Paciocco v Australia and New Zealand Banking Group Limited

AUSTRALIA AND NEW ZEALAND BANKING GROUP LIMITED RESPONDENT

Paciocco v Australia and New Zealand Banking Group Limited [2016] HCA 28 27 July 2016 M219/2015 & M220/2015

D F Jackson QC with M B J Lee SC and W A D Edwards for the appellants (instructed by Maurice Blackburn)

A C Archibald QC and M H O'Bryan QC with C van Proctor for the respondent (instructed by Ashurst Australia)

Notice: This copy of the Court's Reasons for Judgment is subject to formal revision prior to publication in the Commonwealth Law Reports.

Paciocco v Australia and New Zealand Banking Group Limited

Banker and customer - Rule against penalties - Consumer credit card accounts - Late payment fees - Where late payment fees were $35 and $20 - Where costs actually incurred by respondent upon failure by first appellant to make timeous payment of amounts owing were approximately $3 - Where late payment fees not genuine pre‑estimates of damage - Where respondent alleged it could conceivably have incurred loss provision costs, collection costs and regulatory capital costs as a result of first appellant's default - Whether late payment fees penalties - Whether late payment fees extravagant, exorbitant or unconscionable - Whether late payment fees out of all proportion to interests damaged - Whether respondent's legitimate interests confined to reimbursement of expenses directly occasioned by first appellant's default.

Contract - Rule against penalties - Essential characteristics of a penalty - Whether sum disproportionate to actual loss suffered amounts to a penalty - Whether sum incorporating loss too remote to be recoverable in action for damages amounts to a penalty - Relevance of Dunlop Pneumatic Tyre Co Ltd v New Garage and Motor Co Ltd [1915] AC 79.

Trade practices - Consumer protection - Late payment fees - Unconscionable conduct - Unjust transactions - Unfair terms - Whether late payment fees unconscionable, unjust or unfair.

Precedent - Apex courts of foreign jurisdictions - Status of unwritten law of United Kingdom in Australia.

Words and phrases - "exorbitant", "extravagant", "genuine pre-estimate", "in terrorem", "late payment fees", "liquidated damages", "out of all proportion", "penalty", "unconscionable", "unconscionable conduct", "unfair terms", "unjust transactions".

Australian Securities and Investments Commission Act 2001 (Cth), ss 12BF, 12BG, 12CB, 12CC. National Consumer Credit Protection Act 2009 (Cth), Sched 1 s 76. Fair Trading Act 1999 (Vic), ss 8, 8A, 32W, 32X.

FRENCH CJ. These appeals concern the enforceability of late payment fee provisions in contracts between the first appellant and the respondent bank in relation to consumer credit card accounts. The terms of the impugned provisions are set out in the reasons for judgment of Nettle J. Broadly speaking they required the cardholder, following receipt of a monthly statement of account, to make the "Minimum Repayment" set out on each statement by the due date shown on it. A "Late Payment Fee" was to be charged to the credit card account if the minimum monthly payment, plus any "Amount Due Immediately" shown on the statement of account, was not paid by a specified date. The first appellant, Lucio Paciocco ("Mr Paciocco"), held consumer credit card and deposit accounts with the respondent, Australia and New Zealand Banking Group Limited ("the Bank"). The second appellant, Speedy Development Group Pty Ltd, is a company controlled by Mr Paciocco. It held a business deposit account with the Bank. All of the accounts were charged various fees by the Bank. The consumer and business deposit accounts were charged honour fees, dishonour fees and non-payment fees. The consumer credit card accounts held by Mr Paciocco were charged over-limit fees and late payment fees. Both appellants were applicants in representative proceedings against the Bank under Pt IVA of the Federal Court of Australia Act 1976 (Cth). The appellants alleged that the provisions for the various fees were unenforceable as penalties and, alternatively, that their inclusion contravened various statutory provisions relating to unconscionable conduct and, with respect to Mr Paciocco only, unjust and unfair contract terms. The primary judge, Gordon J, found that the provisions for the late payment fees were penalties at common law and in equity. It was therefore not necessary for her Honour to deal with the statutory claims regarding the late payment fees. Her Honour held that none of the other fees constituted a penalty, nor contravened any of the identified statutory provisions. On appeal, the Full Court of the Federal Court held that the late payment fees were not penalties and did not fall within any of the statutory categories of unconscionable conduct, unjustness or unfairness. The Full Court upheld the primary judge's findings with respect to the other fees. That conclusion is not challenged in these appeals. These appeals, by grant of special leave from the Full Court's decision, are concerned only with the correctness of that decision in respect of the late payment fees. The facts relevant to the appeals and the evidence of contending expert witnesses at trial are set out in the judgment of Gageler J and it is unnecessary to repeat them here. For the reasons given by Kiefel J and, in relation to the statutory claims, for the reasons given by Keane J, I agree that the appeals to this Court should be dismissed. I will, however, add some comments to that concurrence. The question whether various fees charged by the Bank to its credit card customers were unenforceable as penalties was raised in an earlier representative proceeding before Gordon J, sub nom Andrews v Australia and New Zealand Banking Group Ltd. An application for leave to appeal from an interlocutory decision of Gordon J in those proceedings, to the Full Court of the Federal Court, was removed into this Court pursuant to s 40(2) of the Judiciary Act 1903 (Cth). The interlocutory decision responded to separate questions asked by the applicants in that case. They included questions whether honour, dishonour, non-payment and over-limit fees, and the late payment fees in issue in these appeals, were payable on breach of the relevant contract by the customer, or upon the occurrence of events amounting to a default under the contract which the customer had an obligation to avoid, and whether they were capable of being characterised as penalties by reason of either of those facts. The framing of those questions required consideration of whether the unwritten law making penalties unenforceable was limited to cases in which the putative penalty was enlivened by a breach of contract. The primary judge held that the rule against penalties was limited to penalties arising out of breach of contract, that only the late payment fees were payable upon breach, and that the rule could therefore only be applied to those fees. In so doing, her Honour properly followed the decision of the Court of Appeal of the Supreme Court of New South Wales in Interstar Wholesale Finance Pty Ltd v Integral Home Loans Pty Ltd. This Court, on the removed application for leave to appeal from her Honour's interlocutory decision, granted leave to appeal and allowed the appeal. The Court held that equitable relief against penalties had not been subsumed into the common law rule and that the rule against penalties was not limited to cases arising out of a breach of contract. Subsequently, the present appellants commenced these proceedings, which were also heard before Gordon J as the primary judge. In Andrews, the Bank did not seek to appeal against her Honour's finding in her interlocutory decision about how the alleged penalty provision with respect to late payment fees operated. There was no attempt to argue in this Court that the penalty provisions in the consumer credit card accounts to which Mr Paciocco was a party operated any differently. This case thus came to this Court as one involving characterisation of a provision for payment of a fee which was, if enforceable, enlivened upon a breach of contract. As Gageler J points out, the decision in Andrews and that of the House of Lords in Dunlop Pneumatic Tyre Co Ltd v New Garage and Motor Co Ltd set out the governing principles so far as they apply to penalties for breach of contract. A difference has emerged since the decision in Andrews between the Supreme Court of the United Kingdom and this Court in relation to the scope of the law relating to penalties. It is not necessary to reflect upon the merits of the different positions as the present appeal on the penalty question falls within essentially undisputed territory. It is, however, desirable to say something about the fact of divergence between our jurisdictions, which have an historical connection that Australia does not have with any other jurisdiction. In Cavendish Square Holding BV v Makdessi, Lord Neuberger of Abbotsbury PSC and Lord Sumption JSC (with whom Lord Carnwath JSC agreed) held that the rule against penalties was confined to cases arising out of contractual breach. Their disagreement with the scope of the law as stated in Andrews was emphatic, describing the decision as "a radical departure from the previous understanding of the law". Their Lordships' language echoed that of Menzies J in this Court half a century earlier in Uren v John Fairfax & Sons Pty Ltd when he declared the limitation on recovery of exemplary damages prescribed by Lord Devlin in Rookes v Barnard to be "a radical departure from what has been regarded as established law." It is not necessary for present purposes to engage with that characterisation of Andrews. Gageler J expresses the view that it was incorrect and based upon a misunderstanding of the scope of what was actually decided in Andrews. In any event, emphatic disagreement between our jurisdictions in relation to the common law and equitable doctrines, while infrequent, is not novel. The countries of the common law world have a shared heritage which they owe to the unwritten law of the United Kingdom. That shared heritage offers the undoubted advantage, but does not import the necessity, of development proceeding on similar lines. It is more than half a century since Dixon CJ said that this Court would no longer adhere to the policy that it ought to follow the decisions of the House of Lords at the expense of its own opinions. That change of direction was occasioned by the judgment of the House of Lords in Director of Public Prosecutions v Smith, which the Chief Justice considered laid down propositions which were "misconceived and wrong". Twenty-five years later that evolutionary change was well entrenched. Mason CJ observed extra-judicially that the value of English judgments, like those of Canada, New Zealand and the United States, "depend[ed] on the persuasive force of their reasoning." So too, no doubt, for the courts of the United Kingdom as they consider the decisions of courts of other common law jurisdictions. The common law in Australia is the common law of Australia. So much was affirmed in the unanimous judgment of this Court in Lange v Australian Broadcasting Corporation and in cases that followed it. Following the enactment of the Australia Acts and the abolition of the last remaining avenue of appeal to the Privy Council from the Supreme Courts of the States, s 80 of the Judiciary Act was amended by substituting the term "common law in Australia" for the term "common law of England". The common law of England was a source of law for legal development in Australia, but not the only source. Moreover, as the alternative claims in the present case demonstrate, there are few areas of the common law which are untouched by statutory regimes reflecting public policy settings which may differ from one jurisdiction to another. Differences have emerged from time to time between the common law of Australia and that of the United Kingdom in a number of areas. Those differences have not heralded the coming of winters of mutual exceptionalism. All of the common law jurisdictions are rich sources of comparative law whose traditions are worthy of the highest respect, particularly those of the United Kingdom as the first source. No doubt in a global economy convergence, particularly in commercial law, is preferable to divergence even if harmonisation is beyond reach. The common law process will not always be the best way of achieving convergence between common law jurisdictions. The penalty rule in the United Kingdom, a product of that process, was described by Lord Neuberger and Lord Sumption in their joint judgment in Cavendish as "an ancient, haphazardly constructed edifice which has not weathered well". More than one account of its construction and more than one view of whether it should be abrogated or extended or subsumed by legislative reform is reasonably open. There has been much activity in this area within national jurisdictions and in the development of internationally applicable model rules and principles which were discussed in Cavendish in the judgments of Lord Mance and Lord Hodge. It may be that in this country statutory law reform offers more promise than debates about the true reading of English legal history.

