Management of Choice’ (1986) 59 Southern California Law Review 329 and accompanying text; ‘decision theory’ eg R Prentice, ‘Chicago Man, K-T Man, and the Future of Behavioural Law and Economics’ (n 219) 1671–8.
228 Bounded rationality is also a cornerstone of so-called transaction cost or neo-institutional economics, ie ‘[economic] studies which assume bounded rationality differ qualitatively from studies based upon rational choice’. See Thrainn Eggertsson in P Newman (ed), The New Palgrave Dictionary of Economics and the Law, Vol 2 (Palgrave Macmillan 1998) 665. This tradition of law and economics scholarship emphasises the role of ‘transaction costs’ and the discriminating matching of exchange activity and governance structures. For an overview by one of its leading exponents see Oliver Williamson, ‘Markets and Hierarchies’ (1977) 24 Journal of Law and Society 574; Oliver Williamson, ‘Transaction Cost Economics: The Governance of Contractual Relations’ (1979) 22 Journal of Law and Economics 233; O Williamson, ‘The Economics of Organisation: The Transaction Cost Approach’ (1981) 87 American Journal of Sociology 548; O Williamson, The Economics of Capitalism (The Free Press 1985); P Newman (ed), The New Palgrave Dictionary of Economics and the Law, Vol 2 (Palgrave Macmillan 1998) 703–10.
229 H A Simon, Administrative Behaviour (2nd edn, The Free Press 1961) xxiv. See also Garvin ‘Disproportionality and the Law of Consequential Damages: Default Theory and Cognitive Reality’ 59 Ohio St L J 339 at 391–4 for a discussion of so-called bounded rationality.
230 See generally A Tversky and D Kahneman, ‘Judgment under Uncertainty: Heuristics and Biases’ in D Kahneman, P Slovic, and A Tversky (eds), Judgment Under Uncertainty: Heuristics and Biases (Cambridge University Press 1982) ch 1.
231 To illustrate the kind of ‘experiments’ that support the heuristics, biases, and effects identified by BDT the availability heuristic was supported in an experiment where researchers recited lists of well-known people to subjects. In some lists, the men were more famous, in others the women. When asked if men or women were the most represented in a list subjects mistakenly concluded that the class with the most famous personalities was the most numerous. See A Tversky and D Kahneman, ‘Availability: A Heuristic for Judging Frequency and Probability’ (1973) Cognitive Psychology 207, 220–4.
232 A related but distinct bias is overconfidence meaning the tendency to be more confident in behaviours, attributes, and physical characteristics than is justified: see B D Pulford and A Colman, ‘Overconfidence, Base Rates and Outcome Positivity/Negativity of Predicted Events’ (1996) 87 British Journal of Psychology 431, 431. Overconfidence ‘may also cause people to persist in situations where their expected outcome is poor’ according to Baffi, ‘Efficient Penalty Clauses with Debiasing’ (n 131) 1000, earlier referring to Pulford and Coleman’s article.
234 O Svenson, ‘Are We All Less Risky and More Skilful than Our Fellow Drivers Are?’ (1981) 47 Acta Psychologica 143 referred to by Eisenberg, ‘The Limits of Cognition and the Limits of Contract’ (n 217) 216.
236 B Korobkin, ‘The Status Quo Bias and Contract Default Rules’ (1998) 83 Cornell Law Review 608, 625. This is closely related to something known as the ‘offer asking gap’ or ‘endowment effect’, according to which individuals will often value an entitlement more highly if they own it than if they do not. See respectively D Kennedy, ‘Cost-Benefit Analysis of Entitlement Problems: A Critique’ (1981) 33 Stan L Rev 387 at 401 and R Thaler, ‘Toward a Positive Theory of Consumer Choice’ (1980) 1 J Econ Behav & Org 39, 44. For example, I have a bottle of wine in my cellar (a 1982 Château Gruaud Larose) that I would not consider selling for less than £400 when I could buy another for £300. The example and the bottle belong to the author, although the latter may be consumed when this book is published.