KIEFEL J. M220/2015 The first appellant, Mr Paciocco, held two credit card accounts with the respondent ("the ANZ"). One of the terms and conditions to which Mr Paciocco agreed in respect of the provision of credit was that a "Late Payment Fee" would be charged to his account if the "Monthly Payment" plus any "Amount Due Immediately" shown on the statement of account which the ANZ issued was not paid by a specified date (being, until December 2009, by 28 days of the end of the "Statement Period" shown on the statement and, from December 2009, by the "Due Date" shown on the statement). The "Monthly Payment" was a reference to a minimum amount which an account holder was required to pay by a certain date. Customers were advised by the ANZ, in the document "ANZ Credit Cards Conditions of Use", that the Late Payment Fee could be avoided by paying the minimum Monthly Payment shown on the statement by the due date. The ANZ fixed the Late Payment Fee from time to time. It did so without consultation with its customers. Until December 2009, the fee was fixed at $35.00 and from December 2009, at $20.00. The ANZ did not suggest that it had determined the level of the fee by estimating the losses which might be occasioned to it by Mr Paciocco's delays in making payments. The fee is in any event one charged generally to customers conducting credit card accounts of this kind who are late in making the Monthly Payment and it is charged regardless of the amount of the Monthly Payment outstanding. In the proceedings below, the ANZ did not explain how the level of the fee had been arrived at. The primary judge in the Federal Court of Australia (Gordon J) referred to Andrews v Australia and New Zealand Banking Group Ltd, in which the High Court held that a stipulation prima facie imposes a penalty if it is a collateral stipulation which, upon failure of a primary stipulation, imposes upon one party an additional detriment to the benefit of another party. Her Honour also had regard to the "tests" stated by Lord Dunedin in Dunlop Pneumatic Tyre Co Ltd v New Garage and Motor Co Ltd. Her Honour reasoned that the Late Payment Fees were prima facie a penalty. Given the ANZ's admission that the Late Payment Fees were not genuine pre‑estimates of its damage, the question which remained was "to what extent (if any) did the amount stipulated to be paid exceed the quantum of the relevant loss or damage which can be proved to have been sustained by the breach, or the failure of the primary stipulation, upon which the [collateral] stipulation [for the Late Payment Fee] was conditioned". Her Honour held that the stipulation for the Late Payment Fee was to be viewed as "security for, or in terrorem of, the satisfaction of the primary stipulation" and that "each of [the sums charged] is extravagant and unconscionable." In reaching these conclusions, and consistently with the question which had been identified as relevant, the primary judge accepted and applied evidence, adduced by the appellants, which assumed that the only damage that the ANZ could be said to have suffered as a result of the late payments was direct costs associated with the recovery of the minimum payment outstanding. Her Honour rejected other evidence, given for the ANZ, which calculated the costs to it more widely and by reference to certain of its financial interests which, it was said, were adversely affected by the late payments. On appeal, the Full Court of the Federal Court (Allsop CJ, Besanko and Middleton JJ) held that this evidence should have been taken into account and that it showed that the fees were not of the nature of penalties, having regard to the legitimate interests of the ANZ in the performance of the terms for payment. The Full Court allowed the appeal from her Honour's decision. What is a penalty? In Dunlop, Lord Dunedin described the "essence" of a penalty as "a payment of money stipulated as in terrorem of the offending party". By way of comparison, the essence of liquidated damages is "a genuine covenanted pre‑estimate of damage" by the parties. Lord Dunedin's speech in Dunlop has been described as containing a "potpourri of old learning and new" and in the former respect to reflect "centuries of equity jurisprudence". His Lordship's description of the essence of a penalty would fall into this category. The contrasting concept of liquidated damages for breach of contract belongs to a later period. It has been suggested that the reference to a penalty terrorising persons may not be especially helpful, for penalties may be readily agreed to "by parties who are not in the least terrorised by the prospect of having to pay them and yet are … entitled to claim the protection of the court". The Late Payment Fee charged by the ANZ would not appear to have caused Mr Paciocco undue concern, as he would regularly pay the minimum Monthly Payment late and incur the fee, of which he was fully aware. However, the point to be made is that threats and punishment were regarded as the essential characteristics of a penalty. A sum stipulated to be paid on default, which amounted to a threat to the person obliged to pay it if the principal obligation was not performed, bore the character of a penalty, as did a sum stipulated to be paid which could not be accounted for other than as a punishment for default. The distinction drawn in Andrews, between the primary stipulation and the penalty which is collateral to it, directs attention to penal bonds, which were largely used historically to bind persons to the performance of an obligation. Professor Simpson gives the example of a simple common money bond, where A loans B £100. B would execute a bond for a larger sum, which was normally twice the sum lent, thus binding himself to pay £200 on a fixed day. The bond would be subject to a condition of defeasance, which provides that if B pays £100 before the due date, the bond will be void. A similar method was employed for conveying property. The penal bond with conditional defeasance was the principal device for framing substantial contracts in the later medieval and early modern periods. It was adaptable to different transactions and provided certainty. It was the bond that created the debt; it did not just evidence the debt. Thus, it allowed for an action in debt to be brought upon the bond, rather than upon the covenant or agreement it secured. There were limited defences which could be raised in the action (namely, that the condition had been performed, the condition had been substantially performed or the condition was impossible to perform). But penal bonds could operate harshly because of the amount usually required to be paid on default and because any act of default meant the monies were payable. Nevertheless, the law enforced penal bonds strictly, because it regarded their function as compensatory. A creditor could legitimately contract for compensation for loss suffered through the debtor's failure to pay on time. It was on this basis that the law distinguished between such transactions and transactions containing usurious terms (which were payment for the use of money and therefore illegal). It was also considered that a debtor could prevent paying a penalty by paying promptly. Equity also viewed the purpose of penal bonds as compensatory and this was the basis for its intervention. Equity looked to what condition the bond was security for and allowed the obligee compensation for the loss flowing from failure of the condition (usually limited to principal, interest and costs). The purpose of a bond was only to secure the interest of the obligee in the promise or undertaking to be performed. Where compensation was possible for default, the exaction of a penalty was deemed inequitable. The aim of the equity courts was to compensate in the event of default, not to punish. It follows that they would not tolerate individuals exacting punishment. This early understanding of what constituted a penalty finds expression today in the definition given by Mason and Deane JJ in Legione v Hateley: "A penalty, as its name suggests, is in the nature of a punishment for non-observance of a contractual stipulation; it consists of the imposition of an additional or different liability upon breach of the contractual stipulation". This definition was referred to with approval in Andrews and, more recently, by the United Kingdom Supreme Court in Cavendish Square Holding BV v Makdessi (albeit in a more qualified sense), where arguments that the penalty doctrine should be abolished or restricted were rejected. As Lord Neuberger of Abbotsbury and Lord Sumption observed, the innocent party may have interests in the enforcement of the primary obligation but can have no proper interest in simply punishing the defaulter. The consequence of compensation forming the basis of equitable intervention was that where compensation was not possible, or damages could not be assessed, relief could not be given by equity. Compensation might not be possible because the condition on which the bond was made was in respect of an interest not measurable in damages. As explained in Andrews, it is the availability of compensation which generated the "equity" upon which the court intervened; without it, the parties were left to their legal rights and obligations. The primary factor in the decline of the conditional penal bond and the rise of the modern law of penalties has been said to be the practice of the Court of Chancery in relieving against forfeiture. By the time cases such as Dunlop came to be decided, the conditional penal bond may not have been much in use, although it was not wholly obsolete when Professor Simpson was writing and is not today. Examples referred to in Andrews are irrevocable letters of credit and "performance bonds" which are used in the construction industry. While Dunlop does not contain any such discussion of the origins and purposes of the penalty doctrine (as canvassed above), much of what is said in Dunlop is better understood by reference to them. Relevant aspects of Dunlop The aspect of Dunlop which assumes particular importance in this case is the recognition that a sum stipulated for payment on default may be intended to protect an interest that is different from, and greater than, an interest in compensation for loss caused directly by the breach of contract. This is most evident from the speech of Lord Atkinson. It has already been observed that equity recognised that there may be injury to interests for which compensation cannot be made and to which the doctrine of penalties cannot be applied to provide relief. That will usually be because of the nature of the interest protected by the provision for payment on default. In Ringrow Pty Ltd v BP Australia Pty Ltd, it was said that Dunlop continues to express the law to be applied with respect to penalties in Australia. As the primary judge in these proceedings observed, the principles in Dunlop were not affected by the decision of this Court in Andrews. But this does not mean that those principles are confined, or that they are limited, to the "tests" propounded by Lord Dunedin, or that what was said in Dunlop does not require further explication. In Ringrow the Court was concerned with an argument which focused upon Lord Dunedin's speech in Dunlop and the "tests" which were offered to assist in the determination of whether a sum stipulated to be paid on default is, or is not, a penalty. In comparison, in Andrews reference was made to Dunlop, not to Lord Dunedin's "tests", but rather to Lord Atkinson's identification of the interests which were sought to be protected by the provision stipulating for payment of monies on breach and which accounted for that provision not being a penalty. It was said that "the critical issue, determined in favour of the appellant [Dunlop], was whether the sum agreed was commensurate with the interest protected by the bargain." The fact that the decision in Dunlop itself, and Lord Atkinson's reasons with respect to it, assume importance in this case does not deny the significance of the requirement stated by Lord Dunedin, that the sum stipulated be "extravagant and unconscionable" before it can be characterised as a penalty. As explained below, it is these words that, by their extreme nature, identify the penal character of a penalty. The question which may be identified as arising from this aspect of the decision in Dunlop, which is appropriate to a case of this kind, is whether a provision for the payment of a sum of money on default is out of all proportion to the interests of the party which it is the purpose of the provision to protect. This interest may be of a business or financial nature. The Dunlop "tests" The distinction drawn by Lord Dunedin between liquidated damages and a penalty, whilst useful, should not be understood as a limiting rule. It does not mean that if no pre-estimate is made at the time a contract is entered into, as is the case here, a sum stipulated will be a penalty. Nor does it mean that a sum reflecting, or attempting to reflect, other kinds of loss or damage to a party's interests beyond those directly caused by the breach will be a penalty. Indeed the provision in Dunlop, which was held not to be a penalty, was of this kind. The question whether a sum to be paid on default is a penalty, as distinct from liquidated damages, was said by Lord Dunedin to be a question of construction, but his Lordship is not to be taken to suggest that it will be answered by the language of the contract alone. This is evident from the reference to the "inherent circumstances" of the contract, which includes the position of the party whose interests are to be protected by the stipulation for the payment of the sum on default. Lord Dunedin offered four "tests" to assist "this task of construction". They were couched in the language of their time and were intended as guidance only. Tests tend, over time, to encourage literal application. Especially is this so where 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 payment on 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. This latter prohibition has found expression in modern times, as is evident from the passage from Legione v Hateley referred to above and also from judgments in Cavendish. It may be inferred from this policy that a sum stipulated for payment on default is a penalty if it bears no relation to the possible damage to or interest of the innocent party. The first, and principal, "test" stated by Lord Dunedin is that a sum stipulated will be a penalty if it is: "extravagant and unconscionable in amount in comparison with the greatest loss that could conceivably be proved to have followed from the breach." If the "test" is understood to convey that only loss in the nature of damages directly flowing from the breach is to be considered, then it is unduly restrictive, though no doubt it remains useful to many cases. The terms "extravagant" and "unconscionable" (and also "exorbitant") had been used in Clydebank Engineering and Shipbuilding Co v Don Jose Ramos Yzquierdo y Castaneda, where the Earl of Halsbury LC said that the jurisdiction given to the court to interfere in an agreement between parties was with respect to an agreement which was "unconscionable and extravagant, and one which no Court ought to allow to be enforced." Even earlier, the Scottish Court of Session in Forrest and Barr v Henderson, Coulborn, and Co had said that "equity will interfere to prevent the claim being maintained to an exorbitant and unconscionable amount." As explained below, "extravagant", "exorbitant" and "unconscionable" are "strong words"; despite the different expressions used, they all describe the plainly excessive nature of the stipulation in comparison with the interest sought to be protected by that stipulation. The second "test" was said to be merely a corollary of the first, and concerns the case where the breach is constituted by a mere failure to pay a sum of money. The sum stipulated to be paid in the event of a breach will be a penalty if it is greater than the sum which ought to be paid. This reflects equity's concerns about penal bonds and its view that the tender of principal together with interest thereon is sufficient compensation. This "test" has a narrow range of operation and is confined to the simplest of cases. It does not take into account that damages for breach may now include interest by way of damages and opportunity costs. It says nothing about the damage to a party's wider commercial interests, for example to its trading, which was the real issue in Dunlop. And it says nothing about the financial effects for which the ANZ contends. The third "test" is stated as a presumption ("(but no more)") that a sum will be a penalty where it is a single sum made payable on the occurrence of one or more of several events, some of which may occasion serious, and others only inconsequential, damage. The presumption derives from what was said by Lord Watson in Lord Elphinstone v Monkland Iron and Coal Co: "When a single [lump] sum is made payable by way of compensation, on the occurrence of one or more or all of several events, some of which may occasion serious and others but trifling damage, the presumption is that the parties intended the sum to be penal, and subject to modification." However, the provision for payment in that case was not in fact of that kind. It was referable to a single obligation and the sum to be paid bore "a strict proportion to the extent to which that obligation is left unfulfilled." In Ringrow, this Court said that this reasoning "did not require there to be a strict proportion; it merely relied, as a step towards the conclusion that the compensation was not inordinate or extravagant, on the fact that the compensation bore a strict proportion to the unfulfilled obligation." Further, because a provision of the kind mentioned is merely a presumption, it may be rebutted. In Dunlop, Lord Atkinson observed that it was there rebutted by the fact that the damage caused by default may be of such an uncertain nature that it cannot be accurately ascertained. The last "test" stated by Lord Dunedin in Dunlop refers to just such a circumstance. It identifies the case where the parties agree a figure although a forecast of loss, in reality, is almost impossible. Nevertheless, the sum agreed may not be a penalty, indeed it is likely that in circumstances such as these it is not. His Lordship said: "It is no obstacle to the sum stipulated being a genuine pre-estimate of damage, that the consequences of the breach are such as to make precise pre-estimation almost an impossibility. On the contrary, that is just the situation when it is probable that pre-estimated damage was the true bargain between the parties". A similar observation had been made in Clydebank. What Lord Dunedin was pointing to is damage of a kind which is different from that for which liquidated damages could be assessed. It will be different because the interests of the party which are intended to be protected by the provision in question extend beyond an interest in the recovery of compensation for loss caused by the obligation. This was the situation in Dunlop. Interests: Clydebank, Dunlop and Cavendish The agreement in Dunlop was headed "Price Maintenance Agreement" and contained provisions for resale price maintenance, which was clearly not then a prohibited practice. It bound the respondents, as dealers in goods manufactured by Dunlop, inter alia, not to sell or offer the goods at less than Dunlop's list price and to pay £5 for each item sold at less than that price. It followed from these terms that the sale of even one tyre, cover or tube at less than the listed price would attract the sum stipulated to be paid. An argument, reminiscent of one raised in this case concerning the Late Payment Fee, that Dunlop could not possibly lose that sum on the occasion of each sale, was rejected. The argument, Lord Atkinson observed, missed the point about the purpose of the agreement and the nature of the possible injury to Dunlop's trade. Dunlop's object in making the agreement, Lord Atkinson said, was to prevent disorganisation of its trading system. His Lordship said: "[Dunlop] had an obvious interest to prevent this undercutting, and on the evidence it would appear to me impossible to say that that interest was incommensurate with the sum agreed to be paid." In Clydebank, having observed that agreements for the payment of sums on default operate as "instruments of restraint", Lord Robertson identified, similarly to Lord Atkinson in Dunlop, the relevant question as: "Had the respondents no interest to protect by that clause, or was that interest palpably incommensurate with the sums agreed on?" In Clydebank, the contract between the appellant shipbuilders and the Spanish government for the building of torpedo boats contained a clause providing for a "penalty for later delivery … at the rate of £500 per week for each vessel". The fact that the sum was called a penalty was not, of course, conclusive. In holding that the stipulated sum was not a penalty, it was acknowledged that the interests of the Spanish government in having the vessels delivered on time were complex and that how those interests would sound in damages was extremely difficult to prove. In Cavendish, non-competition provisions, in an agreement for the sale of a controlling interest in a business, which had the effect that, upon breach, the seller would not be entitled to any further payments of the purchase price were held not to be penalties. The provisions were seen as protective of the interests of the purchaser in the goodwill of the business, such goodwill being critical to the value of the business. It was of some importance in Dunlop and Clydebank that the nature of the innocent party's interests, which would be injured by breach, was such that it would be difficult to estimate and to prove damage. This difficulty of proof, and the uncertainty of the loss which could arise, made it reasonable for the parties to agree beforehand what the figure for damages should be in order to avoid the problem. In Cavendish it was observed that there is good reason to leave the assessment of the value of a complex interest as a matter of negotiation between the parties, especially since the court may not be in a position to value the interest itself. For present purposes it is perhaps more relevant to observe that difficulties of this kind may render problematic proof that a sum stipulated is a penalty. It was not suggested in either Clydebank or Dunlop that the damage to the Spanish government's or to Dunlop's interests was impossible to estimate; rather, it appears that the damage was capable of estimation, albeit with little precision. It might be thought that the damage to the interests identified in Clydebank in particular might have qualified as impossible to prove, but the Earl of Halsbury LC went only so far as to say that it would be "extremely complex, difficult, and expensive" to do so. And in Dunlop, the estimation was referred to as "almost an impossibility". It will be recalled that equity's jurisdiction was considered not to extend to a case when compensation was not thought to be possible, as is the case when damages could not be assessed. In these circumstances the parties would be left to their bargain. What was said in Dunlop, and in Clydebank, about it being reasonable, in cases of difficulty in the estimation of possible loss, to leave the parties to contract for themselves for a sum to be paid on default might be thought to come close to an acceptance that they be left to their bargain. However, this would overlook the fact that the courts in those cases went on to determine whether the figure arrived at was a penalty and that they did so by considering whether it was unconscionable or extravagant in amount. A sum out of all proportion to the interests protected Lord Dunedin said in Dunlop that there may be no reason to suspect that the figure agreed by the parties, in the case where loss is difficult to estimate, is "a penalty to be held in terrorem", "provided that figure is not extravagant". Lord Atkinson and Lord Parmoor also held that the figure in question was not extravagant, unconscionable or extortionate. The process to be undertaken in order to determine whether an amount is unconscionable or extravagant was not further explained in Dunlop and Clydebank. The figure agreed to be paid cannot be compared with a sum certain, as is the case with Lord Dunedin's first "test". It can only be gauged against the identified interests of the party in whose favour the stipulation is made. It may be inferred from Dunlop and Clydebank that the interests in question were regarded as substantial and the possibility of damage to them real. The sum agreed to be paid in those cases was not incommensurate with the relevant interests. In Clydebank, the Earl of Halsbury LC did not consider that a rule could be laid down as to when a stipulation could be said to be extravagant or unconscionable and that much would depend upon the circumstances of each case. However, it is to be inferred from the adjectives chosen that not every sum in excess of what might be strictly compensatory will amount to a penalty. This is confirmed by the example, admittedly extreme, which his Lordship then gave of an agreement to build a house for £50 but "to pay a million of money as a penalty" if the house was not built. This suggests that a person contending that a sum is a penalty will be facing a high hurdle. Lord Hodge was later to observe in Cavendish that the criterion of exorbitant or unconscionable should prevent the enforcement of only egregious contractual provisions. In Ringrow, it was held that a sum which was merely disproportionate to the loss suffered would not qualify as penal. It was explained that exceptions from freedom of contract "require good reason to attract judicial intervention to set aside the bargains upon which parties of full capacity have agreed", which is why the law on penalties is expressed as an exceptional rule and in exceptional language. The Court went on: "It explains why the propounded penalty must be judged 'extravagant and unconscionable in amount'. It is not enough that it should be lacking in proportion. It must be 'out of all proportion'." In Cavendish, Lord Neuberger and Lord Sumption said that the true test is whether the provision is a secondary obligation which imposes a detriment on the party in breach "out of all proportion to any legitimate interest of the innocent party in the enforcement of the primary obligation." Australian and United Kingdom law are not alone in maintaining a standard to be applied to a requirement to pay money, or some other detriment, which is imposed in the event of default. In many other western legal systems something like the penalty doctrine exists. In Andrews, reference was made to s 343 of the German Civil Code, which provides that a "disproportionately high" penalty may be reduced by a court after taking into account "every legitimate interest" of the party for whose benefit the stipulated sum is made. Such interests are not limited to that party's economic interests. In Cavendish, Lord Hodge referred to provisions in other modern civil codes and international instruments which use tests such as whether the sum stipulated is "manifestly excessive" or "substantially disproportionate" in order to modify or restrict contractual penalties. It has earlier been observed that the nature of an interest and of the injury to it may make for difficulties of proof that the sum stipulated is a penalty. In Clydebank, Lord Robertson acknowledged that the problem was not one for the Spanish government: "But, in truth, the only apparent difficulty in the present case arises from the magnitude and complexity of the interests involved and of the vicissitudes affecting them, and as the question is whether this stipulation of £500 a week is unconscionable or exorbitant, these considerations can hardly be considered a formidable difficulty in the way of the respondents." The Late Payment Fee: a penalty? The ANZ's interests in this case are not as diffuse as those considered in Dunlop, Clydebank and Cavendish. The ANZ did not suggest that the injury to its interests was not capable of some kind of estimation in money's worth. In the hearing before the primary judge it abandoned the claim, made in its defence, that the costs occasioned to it by late payments were impossible to calculate and argued instead that they were very difficult to calculate. On this appeal the appellants accepted that, being realistic, the law should allow a "measure of latitude" where pre-estimation of loss is difficult. Certainly there needs to be some recognition of the difficulties attending any such exercise and that there may, in some cases, be differences in approach to the proper methodology to be employed. But it also needs to be borne in mind that this task is not one which calls for precision. The conclusion to be reached, after all, is whether the sum is "out of all proportion" to the interests said to be damaged in the event of default. It is important at this point to identify the ANZ's interests. The ANZ had an interest in receiving timeous repayment of the credit that it extended to its customers, including the appellants. As explained below, late payment impacted the ANZ's interests in three relevant respects: through operational costs, loss provisioning and increases in regulatory capital costs. Evidence of the costs to the ANZ by reason of the late payments was given by Mr Regan for the appellants and by Mr Inglis for the ANZ. Their approaches were fundamentally different because of the instructions they had been given. As the primary judge observed, Mr Regan was instructed to identify the amounts necessary to restore the ANZ to the position it would have been in had the late payments not been made. In contrast, Mr Inglis was instructed to consider the maximum amount of costs that the ANZ could conceivably have incurred as a result of a late payment. As a consequence Mr Regan calculated only the costs to the ANZ of ensuring that the late payments were made ("operational costs"), such as those costs incurred by the use of staff contacting Mr Paciocco and other administration costs. Mr Inglis calculated those costs and came to a higher figure than Mr Regan, but he also calculated other impacts on, or costs to, the ANZ's financial interests which were referred to in the proceedings below as "Increase in loss provisions" and "Increase in the cost of regulatory capital". As to the first category of costs, as the primary judge explained, the ANZ is required to estimate the impairments to its financial assets in order that its financial statements reflect a fair value of what is likely to be collected from what is outstanding. It is required to make provision in its accounts for what it may not recover, albeit that the potential loss is expressed as a current cost. The primary judge does not appear to have cavilled with the opinion of Mr Inglis - that the reduction in the value of a customer's loan, as recorded in the accounts, was an accepted category of loss. However, her Honour held that the difficulty was that a provision of this kind is merely an accounting entry. At the time it is made it cannot be known whether all the cardholders recorded will default. In the case of Mr Paciocco he did not fall into this category because in fact he did not default; he merely paid late. This reasoning is consistent with the primary judge's overall approach, which was to limit the ANZ's "costs" to actual damage incurred. However, this overlooks that the estimation is to be made at the time the Late Payment Fee is agreed upon; and it does not acknowledge that an effect upon the ANZ's interests may include the provision that it has to make concerning its overall position. As to the second category of costs, the ANZ is also required to hold regulatory capital to cover unexpected losses, a buffer of a kind. An increase in the risk of default increases the amount of regulatory capital which is required to be held. The primary judge accepted that regulatory capital has a cost to the ANZ, by way of the loss of additional return it could otherwise make on the amount held as regulatory capital. But her Honour did not accept that it should be taken into account in calculating loss or damage as a result of late payment. It was her Honour's view that loss provisions and regulatory capital costs are part of the costs of running a bank in Australia. Banks may, and do, seek damages for default, but they are limited to the sums outstanding, enforcement costs and interest. However, as has been explained, the question is not what the ANZ could recover in an action for breach of contract, but rather whether the costs to it and the effects upon its financial interests by default may be taken into account in assessing whether the Late Payment Fees are penalties. The primary judge accepted and applied Mr Regan's evidence. Her Honour considered that the main difficulty with Mr Inglis' evidence was that he did not calculate actual loss or damage, but rather engaged in a broad-ranging exercise of identifying "costs" that might be affected by late payment, in a more theoretical, accounting, sense. In her Honour's view this did not assist in answering the question which she had earlier identified: to what extent (if any) did the amount stipulated to be paid exceed the quantum of the relevant loss or damage which can be proved to have been sustained by the breach. But of course framing the question in this way takes no account of the ANZ's other interests which were said to be addressed by the Late Payment Fees and which extend beyond the recovery of compensation for loss. The primary judge accepted that whilst the actual losses suffered by the ANZ by reason of the late payments could not be precisely determined, they were probably no more than $3.00 for each event of late payment (based on Mr Regan's evidence) and in any event much less than the $20.00 or $35.00 charged as a Late Payment Fee. They were therefore extravagant and unconscionable. Mr Inglis' evidence identified the costs to which the ANZ would be subject in the event of a late payment as a range which exceeded the amounts of the Late Payment Fee. His calculations were criticised as overly generous. It is not necessary to resolve any such controversy. The effect of Mr Inglis' evidence was to identify potential costs to the ANZ, from late payments, which reflect injuries to its financial position. They were real because they had to be taken into account by the ANZ. The evidence called for the appellants did not address damage of this kind. It cannot therefore be concluded that the sums of $20.00 and $35.00 were out of all proportion to the interests so identified. Conclusion and orders It may be accepted that it is difficult to measure the loss to the ANZ as a result of a late payment. Consistently with Clydebank, Dunlop, Ringrow and Andrews, the relevant question in this case is whether the Late Payment Fee is out of all proportion to the ANZ's interest in receiving timeous payment of the minimum Monthly Payment. Applying this test, the appellants did not establish that the Late Payment Fee was a penalty. The appeal should be dismissed with costs. M219/2015 This appeal concerns whether the Late Payment Fee contravenes certain statutory provisions. I agree that this appeal should be dismissed with costs, for the reasons given by Keane J.

GAGELER J. Introduction Two appeals are brought to this Court from orders made by the Full Court of the Federal Court of Australia (Allsop CJ, Besanko and Middleton JJ) on appeal from a judgment of a primary judge of that Court (Gordon J). The judgment of the primary judge was a final determination of claims made by applicants in a representative proceeding constituted under Pt IVA of the Federal Court of Australia Act 1976 (Cth). Their claims were to recover from Australia and New Zealand Banking Group Ltd ("ANZ") certain "exception fees" charged by ANZ under standard terms and conditions of contract with consumer credit card account holders. The appellants in each appeal were applicants in the representative proceeding; they can be called "the customers". The appeals are a sequel to the decision at an interlocutory stage of the representative proceeding in Andrews v Australia and New Zealand Banking Group Ltd. The controversy that now returns to this Court is confined to legal characterisation of a single class of exception fee - contractually designated "late payment fee" - charged to one customer, Mr Paciocco. The ultimate question in the first appeal is whether the contractual stipulation for the late payment fee was unenforceable as a penalty at common law. The ultimate question in the second appeal is whether that stipulation or its enforcement contravened one or more of several applicable statutory norms prohibiting ANZ from engaging in "unconscionable conduct" and from entering into and enforcing contracts which were "unjust" and "unfair". Before turning to subsidiary issues involved in answering those questions, there is utility in recording the contractual provisions under which the late payment fee was imposed, in giving an outline of the evidence bearing on the quantification of that fee, and in noting the reasoning of the primary judge and of the Full Federal Court. The contracts The customers' claims in the representative proceeding related to the period between September 2006 and September 2013. During that period, Mr Paciocco had two credit card accounts with ANZ. Both were "Low Rate MasterCard" accounts. One had been opened in June 2006 with an initial credit limit of $15,000 which was increased to $18,000 in November 2009. The other was opened in July 2009 and had a credit limit of $4,000. The terms and conditions on which ANZ contracted for the provision of each of those credit card accounts were contained in three standard form documents sent to Mr Paciocco by ANZ: a standard form Letter of Offer; a booklet entitled "ANZ Credit Card Conditions of Use"; and another booklet entitled "ANZ Personal Banking Account Fees and Charges". The standard terms and conditions set out in the ANZ Credit Card Conditions of Use provided for ANZ to issue monthly statements of account. Each monthly statement was to show the closing balance of the account, the "minimum monthly payment" and the "due date". The minimum monthly payment was ordinarily to be the greater of $10 or 2% of the closing balance shown on the statement, but to be the full closing balance if the closing balance shown on the statement was less than $10. The due date was to be a date a specified number of days after the end of the monthly period to which the statement related. The account holder was obliged to make the minimum monthly payment shown on each monthly statement by the due date shown on the statement. If the account holder did not pay the closing balance by the due date, the account holder was to be charged interest on each purchase shown on a statement of account, and on all subsequent purchases, until the closing balance was paid in full by its due date. The account holder was not otherwise to be charged interest on purchases (and therefore received credit for purchases interest free if the closing balance of the statement on which the purchase transaction appeared was paid in full by its due date) but was always to be charged interest on cash advances from the date of each cash advance. The annual percentage rate used to calculate the interest charges on the account was to be that notified by ANZ from time to time. Under the heading "Fees and charges", the ANZ Credit Card Conditions of Use provided: "ANZ reserves the right to charge the credit card account with fees and charges for the provision and operation of the credit card account. The fees and charges applicable to the credit card account are those shown in the Letter of Offer and in the ANZ Personal Banking Fees and Charges booklets, as varied from time to time." The late payment fee was one of the fees and charges applicable to each credit card account shown in the standard form Letter of Offer and in the ANZ Personal Banking Account Fees and Charges booklet. ANZ had the right to charge the late payment fee to the account if the minimum monthly payment was not paid by the due date. The amount of the late payment fee was set by ANZ at $35 until December 2009. It was reduced to $20 from December 2009. Under the heading "Default, cancellation and termination", the ANZ Credit Card Conditions of Use provided for the credit card to be cancelled and the outstanding balance of the account to become immediately due and payable, at the option of ANZ, if the account holder came into default. Coming into default included for that purpose the account holder failing to meet any of the account holder's obligations under the credit card contract. The ANZ Credit Card Conditions of Use went on to provide in that context: "Any reasonable amount reasonably incurred or expended by ANZ in exercising its rights in relation to the credit card account arising from any default (including expenses incurred by the use of ANZ's staff and facilities) are enforcement expenses and become immediately payable by the account holder. ANZ may debit the credit card account for such amounts without notice." Finally, the ANZ Credit Card Conditions of Use provided for the account holder to close the credit card account at any time by giving notice to ANZ, and for ANZ to change any term or condition of the credit card contract by giving notice to the account holder. The charges During the period to which the customers' claims related, ANZ charged the late payment fee to Mr Paciocco's accounts on 26 occasions and Mr Paciocco subsequently paid the amounts charged to ANZ. Eight of those occasions were before December 2009, when the applicable charge was $35. Eighteen were after December 2009, when the applicable charge was $20. For the purpose of illustrating their respective arguments in the appeals, the parties chose to focus on six of those 26 occasions. The sample comprised two charges of $35 made to the credit card account having the higher credit limit before December 2009, a single charge of $20 made to that same account after December 2009, and three charges of $20 made after December 2009 to the credit card account having the lower credit limit. The two illustrative charges of $35 were made to the credit card account which had the higher credit limit at a time when that credit limit was $15,000. At the time of the first charge, the outstanding monthly balance was $10,199 and the minimum monthly payment was $203. At the time of the second charge, the outstanding monthly balance was $11,220 and the minimum monthly payment was $223. The single illustrative charge of $20 made to that same account was made when the credit limit had risen to $18,000. The outstanding monthly balance was then $18,025 and the minimum monthly payment was $358. Of the three charges of $20 made to the credit card account which had the lower credit limit of $4,000, the first was made at a time when the outstanding monthly balance was $2,145 and the minimum monthly payment was $43. The second was made at a time when the outstanding monthly balance was $268 and the minimum monthly payment was $10. The third was made at a time when the outstanding monthly balance was $4,055 and the minimum monthly payment was $80. The primary judge found that ANZ's Collections Business Unit contacted Mr Paciocco on most of the occasions when he failed to make a minimum monthly payment by its due date and that Mr Paciocco promised to make a payment each time he was contacted. The primary judge also found that, throughout the period to which the claim in the representative proceeding related, Mr Paciocco was aware of the late payment fee and other fees and charges applicable to his credit card accounts and found it convenient to manage his credit card accounts close to their limits, choosing to accept the risk of incurring fees associated with that course of conduct. The evidence about the amount of the late payment fee ANZ made a formal admission on the pleadings that it did not determine the quantum of the late payment fee by reference to a sum that would have been recoverable as unliquidated damages. Evidence in the representative proceeding showed that fees broadly equivalent to the late payment fee were charged in varying amounts by other banks which provided credit cards to consumer customers in competition with ANZ during the period of the claims made in the proceeding. The reduction of the late payment fee from $35 to $20 in December 2009 occurred not long after reductions by other banks. Documentary evidence was adduced concerning the considerations which led ANZ to set and maintain the amount of the late payment fee at $35 and then to reduce it to $20, but that evidence was described by the primary judge as "incomplete" and "of limited use". The primary judge drew no conclusions from it. The main evidence bearing on the amount of the late payment fee was opinion evidence given by two accountants: Mr Regan and Mr Inglis. They were asked to, and did, perform two quite different tasks. Mr Regan was asked by the customers to calculate, in respect of the late payment events which gave rise to the charge of the late payment fee to Mr Paciocco, "how much money it would take to restore ANZ to the position it would have been in if the particular event giving rise to the entitlement to charge such fees had not occurred". Performing that task, Mr Regan calculated the variable or incremental operational costs incurred within ANZ's Collections Business Unit in making or attempting to make contact with Mr Paciocco following each of the 26 late payment events which were the subject of his claim. Mr Regan calculated those costs to have ranged from $5.50 to 50c, and to have been on average $2.60. Mr Inglis was asked by ANZ to assess, in respect of all of the consumer credit card accounts offered by ANZ in each financial year to which the customers' claims related, "the costs that may have been incurred by ANZ in connection with the occurrence of events that gave rise to an entitlement to charge [the late payment fee]". To perform that task, Mr Inglis undertook an assessment of costs incurred by ANZ in connection with the occurrence of events that gave rise to an entitlement to charge the late payment fee to all of the holders of consumer credit card accounts who had been charged during the financial year ended September 2009. He then extrapolated those results to other years. During the financial year ended September 2009, ANZ had around two million consumer credit card accounts. It charged the late payment fee on around 2.4 million occasions. Its revenue from charging the late payment fee was around $75 million. Mr Inglis identified three categories of costs as having been incurred by ANZ in connection with the occurrence of the events that gave rise to an entitlement to charge the late payment fee. Those categories comprised provisioning costs and regulatory capital costs, in addition to operational costs which were principally costs associated with the activities of ANZ's Collections Business Unit. Each of those categories warrants a short explanation. Provisioning costs were expenses which, in accordance with applicable accounting standards, ANZ recognised in its profit and loss account representing reductions in the value of customer accounts attributable to risk of default. Because the probability of default increased with late payment, late payment of balances contributed to the overall level of the expense required to be recognised. Mr Inglis assessed the average contribution of a late payment event to provisioning costs during the period relevant to ANZ's charge of the late payment fee to Mr Paciocco's accounts at $23 for the account with the higher credit limit or $27 for the account with the lower credit limit. Regulatory capital costs were costs which ANZ incurred in funding capital which ANZ was required by applicable prudential standards to hold as a buffer against unexpected losses. Because the amount of capital required to be held increased with the probability of default associated with late payment, late payment of balances contributed to the overall level of capital required, and with it the costs of funding that capital reserve. Mr Inglis assessed the average contribution of a late payment event to regulatory capital costs during the period relevant to ANZ's charge of the late payment fee to Mr Paciocco's accounts at $23 for the account with the higher credit limit or $5 for the account with the lower credit limit. The operational costs identified by Mr Inglis to be associated with ANZ's Collections Business Unit were costs attributable to the same collection activities as those identified by Mr Regan. However, unlike Mr Regan, who looked only to the variable or incremental costs of individual collections, Mr Inglis made an allowance for the recovery of a proportion of common costs and of fixed costs associated with overall collection activities. Mr Inglis assessed the average collection costs attributable to a late payment event during the period relevant to ANZ's charge of the late payment fee to Mr Paciocco's accounts as in excess of $5. For the period during which a late payment fee of either $35 or $20 was charged to Mr Paciocco's accounts, Mr Inglis thereby assessed the average costs incurred by ANZ in connection with the occurrence of an event giving rise to an entitlement to charge the late payment fee as in excess of $50 for the account with the higher credit limit and in excess of $35 for the account with the lower credit limit. The approach of the primary judge The primary judge drew particular attention to two features of the late payment fee. One was that "the breach (or failure of the stipulation)" which gave rise to the obligation to pay the fee "consisted only in not paying a sum of money". The other was that "[t]he same fee was payable regardless of whether the customer was 1 day or 1 week late (or longer), and regardless of whether the amount overdue was $0.01 (trifling), $100, $1000 or even some larger amount". The second of those features was described as sufficient to give rise to a "presumption" that the late payment fee had the character of a penalty. That "presumption" having arisen, the primary judge saw as the determinative question "to what extent (if any) did the amount stipulated to be paid exceed the quantum of the relevant loss or damage which can be proved to have been sustained by the breach, or the failure of the primary stipulation, upon which the stipulation was conditioned". Mr Regan's evidence was seen by the primary judge to have addressed that question; Mr Inglis' evidence, concluded the primary judge, did not. Addressing the additional categories of costs which Mr Inglis had identified and which Mr Regan had not, the primary judge in any event considered that: a provisioning cost might have been "a 'cost' in an accounting sense" but it did "not represent a loss or damage incurred as a result of [a late payment] event"; "provisions and regulatory capital [were] part of the costs of running a bank in Australia", and "[n]o increase in them [could] be directly or indirectly related to any of the late payments by Mr Paciocco". The primary judge assessed the loss or damage which ANZ sustained by each of Mr Paciocco's failures to pay the balance owing by the due date at "no more than $3", that figure being "comprised of the operational (collections) costs identified by Mr Regan rounded up to the nearest half dollar". When compared with that assessed loss or damage, each late payment fee of either $35 or $20 was "extravagant and unconscionable", and for that reason each late payment fee was unenforceable as a penalty. Having concluded that the customers had succeeded in establishing that the stipulation for the late payment fee was unenforceable at common law or in equity, the primary judge did not go on to consider whether the contractual imposition of the late payment fee contravened any of the statutory norms on which the customers relied in the alternative. The primary judge's orders included a declaration that the late payment fee charged to Mr Paciocco's accounts on each of the 26 occasions "constituted a penalty at common law and a penalty in equity", a declaration that Mr Paciocco was entitled to the amount by which the late payment fee exceeded the costs of ANZ by reason of the events which gave rise to the charging of the late payment fee, and judgment for Mr Paciocco in an amount so calculated together with statutory interest. The approach of the Full Court ANZ and the customers each separately appealed to the Full Court. The Full Court dismissed the appeal of the customers, but allowed the appeal of ANZ and set aside each of the orders made by the primary judge. Allsop CJ gave the principal reasons for judgment. His Honour took the view that the primary judge had "undertaken an ex post inquiry of actual damage as a step in assessing whether the prima facie penal character of the late payment fee was rebutted" and "impermissibly narrowed the content of the notion of genuine pre-estimate of damage as a reflex of penalty". The correct approach, his Honour said, was "to look at the greatest possible loss on a forward looking basis" and to assess that loss by reference to "the economic interests to be protected". Allsop CJ concluded that Mr Inglis' perspective had been correct, and that each of the categories of costs which Mr Inglis had identified and quantified on an average basis was a legitimate object of commercial interest for protection. Having regard to the totality of the circumstances - which his Honour identified as including the interests of ANZ protected and the nature of the relationship between ANZ and a customer under which the customer could terminate a credit card account at will and choose whether and when to engage in transactions on the account - Allsop CJ concluded that "the fees were not demonstrated to be extravagant, exorbitant or unconscionable". From that conclusion, it followed that the late payment fee was not shown to be a penalty. Turning to the customers' argument that ANZ contravened applicable statutory norms by contractually imposing the late payment fee, Allsop CJ observed that a conclusion that ANZ acted unconscionably in relation to Mr Paciocco would be "difficult" in light of the facts that Mr Paciocco was aware of the fees and charges applicable to his credit card accounts, that it was convenient for him to manage his credit card accounts close to their limits, and that he chose to accept the risk of incurring fees associated with so doing. After an extensive consideration of principle, his Honour concluded that ANZ did not engage in unconscionable conduct and that the credit card contracts were not unfair or unjust. Besanko and Middleton JJ agreed with the reasoning of Allsop CJ in relation to both appeals. Choosing to address an argument which Allsop CJ and Middleton J did not find it necessary to consider, Besanko J added in relation to the appeal by ANZ that he considered the claim to recover the earliest of the amounts paid by Mr Paciocco not to be statute barred. The appeals to this Court The explanation for there being two appeals to this Court is that they are brought by the customers from the two distinct sets of orders of the Full Court, allowing the appeal to that Court by ANZ and dismissing the appeal to that Court by the customers. In the first of the appeals to this Court, in which the ultimate issue is whether the late payment fee was unenforceable as a penalty at common law, the parties agree that the governing principles are to be found in Andrews and in Dunlop Pneumatic Tyre Company Ltd v New Garage and Motor Company Ltd. The heart of the dispute between them concerns the nature and content of the propositions of law for which Dunlop is to be treated as continuing authority in Australia. Their competing contentions reflect the competing approaches in the Federal Court. The customers support the approach of the primary judge. ANZ supports the approach of the Full Court. That dispute about Dunlop cannot adequately be addressed without some explanation of the significance and effect of the holding in Andrews. In the first appeal, ANZ seeks by notice of contention to re-agitate the argument, not considered by Allsop CJ or Middleton J but resolved against ANZ by Besanko J, that the claim to recover the earliest of the amounts paid by Mr Paciocco following charging of the late payment fee is statute barred. There will be no need to consider that argument. In the second of the appeals to this Court, in which the ultimate issue is whether the imposition of the late payment fee contravened one or more applicable statutory norms, the customers make a number of specific arguments which they say were lost in the compendious manner in which their statutory claims were dealt with in the Full Court. Those specific arguments are best left to be noted and addressed after setting out the analysis necessary to dispose of the first appeal. The holding in Andrews Andrews explained the term "penalty" to refer to punishment, consisting of the imposition of an additional or different contractual liability, for non-observance of a "primary" contractual stipulation. Thus, a penalty is a "collateral" stipulation "in the nature of a security for and in terrorem of the satisfaction of the primary stipulation". The precise holding in Andrews was reflected in the formal declaration made in that case. The holding was that there is no reason in principle why the primary stipulation to which a penalty is collateral cannot consist of the occurrence or non-occurrence of an event which is neither a breach of contract nor another event which it is the responsibility or obligation of the party subjected to the penalty to avoid. The question determined by that holding had quite recently been considered in and determined to the contrary in a decision of an intermediate court of appeal in Australia, which was overruled. The question had not previously squarely arisen for the determination of this Court. The Supreme Court of the United Kingdom subsequently held in Cavendish Square Holding BV v Makdessi that only a detriment imposed on breach of contract can amount to a penalty. The common law of the United Kingdom has to that extent now taken a different path from the common law of Australia. This is not the occasion to critique the relative merits of the divergent contemporary approaches or to debate the competing perspectives on our common legal history which contributed to that divergence. There is need to respond only to the statement in Cavendish that Andrews involved "a radical departure from the previous understanding of the law". To the extent that the statement refers to the common law of Australia, the statement is wrong and appears to be based on a misunderstanding of Andrews. Andrews did nothing to disturb the settled understanding in Australia that a contractual provision imposing a penalty is unenforceable at common law without the discretionary intervention of equity. That proposition was accepted by at least three members of this Court in AMEV-UDC Finance Ltd v Austin and was the proposition on which the decision of the majority in that case pivoted. Andrews rejected the understanding that AMEV-UDC stood for the further and more sweeping proposition that "[t]he modern rule against penalties is a rule of law, not equity". Equitable jurisdiction to relieve against penalties, Andrews emphasised, has not been abolished. The statement of Mason and Wilson JJ in AMEV-UDC that "the equitable jurisdiction to relieve against penalties withered on the vine for the simple reason that, except perhaps in very unusual circumstances, it offered no prospect of relief which was not ordinarily available in proceedings to recover a stipulated sum or, alternatively, damages" was criticised, but only insofar as the statement might be taken to have drawn a causal link between the withering of the equitable jurisdiction to relieve against penalties and the advent of the Judicature system. To fuse the administration of law and equity is not thereby to destroy an equitable doctrine, and for an equitable doctrine to wither is not necessarily for an equitable doctrine to die. Nothing in Andrews contradicts the fuller explanation given by Deane J in AMEV-UDC that "acceptance by the common law of the unenforceability of penalties largely removed the occasion for the exercise of the equitable jurisdiction to relieve against enforcement with the result that the terms upon which equity would grant such relief became ordinarily of but academic or historical interest". His Honour's explanation, with which Andrews is consistent, continued by pointing out that "[t]he equitable jurisdiction did not, however, cease to exist and the terms upon which equitable relief against penalties would be granted remain directly applicable in those comparatively rare cases in which the party asserting unenforceability is constrained to seek positive relief (whether primary or ancillary) which is purely equitable in character, such as an order for reconveyance". The statement in Andrews that "[i]t is the availability of compensation which generates the 'equity' upon which the court intervenes" without which "the parties are left to their legal rights and obligations" is, in context, a reference to the historically important, although now comparatively rare, exercise of equitable jurisdiction to grant relief against penalties. The statements that, "[i]n general terms", a penalty is enforced "only to the extent" that compensation can be made for prejudice suffered by failure of the primary stipulation