237 W Edwards and D von Winterfeldt, ‘Symposium: Legal Implications of Human Error: Cognitive Imperfections: Consumer Law Inferences: Cognitive Illusions and their Implications for the Law’ (1986) 59 Southern California Law Review 225, 243.
238 B Fischoff, ‘For Those Condemned to Study the Past: Reflections on Historical Judgment’ in R A Schweder and D W Fiske (eds), New Directions for Methodology of Social and Behavioral Science: Fallible Judgment in Behavioural Research (Jossey-Bass 1980).
239 D Kahneman and A Tversky, ‘Choices, Values and Frames’ (1979) 39 Am Psychologist 341, 341: ‘[p]eople prefer the sure thing over the gamble, although the gamble has higher (mathematical) expectation’. See also C R Fox and A Tversky, ‘Ambiguity Aversion and Comparative Ignorance’ (1995) 110 Quarterly Journal of Economics 585.
243 ibid 228.
244 Hillman, ‘The Limits of Behavioural Decision Theory in Legal Analysis’ (n 9) 738 (original emphasis).
249 Rachlinski, ‘The “New” Law and Psychology: A Reply to Critics, Skeptics and Cautious Supporters’ (n 225) 748–9. Cf Baffi, ‘Efficient Penalty Clauses with Debiasing’ (n 131) 1003, who criticises the overgeneral approach of Rachlinski.
250 Note that in some respects the law in the US is different to that in the UK; see section I below
252 See Posner, ‘Economic Analysis of Contract Law after Three Decades’ (n 12) 860: ‘if we take the argument seriously, we should apply it not only to remedial terms. The same logic applies to the price term or, indeed, any other term of a contract’.
253 Korobkin, ‘The Status Quo Bias and Contract Default Rules’ (n 236) 661 and R H Thaler, ‘The Psychology and Economics Conference Handbook: Comments on Simon, on Einhorn and Hogarth, and on Tversky and Kahneman’ in R M Hogarth and M W Reder (eds), Rational Choice: The Contrast between Economics and Psychology (University of Chicago Press 1987).
256 ibid 1886.
257 ibid 1886–7.
258 ibid 1896.
260 Cf ibid 1901.
262  UKHL 49,  2 AC 732 at . In British Westinghouse Electric and Manufacturing Co Ltd v Underground Electric Rlys Co of London Ltd  AC 673 at 689, HL Viscount Haldane LC said: ‘the fundamental basis [of an award of contractual damages] is thus compensation for pecuniary loss naturally flowing from the breach’ and in Johnson v Agnew  AC 367 at 400 (HL) Lord Wilberforce observed that: ‘the general principle for the assessment of damages is compensatory’.
263 A-G v Blake (n 106).
265 UCC 2-718 Liquidation or Limitation of Damages; deposits (1) Damages for breach by either party may be liquidated in the agreement but only at an amount that is reasonable in the light of the anticipated or actual harm caused by the breach, the difficulties of proof of loss, and the inconvenience or non-feasibility of otherwise obtaining an adequate remedy.
266 Restatement 2d 356 Liquidated Damages and Penalties (1) Damages for breach by either party may be liquidated in the agreement but only at an amount that is reasonable in the light of the anticipated or actual loss caused by the breach, the difficulties of proof of loss. A term fixing unreasonably large liquidated damages is unenforceable on the grounds of public policy as a penalty.
267 Restatement 2d 356(1) speaks of ‘loss’ and the UCC 2-718(1) speaks of ‘harm’ but cf Farnsworth, Farnsworth on Contracts (n 81) 305–11, doubting the continued importance of actual, as opposed to presumed, loss.
268 See Lord Hodge in Cavendish (n 3) at , who expressly said that ‘[t]he court … approaches the matter as a question of construing the particular contract at the time when it was made’.