and that a party who can provide compensation "is relieved to that degree from liability to satisfy the collateral stipulation" are similarly directed to, and broadly descriptive of, the grant of equitable relief. The present case does not involve the grant of equitable relief. Nor does the present case involve non-observance of a non-promissory primary contractual stipulation. The customers' claim to recover the amounts charged as, and paid following the imposition of, the late payment fee in the representative proceeding, although variously and elaborately framed, was in substance a common law action in restitution which turned on the enforceability at common law of an obligation to pay a specified sum of money on breach of contract. For present purposes, the significance of Andrews lies in its explanation of the conception of a penalty as a punishment for non-observance of a contractual stipulation, in its explanation of that conception of a penalty as a continuation of the conception which originated in equity, and in its endorsement of the description of the speech of Lord Dunedin in Dunlop as the "product of centuries of equity jurisprudence". The antecedents of Dunlop The ancient equitable antecedents of the speech of Lord Dunedin in Dunlop can be sufficiently picked up towards the end of the eighteenth century in England in the statement by Lord Thurlow in Sloman v Walter of the "rule", then "too strongly established in equity to be shaken", "that where a penalty is inserted merely to secure the enjoyment of a collateral object, the enjoyment of the object is considered as the principal intent of the deed, and the penalty only as accessional, and, therefore, only to secure the damage really incurred". Those equitable antecedents include the subsequent emphatic denials by Lord Eldon and Jessel MR that the equitable jurisdiction to relieve against a penalty was concerned to relieve a contracting party from the consequence of a collateral stipulation to pay an agreed sum of money on breach of contract merely on the basis that the amount contractually agreed was "excessive" or "oppressive". Before Dunlop, English courts experienced difficulty, rising at times to frustration, attempting to articulate the rationale for the rule stated by Lord Thurlow. To attempt to justify refusal to enforce a collateral stipulation solely on the ground that the refusal gave effect to the principal intention of the parties (that the principal contractual stipulation be observed) failed satisfactorily to explain why there should be a refusal to enforce the totality of the intention of the parties (including that the collateral stipulation also be observed). The difficulty was not shared in Scotland. There the Court of Session, "the supreme court of law and equity", adopted the position by at least the middle of the nineteenth century that "a penalty is a punishment" and that the reason for refusing to enforce a penalty was that "it is not legal to stipulate for punishment": to refuse to enforce a penalty was accordingly to give effect to a policy that "[p]arties cannot lawfully enter into an agreement that the one party shall be punished at the suit of the other". The Court of Session came, moreover, to accept that a contract which stipulated for the payment of an amount by way of "liquidated damages" was "not beyond an equitable control". The Court would "deny effect to the mere words of the instrument" if at the time of the contract the amount stipulated was "so utterly extravagant and unreasonable as to infer that, if awarded, it would not be proper damages, though so called, but would really amount to a penalty or punishment". The immediate context of Dunlop was provided by the decision of the House of Lords, on appeal from the Court of Session, in Clydebank Engineering and Shipbuilding Company v Yzquierdo y Castaneda, as followed and explained in the judgment of the Privy Council in Public Works Commissioner v Hills. Clydebank concerned contracts entered into in 1896 for the supply to the Spanish Government by a Clyde shipbuilding firm of four "torpedo-boat destroyers" to be delivered within a specified period of months. The contracts provided that "[t]he penalty for later delivery shall be at the rate of £500 per week for each vessel". At the time of contracting, the Spanish Government was attempting to suppress an insurrection in Cuba and feared the armed intervention of the United States. The Spanish Government brought, and was successful in, an action against the shipbuilding firm in the Court of Session to recover £500 per week for each week that the vessels were in fact delivered late. That Court rejected the shipbuilding firm's argument that the agreement to pay £500 per week was unenforceable because it was a penalty and that the recoverable loss of the Spanish Government for late delivery of each vessel was limited to "the actual outlay on the pay of the officers and crew through detention in Scotland waiting to take over the vessel, and other actual damages of that sort". The Lord Ordinary (Lord Kyllachy) stated at first instance the principle which he went on to apply: "[T]hat in determining the true character of something called penalty, or something called liquidate damage, it was an important, and perhaps conclusive, consideration that the amount of the so-called penalty, or of the so-called liquidated damage, was on the one hand reasonable and moderate, or on the other hand exorbitant and unconscionable. Prima facie of course the parties were the best judges of that matter. Still the amount stipulated might be such as to make it plain that it was merely stipulated in terrorem, and could not possibly have formed a genuine pre-estimate of probable or possible damage, or, to speak perhaps more correctly, a genuine pre-estimate of the creditor's probable or possible interest in the due performance of the principal obligation." Applying that principle in a factual context in which it was established that the price which the Spanish Government paid for the destroyers reflected a premium for early delivery, Lord Kyllachy concluded that the shipbuilding firm "failed to shew that the £500 per week stipulated was exorbitant and unconscionable". The Second Division of the Court of Session affirmed, and the appeal to the House of Lords was dismissed. The principle stated and applied by Lord Kyllachy can be seen to have been reflected in the reasoning of each member of the House of Lords, and to have been adopted in terms by Lord Robertson when he said: "Now the Court can only refuse to enforce performance of this pecuniary obligation if it appears that the payments specified were - I am using the language of Lord Kyllachy - 'merely stipulated in terrorem, and could not possibly have formed' 'a genuine pre-estimate of the creditor's probable or possible interest in the due performance of the principal obligation.'" Demonstrating the neutrality of legal principle in a time of war, the Supreme Court of the United States around two years later applied essentially the same process of reasoning in United States v Bethlehem Steel Company to conclude that a stipulation in a contract for the supply to the United States of armaments during the Spanish-American war, requiring the armaments supplier to pay a fixed amount for each day of late delivery, was not unenforceable as a penalty. In Hills, the Board of the Privy Council included Lord Dunedin and Lord Atkinson. The judgment of the Privy Council was delivered by Lord Dunedin. After referring to Clydebank as having been decided "according to the rules of a system of law where contract law is based directly on the civil law and where no complications in the matter of pleading had ever been introduced by the separation of common law and equity", Lord Dunedin said: "The general principle to be deduced from that judgment seems to be this, that the criterion of whether a sum - be it called penalty or damages - is truly liquidated damages, and as such not to be interfered with by the Court, or is truly a penalty which covers the damage if proved, but does not assess it, is to be found in whether the sum stipulated for can or can not be regarded as a 'genuine pre-estimate of the creditor's probable or possible interest in the due performance of the principal obligation.' The indicia of this question will vary according to circumstances. Enormous disparity of the sum to any conceivable loss will point one way, while the fact of the payment being in terms proportionate to the loss will point the other. But the circumstances must be taken as a whole, and must be viewed as at the time the bargain was made." Months later, dissenting in the result but not in the identification of basal principle, O'Connor J explained in Lamson Store Service Co Ltd v Russell Wilkins & Sons Ltd that he applied to the facts of that case "the simple test suggested by Lord Dunedin, namely, whether the 'sum stipulated for can or can not be regarded as a genuine pre-estimate of the creditor's probable or possible interest in the due performance of the principal obligation'". The same test was applied by Isaacs J in Hamilton v Lethbridge. Dunlop in perspective The litigation in Dunlop was fully explained in Andrews. That explanation need not be repeated. The context of Clydebank and Hills sheds light on the reasoning in Dunlop in two important respects. First, the context of Clydebank and Hills explains the provenance and import of the language used by Lord Atkinson to formulate the conclusion that the stipulated amount in issue in Dunlop "was not stipulated for merely in terrorem, but was really and genuinely a pre-estimate of the appellants' probable or possible interest in the due performance of the contract". The language derives from Scots law. Its importance lies in its provision of a stable justification for the penalty doctrine founded in concepts of public policy conformable with the doctrine's equitable origins. The context thus adds weight to the observation in Andrews that in Dunlop "the critical issue, determined in favour of the appellant, was whether the sum agreed was commensurate with the interest protected by the bargain". Second, that context also explains the provenance and import of the language used by Lord Dunedin when, after stating that the relevant "branch of the law" had "been handled, and at no distant date" in Clydebank and Hills, he said that he chose to "content" himself with "stating succinctly the various propositions which [he thought were] deducible from the decisions which rank as authoritative". The enumerated propositions which followed were not rules of law. They were distillations of principle intended by Lord Dunedin to be consistent with the then recent statements of principle in both Clydebank and Hills. The second proposition enumerated by Lord Dunedin, for which he cited Clydebank, conformed precisely to the critical issue identified by Lord Atkinson in the manner of formulation of his conclusion. "The essence of a penalty", said Lord Dunedin, "is a payment of money stipulated as in terrorem of the offending party." Lord Dunedin immediately expressed the corollary in terms that "the essence of liquidated damages is a genuine covenanted pre-estimate of damage". That reference to "damage", as distinct from "damages", is significant. The "damage" to which Lord Dunedin referred corresponded to the consequence of non-realisation of what, echoing the words of Lord Kyllachy in Clydebank, Lord Dunedin himself had earlier described in Hills as "the creditor's probable or possible interest in the due performance of the principal obligation". It was not a reference to the extent to which the relevant interest of the creditor in the performance of the bargain would be able to be compensated by an award of unliquidated damages in an action for breach of contract at common law. The third proposition enumerated by Lord Dunedin, for which he again cited Clydebank, was that "[t]he question whether a sum stipulated is penalty or liquidated damages is a question of construction to be decided upon the terms and inherent circumstances of each particular contract, judged of as at the time of the making of the contract, not as at the time of the breach". Lord Dunedin evidently used the word "construction" to refer to something beyond the attribution of legal meaning. He used it to encompass legal characterisation. His added reference to the "inherent circumstances of each particular contract", although not then the subject of further elaboration, was not, in light of Clydebank, confined to circumstances which bore only on the attribution of legal meaning but extended to all of the circumstances which bore on the objective resolution of the ultimate question of characterisation. The fourth proposition enumerated by Lord Dunedin was introduced as being "[t]o assist this task of construction", and amounted to a listing of what he had identified in Hills as "indicia" which would vary with the circumstances, which must always be taken as a whole. The indicia were inappropriately labelled in the course of argument in this Court as "rules 4(a), 4(b) and 4(c)". Viewed in their proper perspective, none can be understood as having the character or effect of an operative legal rule. The statement that a stipulated amount "will be held to be penalty 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" harked back to Lord Kyllachy's statement of principle in Clydebank. The statement that a stipulated amount "will be held to be a penalty if the breach consists only in not paying a sum of money, and the sum stipulated is a sum greater than the sum which ought to have been paid", was explained to be "truly a corollary to the last test". And the statement that a stipulated amount is a penalty "when 'a single lump sum is made payable by way of compensation, on the occurrence of one or more or all of several events, some of which may occasion serious and others but trifling damage'", was explained to be "a presumption (but no more)". The words "test" and "presumption" were evidently used to mean "indicia". They were not used to import either a legal criterion or a shift in the evidentiary or persuasive onus. That Lord Dunedin's fourth proposition amounted to no more than a listing of considerations which might indicate a payment of money to have been stipulated as in terrorem of an offending party is demonstrated by the circumstance that the only case cited by Lord Dunedin as an illustration of the second in fact had the features of all three of the indicia to which he referred. The case was Kemble v Farren. There the relevant contract was between the manager of the Covent Garden Theatre and a comic actor under which the actor was to be paid just over £3 for each day that the theatre was open over the course of four theatre seasons. The contract obliged each party to comply with many covenants, and stipulated that each party pay £1000 on any breach of one of those covenants. The breach which led to the action in which the stipulation was held to be unenforceable at common law on the ground that it imposed a penalty was the actor's refusal to act for a second season. English cases in the nineteenth century had routinely contained statements which accepted that a larger sum payable only on a breach of a covenant to pay a smaller sum would be a penalty at law and in equity. Lord Dunedin was saying that those statements were not, at least by the beginning of the twentieth century, to be understood as embodying a distinct legal rule. It is telling that the well-resourced parties in the present case failed to unearth a single instance of an English or Australian case decided before Dunlop (or afterwards and by reference to Dunlop) where a stipulated amount has been held to be a penalty solely on the basis that a breach of contract triggering the obligation to pay the amount consisted only in not paying a sum of money, and the sum stipulated was greater than that which ought to have been paid. The unintended consequence of lucidity is sometimes rigidity. Over the century after Dunlop, in the course of being "applied countless times in this and other courts", Lord Dunedin's explication of principle came to acquire a quasi-statutory status. Unfortunately, that status sometimes obscured its essential meaning. In AMEV-UDC, after noting with reference to Clydebank and Dunlop the historic concept of an amount which was agreed to be paid on breach of contract being a penalty if it is "extravagant, exorbitant or unconscionable", Mason and Wilson JJ observed: "This concept has been eroded by more recent decisions which, in the interests of greater certainty, have struck down provisions for the payment of an agreed sum merely because it may be greater than the amount of damages which could possibly be awarded for the breach of contract in respect of which the agreed sum is to be paid". The more recent decisions to which their Honours referred were for the most part cases decided towards the middle of the twentieth century in the context of hire purchase agreements, before the advent of modern consumer protection legislation. In terms subsequently to receive the unanimous endorsement of this Court in Ringrow, Mason and Wilson JJ continued: "However, there is much to be said for the view that the courts should return to the Clydebank and Dunlop concept, thereby allowing parties to a contract greater latitude in determining what their rights and liabilities will be, so that an agreed sum is only characterized as a penalty if it is out of all proportion to damage likely to be suffered as a result of breach". All of this demonstrates consistency with the equitable root of the penalty doctrine of the more contemporary explanation of its operation given by Diplock LJ in Robophone Facilities Ltd v Blank. Diplock LJ said: "[T]he parties to a contract may expressly stipulate not only what will be what I there called their primary obligations and rights under the contract, ie, those which are discharged by performance of the contract, but also what will be their secondary obligations and rights, ie, those which arise upon non-performance of any primary obligation by one of the parties to the contract. Of these secondary obligations and rights, the commonest is the obligation of the non-performer to make to the other party and the corresponding right of such other party to claim from the non-performer reparation in money for any loss sustained by the other party which results from the failure of the non-performer to perform his primary obligation." He continued: "But the right of parties to a contract to make such a stipulation is subject to the rule of public policy that the court will not enforce it against the party in breach if it is satisfied that the stipulated sum was not a genuine estimate of the loss likely to be sustained by the party not in breach, but was a sum in excess of such anticipated loss and thus, if exacted, would be in the nature of a penalty or punishment imposed upon the contract-breaker." Frankfurter J had written to substantially similar effect in Priebe & Sons Inc v United States. He referred to the doctrine that "exactions for a breach of contract not giving rise to damages and merely serving as added pressure to carry out punctiliously the terms of a contract, are not enforced by courts". He said that he assumed that the "basic reason" for the doctrine was "that the infliction of punishment through courts is a function of society and should not inure to the benefit of individuals". Consistently with those observations of Diplock LJ and of Frankfurter J, after quoting with approval the observation of Mason and Wilson JJ in AMEV-UDC, this Court said in Ringrow: "Exceptions from … freedom of contract require good reason to attract judicial intervention to set aside the bargains upon which parties of full capacity have agreed. That is why the law on penalties is, and is expressed to be, an exception from the general rule. It is why it is expressed in exceptional language. It explains why the propounded penalty must be judged 'extravagant and unconscionable in amount'. It is not enough that it should be lacking in proportion. It must be 'out of all proportion'." Framing the inquiry In O'Dea v Allstates Leasing System (WA) Pty Ltd, Wilson J asked of the stipulation in issue in that case whether it "can be considered to be a 'genuine pre-estimate of the creditor's ... probable or possible interest in the due performance of the principal obligation'" (citing Hills) or "whether it is a penalty inserted 'merely to secure the enjoyment of a collateral object'" (citing Sloman v Walter). That succinct framing of the inquiry is consistent with Andrews and Dunlop. It is also very useful. The ultimate question of whether a stipulation imposing a detriment on a contracting party in the event of non-observance of another stipulation is a penalty is reflected in the formulation Wilson J drew from Sloman v Walter. To ask whether a stipulation serves merely to secure the enjoyment of a collateral object is to ask whether the conclusion objectively to be drawn from the totality of the circumstances is that the only purpose of the stipulation was to punish: to impose a detriment on a contracting party in the event that a principal contractual stipulation is not observed, in order to deter non-observance of that principal stipulation. To ask that question in the context of a stipulation for the payment of money on breach of contract accords with the statement of Lord Dunedin in Dunlop that "[t]he essence of a penalty is a payment of money stipulated as in terrorem of the offending party". The formulation Wilson J drew from Hills does not indicate some separate or different inquiry but rather indicates the nature of the inquiry involved in considering whether a stipulation has some purpose other than to punish. The formulation captures in positive and more elaborate terms what Lord Dunedin reflected in negative and cryptic terms when he added, as a reflex of the essence of a penalty, that "the essence" of a payment of money on breach of contract that is not a penalty is that the amount of the payment "is a genuine covenanted pre-estimate of damage". To ask whether a stipulated payment is a genuine pre-estimate of the innocent party's probable or possible interest in the due observance of the principal contractual stipulation is to ask whether an interest which the innocent party has in the observance of that principal stipulation explains the stipulation for payment as having a purpose other than to punish the offending party. Such an interest of the innocent party in the observance of a principal contractual stipulation need not be an interest in respect of which the offending party would otherwise be compelled to compensate the innocent party at law (or in equity) in the event of non-observance. Where the stipulated detriment is in the form of an obligation to pay a specified sum of money in the event of a breach of contract, a comparison of the specified sum with the amount of the unliquidated damages which might be expected to be recovered by the innocent party in an action for breach of contract will often be probative of whether the only purpose of the stipulation is to punish. Such a comparison might sometimes be decisive. Not always. The facts in Clydebank and Dunlop both sufficiently illustrate that interests of the innocent party beyond the protection of an award of unliquidated damages in the event of a breach of contract can justify a different conclusion. The protection afforded by the stipulation of an obligation to pay a specified sum of money in the event of a breach of contract might be to interests that the innocent party has in contractual performance which are intangible and unquantifiable. A party seeking contractually to protect its interests by insisting on a stipulation that another party pay a specified sum of money in the event of breach cannot be limited by considerations of common law causation of damage to protecting only against incremental loss that the party would sustain as the direct result of that breach. Nor can a party seeking contractually to protect that party's interests by insisting on such a stipulation be compelled by considerations of common law remoteness of damage to absorb actual damage to those interests consequential on breach merely because the nature and extent of that damage might not be apparent to the other party at the time of entering into the contract. The English Law Commission correctly observed in 1975 that "[t]here would seem to be no reason why the parties in Hadley v Baxendale could not have contracted for liquidated damages assessed on the footing that the mill would continue to be at a standstill". In that respect, another explanation given by Diplock LJ in Robophone is instructive: "The basis of the defendant's liability for the enhanced loss under the 'second rule' in Hadley v Baxendale is his implied undertaking to the plaintiff to bear it. ... But such an undertaking need not be left to implication; it can be express. If the contract contained an express undertaking by the defendant to be responsible for all actual loss to the plaintiff occasioned by the defendant's breach, whatever that loss might turn out to be, it would not affect the defendant's liability for the loss actually sustained by the plaintiff that the defendant did not know of the special circumstances which were likely to cause any enhancement of the plaintiff's loss. And so if at the time of the contract the plaintiff informs the defendant that his loss in the event of a particular breach is likely to be £X by describing this sum as liquidated damages in the terms of his offer to contract, and the defendant expressly undertakes to pay £X to the plaintiff in the event of such breach ... [s]uch a clause is ... enforceable whether or not the defendant knows what are the special circumstances which make the loss likely to be £X rather than some lesser sum which it would be likely to be in the ordinary course of things." And as the facts in Clydebank and Dunlop again both sufficiently illustrate, the fact that the amount of a payment stipulated to be made on breach of contract is set at a level which provides a negative incentive - even a very strong negative incentive - to perform the contract is not enough to justify the conclusion that the stipulation served only to punish. The prospect of paying compensatory damages to be assessed by a court in the event of breach itself provides a negative incentive to perform a contract. The relevant indicator of punishment lies in the negative incentive to perform being so far out of proportion with the positive interest in performance that the negative incentive amounts to deterrence by threat of punishment. To accept the appropriateness of framing the inquiry in terms of whether the stipulation in issue is properly characterised as having no purpose other than to punish is to reject the suggestion, which originated with Lord Radcliffe in Bridge v Campbell Discount Co Ltd, that the description of a penalty as a threat to be enforced in terrorem adds nothing to the requisite analysis. To the contrary, the description captures the essence of the conception to which the whole of the analysis is directed. Framing the inquiry in terms of whether the stipulation in issue is properly characterised as having no purpose other than to punish compels a more tailored inquiry into the commercial circumstances within which the parties entered into the contract containing the stipulation than might be involved in asking, as did the Supreme Court of the United Kingdom in Cavendish, whether the stipulation serves a "legitimate interest". That is not, of course, to say that the differently framed inquiries might not lead to the same result. Was the late payment fee a penalty? The credit card contracts entered into between ANZ and Mr Paciocco were on terms and conditions relevantly no different from the terms and conditions on which ANZ entered into credit card contracts with each of the other holders of consumer credit card accounts, of which there were millions. The stipulation in each of those contracts for the charging and payment of the late payment fee in the event of non-observance by the account holder of the stipulation for payment of the minimum monthly payment by the due date was properly characterised as the imposition of an additional contractual liability by reason of non-observance of a primary contractual stipulation. The customers' claim that that additional contractual liability was unenforceable as a penalty triggered an inquiry whether, within the totality of the circumstances within which ANZ contracted with its consumer credit card account holders, the stipulation for the payment of the late payment fee was properly characterised as: having no purpose other than to punish an account holder in the event of late payment; or conversely serving the purpose of protecting ANZ's interests in ensuring that consumer credit card account holders made the minimum monthly payment by the due date. The customers bore the evidentiary and persuasive onus throughout that inquiry. That the primary contractual stipulation consisted only in the payment of money, and that the amount of the late payment fee did not vary according to the amount overdue or the length of delay in payment, were indicia which could not be ignored in considering the conclusion to be drawn in the totality of the circumstances. But each of those circumstances was only weakly indicative of the character of the late payment fee as a punishment, given that: the minimum monthly payment was to be payable monthly; the amount of the minimum monthly payment was calculated ordinarily to be a very small percentage of the closing balance of the account; the account holder could control the amount of the closing balance by self-regulating the timing and amount of credit card transactions; and the account holder could cancel the account at any time. The evidence of Mr Regan was not wholly irrelevant to the inquiry to be undertaken. That Mr Regan focused, ex post, on costs incurred by ANZ in consequence of late payment by Mr Paciocco was not of itself a reason for discounting the value of his evidence. As has been recognised in a variety of contexts, evidence of the later occurrence of an event can be probative of the earlier probability of that event occurring. The danger to be avoided in weighing evidence of that nature is that which has been described in behavioural science as "availability bias": the natural but flawed human tendency to treat events that occurred as having had a greater probability of occurrence simply because they occurred. That was not a problem with Mr Regan's evidence. There was no reason to consider that, within the context of ANZ's ongoing provision of millions of consumer credit card accounts, the events which occurred during the operation of Mr Paciocco's credit card accounts were not indicative of the events which were foreseeable as likely to occur at the times the credit card contracts were entered between ANZ and Mr Paciocco. Nor is there any reason to consider that Mr Paciocco's circumstances were wholly unrepresentative of those of other account holders. What limited the utility of Mr Regan's evidence to the inquiry was the confinement of his attention to variable or incremental operational costs incurred within ANZ's Collections Business Unit. The calculations which Mr Regan performed would have been entirely appropriate to quantify the compensable loss to ANZ which flowed from each breach of the stipulation for payment of a minimum monthly payment by the due date. Those calculations alone were not sufficient to indicate the totality of ANZ's interests in ensuring that the stipulation for payment of a minimum monthly payment by the due date was observed. Turning from the evidence of Mr Regan to the evidence of Mr Inglis, that neither of the principal categories of costs which Mr Inglis identified in connection with late payment events would be compensable at the suit of ANZ against an individual customer may readily be accepted. As to provisioning costs, the amount to be recovered on a chose in action by the holder of that chose cannot possibly extend to a diminution in the value of the chose itself that is attributable to the potential for non-recovery. As to regulatory capital costs, those costs would plainly be too remote. Each of those two categories of costs nevertheless represented commercial interests which ANZ had in ensuring that its credit card customers, as a cohort, made minimum monthly payments by the due date. Provisioning costs directly affected recorded profit, and the costs of regulatory capital were a real outgoing. The primary judge was undoubtedly correct in describing provisioning costs and regulatory capital costs as part of the costs of running a bank in Australia. But ANZ was not confined by a principle of law to adopting a pricing strategy for its credit card products which involved cross-subsidisation. The calculations of Mr Inglis showed the commercial interests of ANZ in avoiding those two principal categories of costs to be not grossly disproportionate to the amount of the late payment fee, even if operational costs were ignored. Notwithstanding the attention given to them by the parties in argument in this Court, the costs of collection activities within ANZ's Collections Business Unit considered on an average basis were not particularly significant when compared with provisioning costs and regulatory capital costs. Although nothing of substance therefore turns on their precise quantification, ANZ obviously had a commercial interest in avoidance or minimisation of common costs and fixed costs associated with those collection activities in the same way as ANZ had an interest in avoidance or minimisation of the variable or incremental costs of those activities. The calculations of collection costs by Mr Inglis are for that reason for present purposes to be preferred to the calculations of Mr Regan. Although of little consequence in the scheme of things, it is worth noting a discrete argument concerning collection costs made on behalf of the customers for the first time in this Court. The argument is that those costs were recoverable by ANZ under the separate provisions of the standard form credit card contracts which provided for ANZ to recover "enforcement expenses" in the event of default. The answer to the argument is that the collection activities in question fell short of enforcement and were therefore beyond the scope of that separate provision. To include the recovery of the costs of those activities within the commercial interests of ANZ which were capable of providing a non-punitive justification for the stipulation for the late payment fee therefore involves no element of double counting. Each category of costs identified by Mr Inglis represented a commercial interest of ANZ in ensuring observance by its consumer credit card customers of the principal stipulation in each of their contracts for payment of the minimum monthly payment by the due date. The customers' grounds of appeal to this Court do not encompass any challenge to Mr Inglis' evidence of their quantification. In light of those interests, it cannot be concluded that the inclusion in the credit card contracts of the stipulation for charging and payment of the late payment fee properly had no purpose other than to punish the account holder in the event of late payment. The stipulation was not merely in terrorem; the late payment fee was not just a punishment. The Full Court was correct to conclude that the customers failed to demonstrate that the late payment fee was a penalty. The first appeal must for that reason be dismissed. Did ANZ contravene any applicable statutory norm? The customers' claims in the representative proceeding included that ANZ engaged in "unconscionable conduct" within the meaning of ss 12CB and 12CC of the Australian Securities and Investments Commission Act 2001 (Cth) ("the ASIC Act") and ss 8 and 8A of the Fair Trading Act 1999 (Vic) ("the FT Act") by entering into and implementing the standard contractual stipulation for the charging of the late payment fee. They claimed also that the credit card contracts were "unjust" within the meaning of s 76 of the National Credit Code in Sched 1 to the National Consumer Credit Protection Act 2009 (Cth) and "unfair" within the meaning of s 32W of the FT Act and s 12BG of the ASIC Act. ANZ did not dispute that it was bound in its dealings with consumer credit card customers by each of those statutory norms. Not all of those statutory norms were applicable throughout the period to which the customers' claims related, and those that were applicable were each the subject of amendment, but nothing turns on that detail. There was no substantial controversy between the parties as to the content of those norms. The controversy was as to their application. For the purposes of the second of the appeals to this Court, the customers accept that their argument that ANZ engaged in unconscionable conduct is no stronger insofar as it relies on the other and later statutory provisions than insofar as it relies on s 12CB of the ASIC Act in the form that section took until 1 January 2012. Section 12CB then relevantly provided: "(1) A person must not, in trade or commerce, in connection with the supply or possible supply of financial services to a person, engage in conduct that is, in all the circumstances, unconscionable. (2) Without limiting the matters to which the court may have regard for the purpose of determining whether a person (the supplier) has contravened subsection (1) in connection with the supply or possible supply of services to a person (the consumer), the court may have regard to: (a) the relative strengths of the bargaining positions of the supplier and the consumer; and (b) whether, as a result of conduct engaged in by the supplier, the consumer was required to comply with conditions that were not reasonably necessary for the protection of the legitimate interests of the supplier; and (c) whether the consumer was able to understand any documents relating to the supply or possible supply of the services; and (d) whether any undue influence or pressure was exerted on, or any unfair tactics were used against, the consumer or a person acting on behalf of the consumer by the supplier or a person acting on behalf of the supplier in relation to the supply or possible supply of the services; and (e) the amount for which, and the circumstances under which, the consumer could have acquired identical or equivalent services from a person other than the supplier. ... (4) For the purpose of determining whether a person has contravened subsection (1) in connection with the supply or possible supply of financial services to another person: (a) the court must not have regard to any circumstances that were not reasonably foreseeable at the time of the alleged contravention". The customers place the weight of their argument on s 12CB(2)(a) and (b) of the ASIC Act. They point out that the effect of those provisions is specifically to include, within the matters to which regard may be had for the purpose of determining whether ANZ (the supplier) contravened s 12CB(1) in connection with the supply of financial services to Mr Paciocco (the consumer), the relative strengths of the bargaining positions of ANZ and Mr Paciocco and "whether, as a result of conduct engaged in by [ANZ], [Mr Paciocco] was required to comply with conditions that were not reasonably necessary for the protection of the legitimate interests of [ANZ]". The customers emphasise that the late payment fee was unilaterally set by ANZ in the context of standard form consumer contracts which were not open to negotiation by a customer. They go on to argue that the late payment fee charged to Mr Paciocco: "was not a price for any service being provided (but an amount payable upon breach of contract)"; "was not a genuine pre-estimate of loss agreed between the parties (on any ordinary understanding of those words)"; "was intended to secure the performance by [Mr Paciocco] of his contractual obligation to make timeous payment of monies borrowed"; and "resulted in windfall gains to [ANZ] by reason of the disparity between the level of the late payment fee and the actual loss sustained by [ANZ] by reason of the late payments of [Mr Paciocco]". The value-laden terminology in which the customers couch the characteristics of the late payment fee on which they rely for the conclusion that its imposition on Mr Paciocco was unconscionable implicitly incorporates conclusions about the legal operation and commercial context of the late payment fee which have already been rejected in the context of rejecting the customers' argument that the late payment fee was to be characterised as a penalty. It would be open to reject their argument concerning statutory unconscionability on the basis that they have failed to establish each of the argument's premises. The argument, however, has deeper problems. To the extent that the customers' argument relies on s 12CB(2)(b) of the ASIC Act, it is based on a misconstruction of that provision. The concern of s 12CB(2)(b) was not, in its primary operation, with the substantive content of a condition with which a consumer was required to comply but rather with conduct on the part of the supplier as a result of which the consumer was required to comply with that condition. Section 12CB(2)(b) was not enlivened, and therefore raised no question about whether or not a condition with which a consumer was required to comply was reasonably necessary for the protection of the legitimate interests of the supplier, unless a requirement on the part of the consumer to comply with a condition was brought about by conduct on the part of the supplier. ANZ did not cause Mr Paciocco to enter into credit card contracts which contained what was, in the language of s 12CB(2)(b) of the ASIC Act, the "condition" for the late payment fee; it presented him with standard form contracts on a take-it-or-leave-it basis. And ANZ did not cause Mr Paciocco to fail to make minimum monthly payments as a result of which he became subject to the requirement to pay the late payment fee in compliance with that condition. There was, in short, no conduct on the part of ANZ which enlivened the operation of s 12CB(2)(b). More significantly, in focusing on s 12CB(2)(a) and (b), the customers' argument implicitly overstates the significance of those provisions within the overall scheme of s 12CB of the ASIC Act. The statutory question raised by the customers' claim that ANZ engaged in unconscionable conduct within the meaning of s 12CB(1) of the ASIC Act when it entered into and then implemented its standard contractual stipulation for the charging of the late payment fee to Mr Paciocco was whether that conduct was objectively to be characterised as "unconscionable" according to the ordinary meaning of that term, requiring as it does a "high level of moral obloquy" on the part of the person said to have acted unconscionably. The answer to that question necessarily turned on a consideration of that conduct in the context of what s 12CB(1) described as "all the circumstances". The word "may" in s 12CB(2) of the ASIC Act was not permissive, but conditional. The import of s 12CB(2) was to spell out that circumstances relevant to the determination of whether conduct was objectively to be characterised as "unconscionable" according to the ordinary meaning of that term might or might not include, in respect of particular conduct, all or any of the particular matters referred to in s 12CB(2). The provision made clear that, where any one or more of those matters existed in respect of particular conduct, each of those extant matters was to form part of the totality of the circumstances mandatorily to be taken into account for the purpose of determining the statutory question posed by s 12CB(1). The provision did not leave it open to a consumer who alleged that conduct of a supplier was in breach of s 12CB(1) to pick and choose. The customers could not choose to rely on matters referred to in s 12CB(2)(a) and (b), yet to ignore matters referred to in s 12CB(2)(c), (d) and (e). The existence and amount of the late payment fee were disclosed to Mr Paciocco in the letters, booklets and telephone calls which he received from ANZ. He was able to, and did, understand them. There has never been any suggestion of undue influence or pressure having been exerted on him or of unfair tactics having been used against him. Mr Paciocco freely chose to enter into the two credit card contracts with ANZ and could have terminated those contracts at any time at will. He could at any time have sought to obtain a credit card from another bank. Other banks were in fact charging broadly equivalent fees. Mr Paciocco chose instead to maintain his accounts with ANZ, to manage those accounts at close to their limits and to bear the risk of being charged the late payment fee on those occasions when he failed to comply with the standard stipulation to make the minimum monthly payment by the due date. Taking all those circumstances into account, even if it were accepted that charging Mr Paciocco the late payment fee resulted in windfall gains to ANZ (which, on the evidence of Mr Inglis, it is to be concluded that ANZ did not obtain), the proper conclusion still to be drawn would be that ANZ did not engage in unconscionable conduct within the meaning of s 12CB(1) of the ASIC Act when it entered into and then implemented its standard contractual stipulation for the charging of the late payment fee to Mr Paciocco. That conclusion makes it unnecessary separately to consider the other and later statutory provisions on which the customers relied in making their statutory unconscionability claim. The customers' argument that the standard contractual stipulation for the charging of the late payment fee resulted in the two credit card contracts between Mr Paciocco and ANZ being "unjust" within the meaning of s 76 of the National Credit Code mirrors their statutory unconscionability argument in structure and content. The argument must suffer the same fate, for substantially the same reasons. Section 76(1) of the National Credit Code permitted a court, on the application of a debtor, to "reopen" a transaction that gave rise to a relevant credit contract if satisfied that, at the time it was entered into, the contract was "unjust" - an expression defined for that purpose to include "unconscionable, harsh or oppressive". On reopening the contract, the court was empowered by s 77(b) to make orders which included relieving the debtor from payment of any amount in excess of such amount as the court, having regard to the risk involved and all other circumstances, considered to be reasonably payable. Section 76(2) of the National Credit Code provided that, in determining whether a term of a particular credit contract was unjust in the circumstances relating to it at the time it was entered into, a court was "to have regard to the public interest and to all the circumstances of the case". Section 76(2) added that the court "may" have regard to a long list of specifically enumerated factors, the last of which was expressed to be "any other relevant factor". One of those factors, enumerated as s 76(2)(e), was "whether or not any of the provisions of the contract ... impose[d] conditions that [were] ... not reasonably necessary for the protection of the legitimate interests of a party to the contract". That factor differed from the factor enumerated in s 12CB(2)(b) of the ASIC Act in that it was concerned in its primary operation with the substantive content of the contract in question. Section 76(4) was similar to s 12CB(4)(a) of the ASIC Act in prohibiting a court from considering "injustice arising from circumstances that were not reasonably foreseeable when the contract ... was entered into". Having regard to the evidence of Mr Inglis quantifying the average costs incurred by ANZ in connection with the occurrence of an event giving rise to an entitlement to charge the late payment fee as having been in excess of $35 or $50, it could not be concluded that the conditions of Mr Paciocco's credit card contracts which imposed the late payment fee of either $35 or $20 were not reasonably necessary for the protection of the legitimate interests of ANZ. The factor enumerated in s 76(2)(e) was for that reason not shown to be engaged. But, again, even if it were to be accepted that charging Mr Paciocco the late payment fee resulted in windfall gains to ANZ (which on the evidence of Mr Inglis it did not), that would not be enough to justify the conclusion that the credit card contracts were unjust. Section 76(2) made clear that a conclusion about whether a contract was unjust could only be drawn having regard to all of the circumstances of the case. The customers identified no element of the public interest which might bear on the evaluation required by s 76(2) of the National Credit Code beyond considerations of public interest which inhere in any evaluation of whether conduct is unconscionable, harsh or oppressive. When regard is had to all the circumstances of the case, it could not be concluded that the standard contractual stipulation for the charging of the late payment fee resulted in the two credit card contracts between Mr Paciocco and ANZ being "unjust" for essentially the same reasons that it could not be concluded that ANZ's entering into and implementation of that standard contractual stipulation was "unconscionable". The customers' argument focusing on the "unfair" limb of s 32W of the FT Act is a variation on the same theme, and must again meet the same fate. They do not suggest that the argument would yield a different result were it to be focused on s 12BG of the ASIC Act. Section 32Y of the FT Act provided that "[a]n unfair term in a consumer contract is void". The meaning of the expression "unfair term" was addressed in s 32W as follows: "A term in a consumer contract is to be regarded as unfair if, in all the circumstances, it causes a significant imbalance in the parties' rights and obligations arising under the contract to the detriment of the consumer." Section 32W needed to be read with s 32X, which went on to provide that "[w]ithout limiting section 32W, in determining whether a term of a consumer contract is unfair, a court ... may take into account, among other matters, whether the term was individually negotiated" and "whether the term has the object or effect of", amongst other things, "penalising the consumer but not the supplier for a breach or termination of the contract". Tailoring their more general statutory arguments to the language of s 32W, the customers argue that the contractual stipulation for charging the late payment fee caused a significant imbalance in the rights and obligations of ANZ and of Mr Paciocco under the credit card contracts, to the detriment of Mr Paciocco as the consumer. That was because ANZ was put "in a situation where it [could] profit from breaches of contract by the consumer without any quid pro quo". The significance of that imbalance, they argue, was "demonstrated by the lack of any meaningful relationship between the amount of the late payment fee and the reasonably foreseeable loss which would result to [ANZ] from late payment". One answer to the argument is that its minor premise has not been established: there was a meaningful relationship between the amount of the late payment fee and the accrual of costs to ANZ from the occurrence of late payment events and that those costs were reasonably foreseeable at the time of contracting. The evidence of Mr Inglis showed that relationship. The more complete answer to the argument is that its major premise is flawed. To demonstrate that the stipulation for payment put ANZ in a situation where it might profit from breaches of contract by a credit card customer without the customer in breach acquiring something in return would not alone be sufficient to allow it to be concluded that the stipulation caused a significant imbalance in the parties' rights and obligations arising under the contract. Even if the stipulation could be characterised as a matter of ordinary language as "penalising the consumer but not the supplier for a breach or termination of the contract", that was only one factor amongst many to be taken into account. The Full Court was correct to conclude that the customers failed to demonstrate that the contractual stipulation to pay the late payment fee contravened any of the applicable statutory norms. The second appeal must for that reason be dismissed. Orders Both appeals should be dismissed with costs.