269 For a discussion of this category see ch 2F.
272 ibid 451, apparently endorsing the argument of Clarkson, Miller, and Muris ‘Liquidated Damages v Penalties: Sense or Nonsense’ (n 86).
274 A-G v Blake (n 106).
275 Burrows, Remedies for Torts and Breach of Contract (n 271) 449.
276 A-G v Blake (n 106). The ‘solitary’ (at 637) example where a non-compensatory award was given in respect of a simple breach of contract was Wrotham Park Estate Co Ltd v Parkside Homes Ltd  2 All ER 321,  1 WLR 798, applied in Experience Hendrix LLC v PPX Enterprises Inc  EWCA Civ 323,  1 All ER (Comm) 830 and One Step (Support) Ltd v Morris-Garner  EWCA Civ 180.
277 A-G v Blake (n 106) at 638, 639, and 640 (Lord Nicholls), Lords Goff and Browne-Wilkinson agreeing.
279 See further ch 2C.
280 See Downes, ‘Rethinking Penalty Clauses’ (n 154) 265. The author (at 267) refers to an older empirical study of engineering contractors in the Bristol area conducted by H Beale and A Dugdale, ‘Contracts between Businessmen: Planning and the Use of Contractual Remedies’ (1975) 2 Brit J Law and Soc 18, 45, and noted that when, infrequently, stipulated damage clauses were included in contracts it was ‘often as a limitation on the loss recoverable which was more acceptable than a complete exclusion clause’.
281 See Muir, ‘Stipulations for the Payment of Agreed Sums’ (n 27) 519.
282 ibid n 83, referring to: doctrines relating to non-severable entire obligations, total failure of consideration where some benefit has nonetheless been conferred, deposits, deviation, stipulations for payment of agreed sums on events other than breach of contract, performance bonds.
284 Following a detailed review of the arguments and case law, the Singapore Court of Appeal endorsed the English position in PH Hydraulics & Engineering Pte Ltd v Airtrust (Hong Kong) Ltd  SGCA 26.
285 In Johnson v Unisys Ltd  UKHL 13 at 15,  1 AC 518,  2 WLR 1076 at 1082, Lord Steyn acknowledged that such awards ‘have never and cannot be awarded for breach of any contract’. See also Perera v Vandiyar  1 All ER 1109,  1 WLR 672, CA; Kenny v Preen  1 QB 499, CA; Tito v Waddell (No 2)  Ch 106 at 332; Reed v Madon  Ch 408; Ruxley Electronics Ltd v Forsyth  AC 344 at 353 and 365 (HL).
286 See Aggravated, Exemplary and Restitutionary Damages (Report No 247 1997) 106, but this proposal would not involve any reform of the law in relation to the availability of damages for breach of contract. Report No 247 199, p 107. In 2009, the Ministry of Justice dismissed the need for legislation and affirmed the distinction between compensation and punishment: Ministry of Justice Civil Law Reform Bill: Consultation CP 53/09.
287 For an explanation of the distinction see ch 3B(2).
288 Bildfell, ‘Exculpatory Clauses and Liquidated Damages Clauses?’ (n 56) 358 examining Canadian law.
290 Cavendish (n 3) . See, to similar effect, Lords Neuberger and Sumption at : ‘[w]hether it is penal … whether it is a pre-estimate of loss. These are not natural opposites or mutually exclusive categories’ and Lord Hodge favouring a ‘[b]roader approach’ which ‘… escapes the [straightjacket]’.
291 The possibility of a third category of stipulated damages provision was suggested by Colman J in Lordsvale Finance Plc v Bank of Zambia  QB 752 at 762 G. This important refinement of the traditional principle was expressly approved by the Court of Appeal in Cine Bes Filmcilik VE Yapimcilik v United International Pictures  EWCA Civ 1669  All ER (D) 312 (Nov). In Murray v Leisureplay Plc (n 124) , Arden LJ suggested that there will be clauses ‘[w]hich may operate on breach, but which fall into neither category’. Buxton LJ, with whom Clarke LJ agreed, at  criticised the approach of Arden LJ stating that it was: ‘[i]mportant to note that the two alternatives, a deterrent penalty; or a genuine pre-estimate of loss; are indeed alternatives, with no middle ground between them’. The denial of the majority in Murray that a deterrent penalty is enforceable cannot stand in the light of the Supreme Court’s decision in Cavendish (n 3).