KEANE J. These appeals concern late payment fees charged by the respondent, Australia and New Zealand Banking Group Limited ("ANZ"), in connection with two credit card accounts held by the first appellant, Mr Lucio Robert Paciocco. Mr Paciocco and the second appellant, Speedy Development Group Pty Ltd ("SDG"), brought an action against ANZ in the Federal Court of Australia as a representative proceeding under Pt IVA of the Federal Court of Australia Act 1976 (Cth). They claimed relief in respect of a range of fees charged in connection with Mr Paciocco's credit card accounts, a consumer deposit account held by Mr Paciocco, and a business deposit account held by SDG. At trial, those fees were referred to by the primary judge (Gordon J) compendiously as "exception fees". The appellants claimed that the exception fees were unenforceable as penalties; alternatively, they claimed that charging the exception fees contravened standards set by a number of statutory regimes which protect consumers against "unconscionable conduct", "unjust transactions" and "unfair contract terms". The primary judge held that only the exception fees charged in connection with Mr Paciocco's credit card accounts, which her Honour referred to as "late payment fees", were unenforceable as penalties. The primary judge did not consider whether the late payment fees breached any of the statutory regimes on which the appellants relied. As to the alleged breaches of the statutory provisions in respect of the other exception fees, the primary judge dismissed the appellants' claims. ANZ appealed to the Full Court of the Federal Court against the primary judge's decision that the late payment fees were penalties. The appellants appealed against the primary judge's decision in relation to the other exception fees. The Full Court allowed ANZ's appeal, and dismissed the appellants' appeal. The appellants appeal to this Court pursuant to a grant of special leave. There are two appeals before this Court. The issue that arises for determination in appeal M219 of 2015 is whether the late payment fees charged by ANZ contravened the proscriptive statutory regimes referred to above. The issue in appeal M220 of 2015 is whether those fees are unenforceable as penalties under the general law. It is convenient to refer to the claims in appeal M219 as "the statutory claims", and to the claims in appeal M220 as "the penalty claims". For the reasons that follow, the issues in the two appeals should be determined adversely to the appellants and the appeals should be dismissed. By a notice of contention filed in appeal M220 of 2015, ANZ argued that the appellants' claims to recover late payment fees incurred more than six years before proceedings were commenced are time‑barred by the operation of the Limitation of Actions Act 1958 (Vic). Because the late payment fees were not penalties, it is not necessary to resolve this issue. The penalty claims It is convenient to deal first with the issue of whether ANZ's late payment fees were unenforceable penalties under the general law. In that regard, some general observations by way of an overview of the appellants' case are necessary. First, it is necessary to be clear as to what the case is not about. The appellants did not advance a case that the late payment fees were a manifestation of an unlawful abuse by ANZ of its market power. No attempt was made to advance a case to that effect under Pt IV of the Competition and Consumer Act 2010 (Cth). No attempt was made to advance a case of misconduct in contravention of any of the provisions of Div 2 of Pt 7.10 of the Corporations Act 2001 (Cth). Accordingly, it should be understood that a rejection of the case which the appellants did advance does not mean that there is no limit to the extent of fees and charges that banks might lawfully charge. Such a concern might be met by invoking laws which are directed to prevent the abuse of market power or dishonest conduct in the market. Secondly, it was no part of the appellants' case that the charges by ANZ or other banks for their financial services should be fixed by the courts at reasonable levels. There are obvious difficulties inherent in determining what level of interest and charges would be "reasonable". In any event, no jurisdiction to make such a determination has been conferred on the courts. That said, the medieval laws against usury serve as a reminder that such laws have been tried in the past. But currently in Australia, no legislation authorises the application by the courts of a standard of reasonableness to determine the lawfulness of bank charges; and it is not suggested that the common law has developed such a standard. Thirdly, the appellants did not advance a case that Mr Paciocco suffered from the kind of disability which attracts the intervention of a court of equity to protect the weaker party to a contract. The strength of the appellants' case lies solely in the finding by the primary judge that the cost actually incurred by ANZ in consequence of each failure by Mr Paciocco to pay his credit card account on time was of the order of $3. That amount can be contrasted with the amount of the late payment fee actually charged by ANZ: initially $35 and later $20. The large disparity between the late payment fee charged by ANZ and the expenses actually incurred by it in each case of late payment is the focus of the appellants' case. That disparity may well mean that the late payment fee could accurately be characterised as an example of profiteering by the bank. But whether a late payment fee is to be characterised as an unenforceable penalty is not to be determined by asking whether the enforcement of the fee will produce profits, even large profits, for the bank. The case advanced by the appellants was that the late payment fee was to be characterised as a penalty because its purpose was to punish Mr Paciocco for breaching his contractual obligation to make timely payment or to deter him from choosing not to perform his contractual obligation. And the appellants sought to make this case good by evidence which showed that the late payment fee exceeded the expenses actually incurred by ANZ on each occasion of default by Mr Paciocco. The disparity was said to be so great that the late payment fee could be seen to be out of all proportion to the bank's interest in recovering the expenses actually incurred by it. To argue from these premises that the contractual purpose which characterised the late payment fee charged by ANZ was the punishment of its customers is fraught with difficulty once it is accepted that the bank's legitimate interests are not confined to the reimbursement of the expenses directly occasioned by the customer's default. The maintenance or even enhancement of ANZ's revenue stream, for the purpose of making a profit, is one explanation of the late payment fee. Indeed, it is the most obvious explanation because, generally speaking, it is the purpose which informs all the terms on which a bank makes its facilities available to its customers. And although interest payments are the primary source of reward to a bank for financial risks involved in the provision of financial accommodation to its customers, there is no legal reason why a bank's fees and charges may not serve the same purpose. In short, the late payment fee is readily characterised by the purpose of ensuring that ANZ's revenues are maintained at the level of profitability required by its shareholders. And the appellants did not seek to advance a case that the pursuit of this level of profitability was itself, in some way, not legitimate. That the contractual purpose of the late payment fee was neither to punish late payment by Mr Paciocco nor to deter him from paying late can be seen by reflecting on the circumstance, noted by Allsop CJ in the Full Court, that the appellants "chose to run their affairs by risking the fees". There was no suggestion at all by the appellants that the exercise of that choice was forced on them by any circumstance beyond their control. Indeed, it was a choice from which Mr Paciocco could expect to derive some benefit if he was prepared, as he evidently was, to accept the risk that ANZ might terminate his accounts if it did not wish to retain his business. By choosing to incur the fees, Mr Paciocco made a rational decision to deploy his available funds to meet other claims on his resources. The rationality of that choice suggests that the amount of the fee was relatively modest, in that it was not of sufficient magnitude as to make the choice inconvenient for him as a matter of business. Alternatively, if Mr Paciocco actually had insufficient funds available to meet his payment obligations to ANZ, his choice to pay the bank late meant that he avoided the transaction costs, and the inconvenience (and probably higher interest rates), which he would have incurred had he chosen to arrange an alternative source of finance to enable him to make timely payment of the amount owing on the credit card account. There was no suggestion that Mr Paciocco was, for any reason, unlikely to be able to make such arrangements had it been in his own interests to do so. On either of these two scenarios, there is no reason to regard Mr Paciocco's choice to incur the fee as other than a rational economic choice on his part. A voluntary and self‑interested choice of this kind is the opposite of the rational response which one might expect to be generated by a penal provision, given that the characteristic purpose of a penalty is to deter non‑compliance. Given the importance of the values of commercial certainty and freedom of contract in the law, the courts will not lightly invalidate a contractual provision for an agreed payment on the ground that it has the character of a punishment. The existence of legislation such as that invoked to support the statutory claims made by the appellants means that it cannot be said that "an untrammelled 'freedom of contract' provide[s] a universal legal value". But in the application of common law rules, the maintenance of freedom of contract is of abiding importance, subject, of course, to statute. Thus, in AMEV‑UDC Finance Ltd v Austin, Mason and Wilson JJ said: "The courts should not … be too ready to find the requisite degree of disproportion lest they impinge on the parties' freedom to settle for themselves the rights and liabilities following a breach of contract." Only in cases where gross disproportion is such as to point to a predominant punitive purpose have agreed payments payable on breach of contract been struck down as penalties. Thus, for example, where that purpose is not discernible because the evaluation and assessment of the loss covered by the agreed payment is "very expensive and very difficult … to calculate precisely", the penalty rule has been held to have no application. It may be that other laws concerned with the unfair or unreasonable use of superior bargaining power will affect the validity of the provision; but, subject to such laws, the penalty rule is not engaged by a provision which achieves a profit for the promisee at the expense of the promisor. That is because, if the provision is not distinctly punitive in its character, the penalty rule does not operate to displace the parties' freedom to settle for themselves the contractual allocation of benefits and burdens and the rights and liabilities following a breach of contract. The leading case of Dunlop Pneumatic Tyre Co Ltd v New Garage and Motor Co Ltd, upon which the appellants principally relied, was itself a case where a clause providing for an agreed payment on any breach was upheld. Dunlop does not encourage invalidating provisions the purpose of which can objectively be seen to be the protection of the legitimate interests of one contracting party against default by the other. Nor does Dunlop suggest a narrow view of what interests may legitimately be protected by a provision for an agreed payment. It is certainly inconsistent with the notion that a contracting party's legitimate interests go no further than the performance of the principal terms of the contract. Finally by way of overview of the appellants' case, the level of interest charged by a bank, while reflecting market forces, may also be affected by the extent to which other means are deployed to cover the risks of the provision of financial accommodation and reward the bank for taking those risks. In this way, the rate of interest demanded of each customer might be expected, other things being equal, to be lower because an enforceable promise is taken from each customer to pay a late payment fee. Such a fee serves to reduce the overall risk assumed by the bank in providing the card facility to its mass of customers and to ensure the level of profitability acceptable to the bank's shareholders. The appellants' claim involves the disturbing irony that, if the challenge to the validity of the late payment fee were to succeed, it could be expected to have the consequence that the cost of financial accommodation to all customers, including those who honour their contractual engagements, will be increased in order to maintain ANZ's revenue streams at a level unaffected by the proscription of the late payment fee. ANZ could be expected to seek to ensure that its revenue streams are maintained; and its evident market power is such that there is no reason to doubt that it would succeed, at least to a large extent, in achieving that end. It was accepted that Mr Paciocco would not have received better terms from ANZ's competitors in the market. This state of affairs is consistent with the oligopolistic character of the market. Accordingly, if the late payment fees (which were relatively uniform among ANZ and its competitors) are unenforceable, interest rates or other charges could be expected to rise at the expense of those customers who adhere to their contractual engagements. That might not be thought to be a good thing. But however that may be, the mere prospect of such a consequence illustrates the danger of pressing the penalty rule into service for a purpose for which it is ill‑adapted. Background Mr Paciocco opened the first credit card account in June 2006, with a credit limit of $15,000, which was increased to $18,000 in November 2009. Mr Paciocco opened the second credit card account in July 2009 with a credit limit of $4,000. It will be convenient to refer to Mr Paciocco's two credit card accounts as "the card accounts". The terms and conditions of the card accounts (set out in various contractual documents identified by the primary judge) provided that an account holder was required to pay a minimum monthly payment plus any other amount immediately due by a certain date shown on the account holder's statement of account. Until December 2009, that date was within 28 days of the end of a statement period. From December 2009, that date was the "due date" on the statement of account. Interest was charged at a rate of 12.24 per cent on the outstanding amount if the full balance of the card account was not paid by the due date shown on the statement of account. The obligation to pay a minimum monthly amount was the primary payment stipulation in favour of ANZ. Until December 2009, in the event that the minimum monthly payment and any other amount immediately due were not paid within 28 days of the end of the statement period, a late payment fee of $35 was charged to the account holder's account. From December 2009, a late payment fee of $20 was charged to the account holder's account in the event that the minimum monthly payment and any other amount immediately due were not paid by the "due date". Until December 2009, eight late payment fees were charged to the card accounts. From December 2009, 18 late payment fees were charged to the card accounts. The decision of the primary judge The primary judge made a number of findings about the circumstances of Mr Paciocco's entry into the contracts for the card accounts. These findings have an important bearing upon both the penalty claims and the statutory claims. First, the terms of the card accounts "were contained in printed forms, which [Mr Paciocco] had no opportunity to negotiate." Secondly, at the time the card accounts were established, other banks charged late payment fees similar to that required by ANZ. It was common ground that ANZ determined the quantum of the late payment fees and that it did not determine that quantum by reference to a sum that would have been recoverable as unliquidated damages. The primary judge held that the contractual stipulation for a fee to be charged if the amount shown on the statement of account was not paid by a certain date was collateral to the primary stipulation in favour of ANZ that Mr Paciocco make a minimum monthly payment by a due date. The primary judge concluded that at law and in equity, the collateral stipulation was to be viewed as security for, or in terrorem of, the satisfaction of the primary stipulation. Her Honour stated the test whether such a stipulation was a penalty under the general law as follows: "[A] stipulation (to pay a sum or other property) will not constitute a penalty at law or in equity unless it is 'extravagant and unconscionable in amount in comparison with the greatest loss that could conceivably be proved'". Her Honour held that the evidence of Mr Inglis, an expert witness called by ANZ, did not assist in the application of that test because it reflected a "theoretical accounting" exercise. Mr Inglis was instructed to assess the maximum amount of costs that ANZ could conceivably have incurred as a result of late payments. Mr Inglis attempted to assess the cost of collecting the unpaid debt, the cost to ANZ of increasing its provisions for the greater risk of late payment or non‑payment or default in repayment by Mr Paciocco and the opportunity costs associated with the need to meet regulations relating to the bank's capital reserves. The primary judge held that "increase in loss provisions" and "increase in the costs of regulatory capital" were not loss or damage suffered as a result of Mr Paciocco's late payments. This was said to be because those matters were too remote to be compensable as loss or damage caused by the late payment. As her Honour put it: "provisions and regulatory capital are part of the costs of running a bank in Australia. No increase in them can be directly or indirectly related to any of the late payments by Mr Paciocco. As the [appellants] submitted, there are many cases each year by banks against customers and guarantors where the principal debtor has defaulted. In those cases, the banks seek damages limited to the sums outstanding, enforcement costs and interests. No one has suggested that a bank would be entitled to recover an increase in provisioning or the cost of its regulatory capital." On the other hand, the appellants' expert, Mr Regan, had been instructed to identify the damage actually suffered by ANZ as a result of the late payments by Mr Paciocco and the amounts needed to restore ANZ to the position it would have been in had the late payments not occurred. The primary judge adopted Mr Regan's methodology and found that, while the actual loss suffered by ANZ as a result of the late payments could not be precisely determined, it was probably no more than $3 per late payment event; and, on any view, considerably less than $35 or $20. On that basis, the primary judge concluded that the late payment fees were extravagant and unconscionable and therefore penalties. The decision of the Full Court The principal judgment in the Full Court was written by Allsop CJ, with whom Besanko and Middleton JJ agreed generally in separate judgments. Allsop CJ held that the primary judge erred in conflating two enquiries: "the anterior enquiry as to whether the provision is penal in character; and the later enquiry as to the remedial consequence of any such characterisation." Allsop CJ said that the two enquiries were separate, and that: "one looks forward and is referable to the time of entry into the contract and the extent of the legitimate interest of the obligee in the performance of the relevant provision of the contract; the other looks backwards to see what damage has been demonstrated to have been caused by the breach or failure of the relevant provision in order to found some relief for such breach or failure." Accordingly, identifying the primary judge's error, Allsop CJ said that: "to the extent that the primary judge is to be taken to have undertaken an ex post enquiry of actual damage as a step in assessing whether the prima facie penal character of the late payment fee was rebutted (as a reading of all the reasons requires) her Honour, in my respectful view, was wrong." Allsop CJ held that, in assessing whether a stipulation is a penalty, "the correct approach [is] to look at the greatest possible loss on a forward looking basis". As his Honour explained: "The question as to whether [a stipulation] is penal is to be assessed by reference to the question whether it is extravagant or exorbitant by reference to the obligee's legitimate interest in the performance of the contract assessed by the greatest loss that could conceivably be proved to have followed from a breach or failure to comply." The Full Court regarded Mr Regan's evidence of the damage actually suffered by ANZ as a result of the late payments by Mr Paciocco as irrelevant to the question whether the late payment fees were extravagant and unconscionable by reference to the greatest loss that could conceivably be proved to have followed from the breach. Allsop CJ said that the enquiry undertaken to answer that question "is not assisted by knowing what the damages from [the late payments] were." Rather, the Full Court regarded Mr Inglis' forward‑looking approach as correct. The Full Court accepted Mr Inglis' evidence as identifying three categories of cost which might conceivably be incurred as a result of late payment. First, ANZ was required to take up a loss provision on its profit and loss account when customers fail to meet a monthly repayment obligation on a credit card account. Secondly, it was required to hold regulatory capital sufficient to cover unexpected losses: as the risk of non‑recovery of its loan assets increases, as upon the failure of an account holder to make a minimum monthly payment, the amount of regulatory capital that ANZ is required to hold increases. Thirdly, there was the cost of collection activity to recover the amounts due when a customer fails to meet a monthly payment. The Full Court held, contrary to the primary judge, that it was appropriate to take into account the three categories of costs in assessing the extravagance or otherwise of the late payment fees because they were "legitimate business cost[s]" and "part of the costs of running a bank". It is apparent that the Full Court resolved the difference between the evidence of Mr Inglis and Mr Regan without descending into close consideration of the accuracy of the calculations made by either witness. The primary judge rejected Mr Inglis' estimate as a theoretical calculation unconstrained by the rules which limit the damages recoverable by suit for breach of contract, while the Full Court proceeded on the basis that the exercise performed by Mr Inglis was to be preferred, being the exercise required to indicate the greatest loss that could conceivably be proved to have followed from the breach, an exercise which was none the worse for being "theoretical". The difference between the Full Court and the primary judge turned on the question whether this theoretical exercise was consistent with the penalty rule's test of the character of a stipulation. The Full Court was not invited to reject Mr Inglis' approach on the basis that his calculations were unreliable. To the extent that the appellants' argument in this Court sought to advance such a contention, it should not be entertained. The Full Court cannot be said to have erred in failing to accept an argument that was not put to it; and, in any event, even if one puts Mr Inglis' calculations to one side, Mr Regan's evidence does not support the conclusion that the late payment fee was a penalty. Mr Regan's evidence proceeds on too narrow a view of the legitimate interests of ANZ protected by the late payment fee. If the expert evidence on either side does not show that the late payment fee was so far out of proportion to the effect upon the legitimate interests associated with the bank's business that its purpose was punitive, then the appellants' case must fail. For the sake of completeness, it should be noted that Allsop CJ also dealt with the admission by ANZ in its amended defence that: "ANZ did not determine the quantum of each [of the late payment fees] by reference to a sum that would have been recoverable as unliquidated damages." Allsop CJ held that an admission that the fee was not, in fact, a pre‑estimate of the sum which might have been recovered as damages for breach did not necessarily mean that the fee in question was a penalty; that determination could only be made in light of all of the "contemporaneous acts and approaches" of the parties to the relevant contract. It may be said immediately that the Full Court was clearly correct in this regard. As was said in Cavendish Square Holding BV v Makdessi by Lord Neuberger of Abbotsbury PSC and Lord Sumption JSC: "[T]he penal character of a clause depends on its purpose, which is ordinarily an inference from its effect. … [T]his is a question of construction, to which evidence of the commercial background is of course relevant in the ordinary way. But, for the same reason, the answer cannot depend on evidence of actual intention." To say, as did Lord Dunedin in Dunlop, that the question whether a provision is a penalty is a question of contractual "construction" is to say that the question is one of identifying the legal character of the provision from the effect of its terms in the commercial context in which it is to operate. The characterisation of a stipulation as a "genuine covenanted pre‑estimate of damage" is a legal question which does not depend upon an evidentiary enquiry into the parties' motivation or subjective intention, purpose or calculations. As Deane J observed in O'Dea v Allstates Leasing System (WA) Pty Ltd, a payment stipulation may be characterised as a penalty notwithstanding that the parties "subjectively intended to make a pre‑estimate of damages in the event of breach." The appellants' submissions as to penalty The appellants argued that the Full Court erred in assessing the greatest loss that ANZ could conceivably have suffered as a result of the late payments by taking into account heads of loss that would not be compensable at law in an action for damages for breach of contract. In this regard, the appellants contended that the Full Court erred in holding that evidence of the actual damage suffered by ANZ was irrelevant to whether the late payment fees were penalties; and in holding that some of the costs said to have been incurred by ANZ as a result of the late payments were relevant to the assessment of whether the late payment fees were penalties. As to the first of these contentions, it may be accepted that evidence of the actual costs incurred by ANZ by reason of the late payments was not irrelevant. Mr Regan's evidence may have assisted in making the case that the late payment fees did not reflect a genuine pre‑estimate of damage if the scope of that exercise were confined to the recoupment of the expenditures necessitated by each particular default. But to show what ANZ actually spent on each occasion of default by Mr Paciocco is not to show what damage might conceivably have been suffered to the interests ANZ was entitled to protect. The second of the appellants' contentions should not be accepted because the view advanced by the appellants of the legitimate interests that ANZ may seek to protect by the late payment fee is too narrow. The acceptance of such a view would broaden the scope of the penalty rule to encompass provisions which are outside its rationale. The penalty rule: a rule in search of a rationale The penalty rule is of ancient but somewhat uncertain origin. In Cavendish Square, Lord Neuberger and Lord Sumption, following the view of Professor Simpson, identified the origin of the penalty rule in the "equitable jurisdiction to relieve from defeasible bonds." The concern of equity was that no more should be recovered from the defaulting party than was necessary to ensure that the innocent party's interest in performance was not diminished. On that approach, a provision which was ancillary to or security for the "substance" of the transaction would not be enforced in equity so long as a defaulting party seeking relief was ready, willing and able to perform the substance of that party's obligations. As Lord Macclesfield said in Peachy v Duke of Somerset, equitable relief was founded on "the original intent of the case, where the penalty is designed only to secure money, and the Court gives him all that he expected or desired". An ancillary or security provision could not be enforced in equity at all to the extent that it went beyond the substance of the transaction. In the 17th and 18th centuries, the courts of common law sought to obviate the need for separate proceedings in Chancery by adopting the penalty rule. As Tindal CJ observed in Smith v Bond, this process was assisted by legislation whose object was to "take away the necessity of applying for relief to a court of equity". Since the work of Professor Simpson to which Lord Neuberger and Lord Sumption referred in Cavendish Square, Professor Biancalana has traced the common law's engagement with penalty clauses back to the 13th century, noting that the penalty rule originated as an aspect of the jurisdictional tussle between the ecclesiastical courts and the courts of common law. According to Professor Biancalana, the common law courts were, at first, disposed to uphold penalty clauses as lawful, but by the turn of the 14th century had come to regard them as objectionable on the basis that they were a form of usury, which was unacceptable to medieval Christianity. It may also be said that the development of the law reflecting the Church's disapproval of usury was aligned with the economic interests of the dominant political class, the landed aristocracy, who, asset rich but cash poor, were chronically disinclined to keep their contractual engagements to those who had the recurring misfortune to have lent them money. Some modern statements of the penalty rule are couched in terms of robust commerciality rather than in terms of a concern about usurious charges by lenders or the nice adjustments of equity. In Ringrow Pty Ltd v BP Australia Pty Ltd, this Court, in a unanimous judgment, said: "The law of contract normally upholds the freedom of parties, with no relevant disability, to agree upon the terms of their future relationships. ... Exceptions from that freedom of contract require good reason to attract judicial intervention to set aside the bargains upon which parties of full capacity have agreed. That is why the law on penalties is, and is expressed to be, an exception from the general rule. It is why it is expressed in exceptional language. It explains why the propounded penalty must be judged 'extravagant and unconscionable in amount'. It is not enough that it should be lacking in proportion. It must be 'out of all proportion'." This passage emphasises that the rule against penalties operates as an exception to the primacy otherwise accorded to considerations of certainty and freedom of contract where neither party is under a relevant disability. And the exceptional nature of the rule, in turn, invites close scrutiny of its rationale. 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. Nor is the function of the penalty rule adequately explained by the concerns which led courts of equity to make adjustments to ensure that both parties obtained what equity saw as the "substance" of their transaction and no more in cases within its jurisdiction. The Supreme Court of the United Kingdom in Cavendish Square conceived of the penalty rule as serving a purpose which stands apart from its equitable roots. According to Lord Neuberger and Lord Sumption, the rule against penalties operates on grounds other than that the putative penalty clause is regarded as a "matter of substance" as mere security for the performance of a principal obligation. As their Lordships explained: "Because [clauses which provided for payment of a specified sum in place of common law damages] were a contractual substitute for common law damages, they could not in any meaningful sense be regarded as a mere security for their payment." In Andrews v Australia and New Zealand Banking Group Ltd, this Court was not concerned with the late payment fee, but with other exception fees which were not charged upon breach by the customer. Given the issues before it, the Court, not surprisingly, stated the penalty rule in terms which reflect the ongoing influence of its equitable origins in cases where the impugned payment is charged otherwise than upon breach of contract, and held that the penalty rule may operate in the absence of a breach of contract. On that point, this Court's decision in Andrews was not followed by the Supreme Court of the United Kingdom in Cavendish Square. But that difference does not matter, either for the resolution of this case, which is a case of breach of contract, or for the identification of the rationale for the operation of the penalty rule in a case of breach of contract. 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. On that point, this Court and the Supreme Court of the United Kingdom are in accord. Courts of common law have long exercised the power to award exemplary damages to punish a tortfeasor in certain circumstances; but the courts have consistently refused to countenance the enforcement of attempts to impose punishment by contract as a sanction for non‑performance or to threaten such punishment. As Lord Hoffmann said in Co‑operative Insurance Society Ltd v Argyll Stores (Holdings) Ltd: "the purpose of the law of contract is not to punish wrongdoing but to satisfy the expectations of the party entitled to performance." Earlier, in Dunlop, Lord Parmoor had described a penalty as a "sum inserted as a punishment on the defaulter"; and in Australia in Legione v Hateley, Mason and Deane JJ described a penalty as a punishment for non‑observance of a contractual provision by the imposition of an additional or different liability. That description was approved in the unanimous judgment of this Court in Andrews. Similarly, in Cavendish Square, Lord Neuberger and Lord Sumption said of the observation in Legione by Mason and Deane JJ: "[I]n so far as it refers to 'punishment' and 'an additional or different liability' as opposed to 'in terrorem' and 'genuine pre‑estimate of loss', this definition seems to us to get closer to the concept of a penalty than any other definition we have seen. The real question when a contractual provision is challenged as a penalty is whether it is penal, not whether it is a pre‑estimate of loss. These are not natural opposites or mutually exclusive categories." To similar effect, this Court in Ringrow rejected the suggestion that the impugned provision must be proportional to the legitimate commercial interests of the party relying upon it in order to avoid being characterised as a penalty. It is only where the impugned provision requires a payment upon breach which is out of all proportion to the legitimate commercial interests of the party relying upon it that the punitive character of the provision stands revealed. "In terrorem" Courts of equity regarded a collateral provision designed to provide an incentive to perform a principal obligation as objectionable on the ground that its enforcement was unnecessary to give the promisee the benefit of the substance of the transaction. Such a collateral provision might be described as operating "in terrorem", because of its evident tendency to deter the promisor from non‑performance. In Campbell Discount Co Ltd v Bridge, Lord Radcliffe said in relation to the use of the phrase "in terrorem" as identifying a purpose characteristic of a penalty: "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' … I do not find that that description adds anything of substance to the idea conveyed by the word 'penalty' itself, and it obscures the fact that penalties may quite readily be undertaken by parties who are not in the least terrorised by the prospect of having to pay them and yet are, as I understand it, entitled to claim the protection of the court when they are called upon to make good their promises." On the other hand, if one regards the function of the penalty rule as being to preclude contracting parties from imposing punishment for breach of contract, it is not difficult to accept that a clause which has a deterrent effect by virtue of the prospect of punishment with which it confronts a defaulting promisor should be characterised as a penalty. In this sense, it may not be unhelpful to use the phrase "in terrorem". Dunlop revisited The preceding discussion of the rationale of the penalty rule affords a contemporary framework for a discussion of Lord Dunedin's statements in Dunlop upon which the appellants' case was based. Lord Dunedin was concerned to offer guidance in drawing the distinction between an agreed payment and a penalty. His Lordship said: "2. The essence of a penalty is a payment of money stipulated as in terrorem of the offending party; the essence of liquidated damages is a genuine covenanted pre-estimate of damage. 3. The question whether a sum stipulated is penalty or liquidated damages is a question of construction to be decided upon the terms and inherent circumstances of each particular contract, judged of as at the time of the making of the contract, not as at the time of the breach. 4. To assist this task of construction various tests have been suggested, which if applicable to the case under consideration may prove helpful, or even conclusive. Such are: (a) It will be held to be penalty 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. (b) It will be held to be a penalty if the breach consists only in not paying a sum of money, and the sum stipulated is a sum greater than the sum which ought to have been paid. … (c) There is a presumption (but no more) that it is penalty when 'a single lump sum is made payable by way of compensation, on the occurrence of one or more or all of several events, some of which may occasion serious and others but trifling damage'. On the other hand: (d) It is no obstacle to the sum stipulated being a genuine pre-estimate of damage, that the consequences of the breach are such as to make precise pre-estimation almost an impossibility. On the contrary, that is just the situation when it is probable that pre‑estimated damage was the true bargain between the parties." (citations omitted) Four points may be made here in relation to his Lordship's propositions. First, it is to be noted that proposition 4(b) harks back to the classic case of a bond upon condition to secure the payment of a lesser sum by a covenant to pay a greater sum. The more general run of cases, where breach engages an obligation to pay a specified sum, is addressed by Lord Dunedin's proposition 4(a). That this is so may also be seen by reference to the observations of Lord Parmoor in Dunlop. His Lordship said: "There are two instances in which the Court has interfered when the agreed sum is referable to the breach of a single stipulation. ... The agreed sum, though described in the contract as liquidated damages, is held to be a penalty if it is extravagant or unconscionable in relation to any possible amount of damages that could have been within the contemplation of the parties at the time when the contract was made. ... ... The second instance in which the Courts have sanctioned interference is in the case of a covenant for a fixed sum, or for a sum definitely ascertainable, and where a larger sum is inserted by arrangement between the parties, payable as liquidated damages in default of payment. Since the damage for the breach of covenant is in such cases by English law capable of exact definition, the substitution of a larger sum as liquidated damages is regarded, not as a pre‑estimate of damage, but as a penalty in the nature of a penal payment." The late payment fee is not within Lord Parmoor's "second instance" or Lord Dunedin's proposition 4(b). As Allsop CJ noted, "[t]he fee may or may not, in fact, be greater than the sum due; [but] that does not appear on the face of the provision, or from an understanding of the facts." Thus, the late payment fee was not necessarily a demand for payment of a larger sum upon failure to pay a smaller sum. The appellants' attempt to rely upon Lord Dunedin's proposition 4(b) is out of step, not only with the rationale of the penalty rule applicable in cases of breach of contract, but with authority. It is proposition 4(a), not proposition 4(b), on which the appellants' case must depend. Secondly, as Mason CJ and Wilson J observed in Hungerfords v Walker, "legal and economic thinking about the remoteness of financial and economic loss have developed markedly in recent times." This observation has much force. It was only in that case, decided in 1989, that Australian jurisprudence finally accepted the (now obvious) economic reality that to be kept out of money due is to suffer real economic loss so that damages should be recoverable in tort for loss of the use of money. More recent decisions have recognised the nature of the consequences for a lender of a default by a borrower in a payment obligation, and that these consequences extend beyond the mere fact of non‑payment of the sum due on the due date. In Lordsvale Finance Plc v Bank of Zambia, Colman J observed that: "the borrower in default is not the same credit risk as the prospective borrower with whom the loan agreement was first negotiated. … [M]oney is more expensive for a less good credit risk than for a good credit risk". The common law's relatively recent acceptance of the economic reality that risky credit is more expensive credit has been accompanied by an appreciation of the nature of the relationship between the greater financial risk assumed by a bank by reason of late payments by customers and the costs to the bank's revenue stream associated with that increased risk. In short, if the adverse effects of late payment upon the bank's revenue stream are not covered by a late payment fee, those expenses can be expected to be covered by other charges, or by way of higher interest rates, imposed across the class of bank customers who use the same lending facility as the contract breaker. In this regard, in the United States, the Court of Appeals for the Second Circuit in Citibank NA v Nyland (CF8) Ltd, referring to earlier authority in that Court which rejected the assertion that an uplift in interest rates upon default was a penalty, said: "[D]ebtors might fare worse … if creditors were not allowed to impose variable rates, because creditors would then impose higher rates for the full life of the loan in order to reallocate the risk." Thirdly, à propos of Lord Dunedin's proposition 4(c), which states a "presumption" (albeit a weak one) that a single lump sum payable on the occurrence of one of several breaches of differing levels of seriousness is a penalty, it may be said that Mr Paciocco's choice to run his affairs by risking the fees affords a practical demonstration that the fixed quantum of the fee was sufficiently modest in amount that it was not apt, in the circumstances of its contemplated operation, to have an effect in terrorem of Mr Paciocco. It is useful here to refer, by way of analogy, to ParkingEye Ltd v Beavis (Consumers' Association intervening). In that case, the Supreme Court of the United Kingdom upheld the validity of a provision with the effect that the maximum permitted stay in the defendant's shopping centre car park was two hours with free parking during that period, but that £85 would be charged to those who stayed longer, reducible to £50 if paid within 14 days. Lord Neuberger and Lord Sumption held that: "although [the car park operator] was not liable to suffer loss as a result of overstaying motorists, it had a legitimate interest in charging them which extended beyond the recovery of any loss." Overstaying motorists reduced the parking spaces available to other customers of the retail outlets in the shopping centre. That created a situation which was apt adversely to affect the goodwill of the business of the operator of the car park by inconveniencing other users of the car park as would‑be customers of the shopping centre. Their Lordships drew the inference that overstayers "must regard the risk of having to pay £85 for overstaying as an acceptable price for the convenience of parking there." In the present case, as noted above, Mr Paciocco freely risked incurring the late payment fee as a matter