292  HCA 28 at . See also Keane J at : ‘The real objection, as a matter of public policy, to a penalty clause which operates upon breach of contract is that it is no part of the law of contract to allow one party to punish the other for non-performance’.
297 Cavendish (n 3) at . Cf in Paciocco (n 39) Kiefel J at  and Keane J at , who both appear to rely upon Lord Neuberger and Sumption’s statement to support the prevention of punishment as a freestanding policy justification for the penalty rule.
300 It was quoted with approval by the Court of Appeal in Cine Bes Filmcilik ve Yapimcilik AS v United International Pictures (n 291) and Murray v Leisureplay plc (n 124) at  and  (Arden LJ and Buxton LJ, respectively).
303 332 US 407 (1947) at 417–18, quoted by Gageler J in Paciocco v Australia and New Zealand Banking Group Ltd (n 39) .
304 This tendency was noted by McCormick as long ago as 1935 in The Law of Damages (n 139) at 603.
305 See Rowan, Remedies for Breach of Contract (n 18) 230. The author goes on to say that: ‘[t]he perception that negative consequences would flow from the advent of agreed remedies is largely misguided’.
306 Cavendish (n 3) at  (Lords Neuberger and Sumption), Lord Carnwath agreeing.
307 Andrews v Australia and New Zealand Banking Group Ltd (n 15).
308 Jobson v Johnson (n 160) 631 (Nicholls LJ).
309 See A W B Simpson, A History of the Common Law of Contract (Clarendon Press 1975), D Ibbetson, A Historical Introduction to the Law of Obligations (Oxford University Press 2001) 150–1 and 213–4; J D Heydon, M Leeming, and P Turner, Meagher, Gummow and Lehane’s Equity: Doctrines and Remedies (5th edn, LexisNexis Butterworths 2014) Pt 4 s 7; A W B Simpson, ‘The Penal Bond with Conditional Defeasance’ (1966) 82 LQR 392; E Henderson, ‘Relief from Bonds in the English Chancery’ (1974) American Journal of Legal History 298; Lanyon, ‘Equity and The Doctrine of Penalties’ (n 167); and Muir, ‘Stipulations for the Payment of Agreed Sums’ (n 27) 503–10. For other, mostly brief, judicial discussions in the UK see Wall v Rederiaktiebolaget Luggude  3 KB 66 at 72 (Bailhache J); Bridge v Campbell Discount Co Ltd (n 63) 630–1 (Lord Denning); Jobson v Johnson (n 160) 631–2 (Nicholls LJ); and Lancore Services Ltd v Barclays Bank plc  EWHC 1264 (Ch) at ,  1 CLC 1039.
310 Cavendish (n 3) at  (Lords Neuberger and Sumption).
311 8 & 9 Will 111 c 11, s 8 and 4 & 5 Anne c 3 ss 12 and 13 repealed respectively by SI 1957/1178, reg 7 under powers conferred by the Supreme Court of Judicature Act 1925, s 99(1)(f), (g), and Sch 1 and the Statute Law Revision Act 1948, s 1 and Sch 1.
315 See Carter and others, ‘Contractual Penalties’ (n 67) 109: ‘[t]he task which the court set itself, namely, to provide a meaningful historical account of the evolution of the penalties doctrine over several hundred years within the space of a few pages seems an impossible and futile exercise’.
316 Cavendish (n 3) .
319 Jobson v Johnson (n 160). A substantial share in a football club was sold for a consideration payable in instalments with a provision that if the purchaser defaults on payment he must retransfer the shares at undervalue. When the purchaser defaulted on payment he was not able to claim relief under the usual relief from forfeiture principles because he had failed to comply with disclosure requirements. The Court of Appeal accepted the purchaser’s argument that the retransfer clause was penal. The usual consequence would be that the clause was not enforced but, as was pointed out in Cavendish (n 3) at , this might have resulted in the purchaser being advantaged by his failure to disclose. Perhaps to avoid this consequence the Court of Appeal applied forfeiture principles to the relief from a penalty and held effectively that the penalty clause was enforceable to the extent of any loss caused by the breach and the seller was therefore offered a choice of remedies to effect this.
320 Cavendish (n 3)  (Lords Neuberger and Sumption).
322 It is further suggested that the indeterminacy of a historical approach should have been anticipated by the High Court of Australia in Andrews because on AMEV-UDC Finance Ltd v Austin (n 67) the judgments of Mason and Wilson JJ and that of Deane J relied upon historical analysis to reach different conclusions at ibid104.
323 See Benjamin, ‘Penalties, Liquidated Damages and Penal Clauses in Commercial Contracts’ (n 73) 609 and the discussion in ch 1, nn 38–40 and accompanying text.
324 See further above n 72.
327 See Muir, ‘Stipulations for the Payment of Agreed Sums’ (n 27) 516.
328 Posner, ‘Let us Never Blame a Contract Breaker’ (n 192) also published as O Ben-Shahar and A Porat (eds), Fault in American Contract Law (n 192) at 18. This quotation is in an article/chapter examining the role of fault in contract law and seeks to apply Oliver Holmes’ famous celebrated statement reproduced at n 191 above, which Posner terms ‘the option feature’ to different contractual doctrines including the penalty rule at 1350 and 4, respectively.
330 ibid . Similarly, see the brief, unannotated reference to BGB s 343 in Paciocco (n 39) at  (Kiefel J, French CJ agreeing).
332 For extensive and dispassionate overviews of civil and common law approaches see Mattei, ‘The Comparative Law and Economics of Penalty Clauses in Contracts’ (n 92); P Hachem, Agreed Sums Payable upon Breach of an Obligation (Eleven International Publishing 2011); and the shorter overview in Scottish Law Commission, Review of Contract Law Discussion Paper on Penalty Clauses (Discussion Paper 162 2016) paras 2.26–2.45.
333 Lord Carnwath agreed with the joint judgment of Lords Neuberger and Sumption and Lord Clarke agreed with their reasoning, except with regard to forfeiture and the precise interpretation of the clauses in the Makdessi appeal.
334 See Cavendish (n 3) .
335 The comparative law comments of Lords Mance and Hodge followed a similar pattern, although the latter looked extensively at the development of Scottish law. See Cavendish (n 3) – (Lord Mance); – (Lord Hodge).
336 See Cavendish (n 3) .
338 E Nordin, ‘The Penalty Clause Bias’ (2014) 21 Maastricht Journal of European and Comparative Law 162, 166. For a similarly spirited promotion of the civil over the common law approach to penalty clauses, see A N Hatzis, ‘Having the Cake and Eating It Too: Efficient Penalty Clauses in Common and Civil Contract Law’ (2003) 22 International Review of Law and Economics 381. The suggested ‘superiority’ of civil law is at a general level contradicted by various surveys analysing the jurisdictional preferences of parties engaged in international contracting including the annual World Bank Doing Business survey which ranks countries in terms of the ease of doing business across key metrics and whose top 10 places are always dominated by common law, as opposed to civil law, jurisdictions. In the latest 2018 survey, 5 of the top 7 jurisdictions ranked overall for the ease of doing business are common law based (New Zealand, Singapore, Hong Kong SAR, the UK, and the US). See generally R Halson and D Campbell, ‘Harmonisation and Its Discontents: A Transaction Costs Critique of a European Contract Law’ in J Devenney and M Kenny (eds), The Transformation of European Private Law (Cambridge University Press 2013) 114–15.
340 Garcia, ‘Enforcement of Penalty Clauses in Civil and Common Law: A Puzzle to be Solved by the Contracting Parties’ (2012) 5(1) European Journal of Legal Studies 95, 100 fn 19 and accompanying text.
342 Eg Spanish law will only permit the moderation of the clause if the undertaking has been partly performed (Spanish Civil Code Art 1154); French law empowers the judge to increase the agreed penalty (Code Civile Art 1152). For further discussions of Spanish and French Law on penalties see, respectively, J Turner, ‘Spain: Contracts—Penalty Clauses’ (case comment) (2002) 13(4) International Company and Commercial Law Review 34; and L Miller, ‘Penalty Clauses in England and France: A Comparative Study’ (2004) 53(1) ICLQ 79. An attempt to impose uniformity on Belgium, the Netherlands, and Luxemburg through the Benelux Convention on Penalty Clauses (1973) failed.
343 Such a clause is a ‘dependent’ penalty clause. The legal obligation imposed under it may be contractual or non-contractual (such as tortious). Where a clause sets out to secure the performance of a particular act the party is not bound to do, it is ‘independent’ and not subject to the power to reduce. See further F Faust, ‘Contractual Penalties in German Law’ (2015) 23(3) European Review of Private Law 285, 288, 289–90.
344 If a merchant, the party may resort only to legal grounds invalidating the clause. As an exception, in ‘extreme cases’ the agreed sum may be reduced, albeit by a smaller margin, where the agreed sum is considered to contradict the principle of good faith: Faust (n 343) 294.
345 At  namely: UNIDROIT Principles of International Commercial Contracts (2010) art 7.1.13 and UNCITRAL Uniform Rules on Contract Clauses for an Agreed Sum Due on Failure of the Performance art 6.
347 For contrasting views as to whether and if so on what basis the CISG’s general provisions may apply to stipulated damages clauses see P Hachem, ‘Fixed Sums in CISG Contracts’ (2010) 13 Vindobona Journal of International Commmerical Law and Arbitration 217; B Zeller, ‘Penalty Clauses: Are They Governed by the CISG?’ (2011) 23 Pace International Law Review 1; and J Graves ‘Penalty Clauses and the CISG’ (2012) 30 Journal of Law and Commerce 153.
349 Farnsworth, Farnsworth on Contracts (n 81) 812, n 5 and I M Garcia, ‘Enforcement of Penalty Clauses in Civil and Common Law: A Puzzle to be Solved by the Contracting Parties’ (2012) 5(1) European Journal of Legal Studies 98, n 86, quoting Farnsworth, Farnsworth on Contracts (n 81).
350 Above n 5.
351 HF Clarke Ltd v Thermidaire Corp  1 SCR 319 (‘and judicial interference with the enforcement of what the Courts regard as penalty clauses is simply a manifestation of a concern for fairness and reasonableness, rising above contractual stipulation, whenever the parties seek to remove from the Courts their ordinary authority to determine not only whether there has been a breach but what damages may be recovered as a result thereof’). For academic support for this view: S M Waddams, The Law of Contracts (5th edn, Canada Law Book Inc 2005) 323, .
352 Elsley v JG Collins Insurance Agencies (n 161) (Dickson J). Similarly, AMEV-UDC Finance Ltd v Austin (n 67) 162 CLR 170, 193–4 (Mason and Wilson JJ). This approach had been rejected in the UK before, eg by Arden LJ in Murray v Leisureplay plc (n 124)  and then confirmed by the Cavendish decision. For more detail see the discussion above in section A.
353 In Banta v Stamford Motor Co 92 A 665 (Conn 1914) 667–8, the test for a valid liquidated damages clause was said to have three elements: (1) the damages to be anticipated as resulting from the breach must be uncertain in amount or difficult to prove; (2) there must have been an intent on the part of the parties to liquidate them in advance; and (3) the amount stipulated must be a reasonable one, that is to say, not greatly disproportionate to the presumable loss or injury. In time, the second requirement was dropped and the third requirement was relaxed so that it could be satisfied if the amount was reasonable either in relation to the loss anticipated at the time of contracting or the loss that actually transpired: Restatement (Second) of Contracts, art 356.
355 See section C above.
356 See section F above.
357 See section G above.
359 See n 209.
360 For a rare judicial endorsement in the UK of ‘economic efficiency’ as a ‘policy goal’ see Leggatt J in MSC Mediterranean v Cottonex Anstalt (n 25) (affirmed  EWCA Civ 789) at , quoted at n 196 above.
363 The Bailii Lecture, ‘Developing commercial law through the courts: rebalancing the relationship between courts and arbitration’ (9 March 2016) para 4 https://www.judiciary.gov.uk/wp-content/uploads/2016/03/lcj-speech-bailli-lecture-20160309.pdf (last accessed 10 July 2017).
364 See section E above.
365 In Elsley v JG Collins Insurance Agencies (n 161) (Dickson J).
366 Cavendish (n 3) at  (Lords Neuberger and Sumption), Lords Carnwath and Clarke agreeing at , at  (Lord Mance), Lord Toulson agreeing at  and at  and  (Lord Hodge). The contrary proposal by the English Law Commission remains unimplemented. See generally the discussion in ch 2B.
367 Andrews v Australia and New Zealand Banking Group Ltd (n 15).
369 This refusal may be supported by reference to behavioural decision theory which suggests that commercial contractors do not learn from their experience in the way one might expect: ‘This unpleasant result prevents the possibility of tracing a distinction between sophisticated parties and naive parties’. See Baffi, ‘Efficient Penalty Clauses with Debiasing’ (n 131) 1004.
370 None of the judgments states any different principles to be applied to consumer contracts. However, only Lord Hodge explicitly rejects the argument at  and Lords Neuberger and Sumption refer to and expressly at  reject the submission that the penalty rule should be abolished ‘in the light of the growing statutory regulation in this field’. In 1977, the Californian Civil Code was amended to make liquidated damage provisions ‘valid unless the party seeking to invalidate the provision establishes that the provision was unreasonable under the circumstances existing at the time the contract was made’ except against a consumer in a consumer case. Cal Civ Code para 1671. See further Farnsworth, Farnsworth on Contracts (n 81) 303.
371 Re-enacting with modification the provisions of Unfair Terms in Consumer Contracts Regulations (1999) SI 1999/2083 and their predecessor the Unfair Terms in Consumer Contracts Regulations (1994) SI 1994/3159.
372 See Andrews v Australia and New Zealand Banking Group Ltd (n 15) : ‘this pattern of remedial legislation suggests the need for caution in dealing with the unwritten law as if laissez faire notions of an untrammelled “freedom of contract” provide a universal value’.
373 See Carter and others, ‘Contractual Penalties’ (n 67) 113–14; and Gray, ‘Contractual Penalties in Australian Law after Andrews’ (n 15).
374 The suggestion from Andrews above that an inference about the proper limits of the common law rule should be made from an examination of the statutory regime is the opposite direction of travel to that proposed by Gageler J in Paciocco v Australia and New Zealand Banking Group Ltd (n 39) , apparently suggesting that the conclusion on ‘statutory unconscionability’ should follow from a consideration of the limits of the common law principle.
375 See section J above.
376 At . See also Lord Mance at : ‘As to abolition, there would have to be shown the strongest reasons for so radical a reversal of jurisprudence which goes back over a century in its current definition and much longer in its antecedents’.
377 See section K above.
378 Or so it would seem. For the law of any other jurisdiction see Benjamin, ‘Penalties, Liquidated Damages and Penal Clauses in Commercial Contracts’ (n 73) 626.