of his own convenience. It can be inferred that having to pay the fee was, in the commercial context in which the fee was to operate, an acceptable cost of avoiding the expense and inconvenience of meeting his obligations as to timely payment of his account. That such an inference is available as to the operation of the provision in its commercial context is inconsistent with a purpose of punishment for breach. Fourthly, Lord Dunedin's summary was meant as a guide to the application of the rule. His Lordship's propositions were not intended to be applied as if they were the provisions of a statute. The terms "extravagant" and "unconscionable" in Lord Dunedin's proposition 4(a) are not used in contradistinction to reasonable, much less as free‑standing criteria of invalidity. In proposition 4(a), the terms "extravagant" and "unconscionable" function as pointers towards the punitive purpose which imbues the challenged provision with the character of a punishment. And as Lord Neuberger and Lord Sumption said in Cavendish Square, "the real question" is whether the impugned provision "is penal, not whether it is a pre‑estimate of loss." In Cavendish Square, Lord Neuberger and Lord Sumption accepted that, although there was "a case" for judicial abrogation of the penalty rule on the ground that it is "antiquated, anomalous and unnecessary, especially in the light of the growing importance of statutory regulation", the penalty rule should be retained because it serves the useful purpose of preventing a party from exercising a remedy where "the adverse impact of [the remedy] on the defaulter significantly exceeds any legitimate interest of the innocent party." Lord Neuberger and Lord Sumption considered that the courts can avoid inappropriate application of the penalty rule: "(i) by [proceeding upon] a realistic appraisal of the substance of contractual provisions operating on breach, and (ii) by taking a more principled approach to the interests that may properly be protected by the terms of the parties' agreement." In Andrews, this Court summarised the "critical issue" as being "whether the sum agreed was commensurate with the interest protected by the bargain." This Court's discussion in Andrews of the decision in Dunlop focused upon the reasons of Lord Atkinson, who accepted that an agreed payment upon breach should not be unenforceable where, though it "appeared imprecise as a pre‑estimate of damage, it protected the [seller's] interest in preventing undercutting, which would disorganise its trading system". Accordingly, the question to be addressed in order to distinguish a penalty from a provision protective of a legitimate interest is: "whether the sum or remedy stipulated as a consequence of a breach of contract is exorbitant or unconscionable when regard is had to the innocent party's interest in the performance of the contract." ANZ's legitimate interest It may be accepted immediately that a bank, like any other party to a contract, has no legitimate interest in punishing its customers for their defaults or in threatening them with punishment in order to discipline their behaviour. But a bank has a multi‑faceted interest in the timely performance of its customers' obligations as to payment. The legitimate interest of ANZ protected by the late payment fee cannot be apprehended without an understanding of the commercial context in which that interest requires protection. It is an awkward irony of the appellants' case that, in a class action representing many claimants unified in their assertion that this aspect of the contract of adhesion made by each of them with the bank is unenforceable as a penalty, the appellants' focus was upon the individual contract between Mr Paciocco and ANZ. Lord Dunedin in Dunlop recognised that the supposedly penal character of the operation of a contractual term must be understood by reference to the "inherent circumstances" in which it is to operate. In the present case, those circumstances include the fact that ANZ is a bank engaged in providing financial accommodation by way of bank card facilities to many customers on standard terms. The agreement, common to each customer, to pay the late payment fee required by ANZ was an aspect of ANZ's business which enabled the bank to provide accommodation to each customer. The fixing of risk and reward on each side of each transaction reflected the circumstance that it was one of many transactions and that the very multiplicity of these transactions was a factor bearing upon the pricing of each facility to each of many customers. The circumstance that the value of an increase in credit risk may be difficult to assess is a consideration which tends against an affirmative conclusion that the stipulation is to be characterised as a punishment. That consideration gains added force where the creditor is a bank which, as such, is exposed to the risk that many borrowers may default at one time in circumstances unforeseen at the time of the original arrangements. One aspect of ANZ's interest is obvious as an ordinary aspect of the business of a bank. Because of the relationship between the financial risks assumed by a lender and the cost of the facility to customers, the economic effect of provisions calculated to secure the protection of the bank by a late payment clause cannot be viewed in isolation from other elements of the cost of the facility to borrowers, such as interest. The Supreme Court of the United States in Smiley v Citibank (South Dakota) NA held that late payment fees charged by a bank on its credit card accounts were not unenforceable penalties. The case turned upon the Court's acceptance of the bank's argument that the late payment fees could be characterised as "interest" which the bank was permitted to charge under what the Court regarded as an ambiguous provision of the National Bank Act of 1864. The Court held that it was reasonable to interpret the statutory term "interest" as including: "any payment compensating a creditor … for an extension of credit … or any default or breach by a borrower of a condition upon which credit was extended. It includes, among other things, the following fees connected with credit extension or availability: numerical periodic rates, late fees, not sufficient funds … fees, overlimit fees, annual fees, cash advance fees, and membership fees." The Court held that it was: "perfectly possible to draw a line … between (1) 'payment compensating a creditor or prospective creditor for an extension of credit, making available of a line of credit, or any default or breach by a borrower of a condition upon which credit was extended,' and (2) all other payments. To be sure, in the broadest sense all payments connected in any way with the loan - including reimbursement of the lender's costs in processing the application, insuring the loan, and appraising the collateral - can be regarded as 'compensating [the] creditor for [the] extension of credit.' But it seems to us quite possible and rational to distinguish … between those charges that are specifically assigned to such expenses and those that are assessed for simply making the loan, or for the borrower's default." Underpinning the Supreme Court's liberal construction of the statute regulating bank lending is an appreciation that a late payment fee is part of the compensation for the risk assumed by the bank in making the facility available to the customer; "compensation" in this discourse being used, not in the sense of making good a loss, but in the sense of reward for risk. Maintaining or enhancing that reward is part of the legitimate business of a bank. Another legitimate interest which is an ordinary aspect of the business of a bank is the freedom the bank obtains, by timely repayment by its customers, to pursue more profitably its business of lending to its customers than would be the case if it is constrained to take into account the effect of defaulting customers upon its revenues. This interest is not limited to the loss of the opportunity to profit by re‑investing funds paid late. If a bank's customers comprised only borrowers who paid on time in accordance with their contractual arrangements, the bank's freedom from the risks associated with late payment would enable it to maximise its revenues by reducing the cost of its facilities to all its customers in order to secure more customers and hence higher revenues. Some lenders may seek to improve their profits by deliberately soliciting the business of chronic defaulters, but it would be distinctly naïve to think that those borrowers would not be required to pay for a less attractive credit rating by higher interest rates. Accordingly, it is worth reiterating that even if one were to put the evidence of Mr Inglis to one side, the evidence of Mr Regan would not support the conclusion that the late payment fee was punitive in character. Mr Regan's evidence was not apt to demonstrate the gross disproportion required to establish the punitive character of the late payment fee because it did not address the full range of ANZ's legitimate interests protected by the late payment fee. Further, for reasons now to be discussed, the primary judge erred in treating characterisation of the late payment fee as turning upon a comparison between the quantum of the fee and the amount that might have been recovered in an action for damages. Pre‑estimate of damage or recoverable damages The appellants argued that the Full Court misapplied Lord Dunedin's proposition 4(a). They noted that the expression used in Dunlop - "the greatest loss that could conceivably be proved to have followed from the breach" - speaks only of damages recoverable by action in consequence of a breach of contract and not of loss which is too remote to be compensated by an award of damages. The appellants' submission was that it is not the case that any set of circumstances resulting in loss that might be hypothesised can be taken into account in assessing whether a clause is penal. The appellants submitted that the primary judge was correct to reject ANZ's submission that "increase in loss provisions" and "increase in the cost of regulatory capital" were losses incurred as a result of late payment and to conclude that these matters were too remote to form part of compensable damage. The appellants also argued that, to the extent that the Full Court held that collection costs exceeding the amount of the late payment fee could be taken into account, that finding should be set aside. It is not the case that authoritative formulations of the test of punitive extravagance are invariably stated in terms of a comparison between the impugned payment and what might be recovered by litigation. This Court said in Ringrow: "The law of penalties, in its standard application, is attracted where a contract stipulates that on breach the contract‑breaker will pay an agreed sum which exceeds what can be regarded as a genuine pre-estimate of the damage likely to be caused by the breach." It will be noted that this formulation of the rule does not depend upon a demonstration of the quantum of damages which would be recovered in an action. To speak of damage, as opposed to damages, is to speak of the loss caused by the breach, not the remedy which might be awarded by a court. To speak of a "genuine pre‑estimate of the damage" is to speak of the damage liable to be suffered by those parts of the bank's legitimate business interest that it is trying to protect. A genuine pre‑estimate of that damage may encompass items of loss actually suffered, albeit too remote to be compensable by way of damages by virtue of the rules in Hadley v Baxendale. An agreement for the recovery of such loss is consistent with the absence of a punitive purpose. For a party to stipulate for a more ample remedy than is available at law is not to visit a punishment on the other party. An agreed provision avoids the uncertainty and expense of litigation. The benefit of such a provision to both parties and to the legal system is obvious. Even in medieval times, the authority we know as Bracton considered that a clause providing for the payment of an agreed sum upon a breach of contract served the legitimate purpose of removing the uncertainty and expense of litigation involved in establishing the quantum of damages recoverable for a breach of contract. In summary as to appeal M220 of 2015, the appellants' claim that the late payment fees were penalties fails. I turn to consider the statutory claims. The statutory claims Three statutory regimes were invoked by the appellants: (a) a regime which prohibited "unconscionable conduct" in connection with the supply of financial services, operated concurrently by s 12CB of the Australian Securities and Investments Commission Act 2001 (Cth) ("the ASIC Act") and s 8 of the Fair Trading Act 1999 (Vic) ("the FTA"), until the latter was replaced by the Australian Consumer Law ("the ACL"); (b) a regime which regulated the provision of credit and allowed for the reopening of "unjust transactions" (the term "unjust" including "unconscionable, harsh or oppressive") under s 76 of the National Credit Code, in force pursuant to the National Consumer Credit Protection Act 2009 (Cth); and (c) a regime which rendered void "unfair" contractual terms pursuant to Pt 2B of the FTA and ss 12BF and 12BG of the ASIC Act. Each of the regimes did not apply throughout the whole of the period to which the statutory claims relate, and the regimes were subject to amendment. However, nothing turns on that fact. Some general observations by way of overview of the statutory claims will aid an understanding of what follows. First, as the primary judge noted, there was no allegation of any dishonesty or abuse of market power by ANZ, or that ANZ concealed the requirements of the late payment fee from Mr Paciocco, or that he was unable to understand the effect of the contracts in that regard, or that he entered into the contracts as the result of the exercise of financial pressure placed upon him by ANZ. The primary judge also noted that Mr Paciocco was under no obligation to use the card accounts and was free to terminate them should he so choose. Importantly, the primary judge observed that: "Mr Paciocco did not contend that there was anything unusual or exceptional in the manner in which the card accounts were entered into or in their terms. On the contrary, it was common ground that similar terms were offered by ANZ's competitors." (citations omitted) Given that the appellants did not suggest that ANZ dealt with Mr Paciocco in any way less favourably than he would have been treated by any other supplier of credit card facilities, and in the absence of an allegation that the market in which this state of affairs prevailed was itself brought about by unlawful conduct, or an allegation that Mr Paciocco was driven to agree to ANZ's terms by financial pressures of which the bank was aware, the appellants' statutory claims take on an air of unreality. The appellants seek to stigmatise as unconscionable or unfair or unjust an activity in the marketplace in which nothing materially distinguishes the situation and conduct of either Mr Paciocco or ANZ from any of the other participants in that activity. It may be said that ANZ and its competitors have dealt "unconscionably" or "unfairly" or "unjustly" with all of their customers in that, in a careless or partisan use of language, all banks may be said to do so as a matter of course. But to argue that conduct by one participant in a market, which is an unremarkable example of conduct engaged in by all participants in that market, is unconscionable, or unjust or unfair, in breach of the statutory norms, without any suggestion that the market itself is unlawfully skewed, is something of a stretch. And the argument falls distinctly short of the mark in this particular case, given that there was no suggestion that Mr Paciocco was driven to seek the card accounts as a result of "pressure put upon [him] by [his] poverty", or that Mr Paciocco incurred the late payment fees as a result of financial difficulties which prevented him from making timely payment in accordance with his contractual obligations. As has been seen, Mr Paciocco chose to pay late, and thereby incur the late payment fee, as a matter of his own convenience. In these circumstances, it would have been surprising if either Court below had held that ANZ took advantage of Mr Paciocco in a way which would meet the statutory descriptions of "unconscionable conduct", "unjust transactions" or "unfair terms". Unconscionable conduct Allsop CJ rejected the "gravamen" of the appellants' attack on ANZ's conduct, which depended upon what was said to be the "huge disparity between the level of the fees and the costs [ANZ] sustained by the exception fee events." His Honour concluded that: "In all the circumstances, in particular, the lack of any proven predation on the weak or poor, the lack of real vulnerability requiring protection, the lack of financial or personal compulsion or pressure to enter or maintain accounts, the clarity of disclosure, the lack of secrecy, trickery or dishonesty, and the ability of people to avoid the fees or terminate the accounts, I do not consider the conduct of ANZ to have been unconscionable. To do so would require the court to be a price regulator in banking business in connection with otherwise honestly carried on business in which high fees were extracted from customers." The appellants submitted that s 12CB of the ASIC Act was introduced to address "the general disparity of bargaining power" between financial services providers and consumers. That submission may be accepted as far as it goes; but it does not go very far. While a disparity in bargaining power may be necessary to attract the operation of the provision, the mere existence of the disparity is not sufficient to do so. The existence of a disparity in bargaining power, which is an all‑pervading feature of a capitalist economy, does not establish that the party which enjoys the superior power acts unconscionably by exercising it. In this latter regard, s 12CB of the ASIC Act, as in force immediately prior to 1 January 2012, relevantly provided: "(1) A person must not, in trade or commerce, in connection with the supply or possible supply of financial services to a person, engage in conduct that is, in all the circumstances, unconscionable. (2) Without limiting the matters to which the court may have regard for the purpose of determining whether a person (the supplier) has contravened subsection (1) in connection with the supply or possible supply of services to a person (the consumer), the court may have regard to: (a) the relative strengths of the bargaining positions of the supplier and the consumer; and (b) whether, as a result of conduct engaged in by the supplier, the consumer was required to comply with conditions that were not reasonably necessary for the protection of the legitimate interests of the supplier; and (c) whether the consumer was able to understand any documents relating to the supply or possible supply of the services; and (d) whether any undue influence or pressure was exerted on, or any unfair tactics were used against, the consumer or a person acting on behalf of the consumer by the supplier or a person acting on behalf of the supplier in relation to the supply or possible supply of the services; and (e) the amount for which, and the circumstances under which, the consumer could have acquired identical or equivalent services from a person other than the supplier. ... (4) For the purpose of determining whether a person has contravened subsection (1) in connection with the supply or possible supply of financial services to another person: (a) the court must not have regard to any circumstances that were not reasonably foreseeable at the time of the alleged contravention; and (b) the court may have regard to conduct engaged in, or circumstances existing, before the commencement of this section." The appellants' argument focused upon s 12CB(2)(a) and (b) of the ASIC Act without regard to the other provisions which may be relevant. The argument that the Full Court should have concluded that the fee was unconscionable on the basis that it was not set at an amount limited to cost recovery only must be rejected because of its erroneously narrow assumption as to the legitimate interests of ANZ. Further, to focus upon the relative strengths of the bargaining positions of Mr Paciocco and ANZ is to ignore the requirement of s 12CB(1) to consider "all the circumstances". Section 12CB(1) does not proscribe the existence of a disparity in bargaining power as opposed to the manner of its exercise. And, as has been noted, nothing in the manner of ANZ's exercise of its superior bargaining strength fell foul of the other provisions of s 12CB(2). Unjust transactions and unfair contract terms The appellants argued that the late payment fees were unjust under s 76 of the National Credit Code. Section 76(1) of the National Credit Code provides: "The court may, if satisfied on the application of a debtor, mortgagor or guarantor that, in the circumstances relating to the relevant credit contract, mortgage or guarantee at the time it was entered into or changed (whether or not by agreement), the contract, mortgage or guarantee or change was unjust, reopen the transaction that gave rise to the contract, mortgage or guarantee or change." Section 76(2) provides that, in determining whether a term is unjust, the court must have regard to the public interest and to all the circumstances of the case. Section 76(2)(a) to (p) also provide that the court may have regard to a range of matters, including "any … relevant factor", in making that determination. A close reading of s 76(2) of the National Credit Code reveals significant overlap between that sub‑section and s 12CB(2) of the ASIC Act. In such circumstances, the appellants' focus upon s 76(2)(e) of the National Credit Code was once again too narrow. Section 76(2)(e) provides that the court may have regard to: "whether or not any of the provisions of the contract, mortgage or guarantee impose conditions that are unreasonably difficult to comply with, or not reasonably necessary for the protection of the legitimate interests of a party to the contract, mortgage or guarantee". As has already been noted, reliance on this sub‑section reflected the appellants' misapprehension of the legitimate interests that ANZ was entitled to protect. The appellants also argued that the late payment fees were unfair contract terms within the meaning of ss 32W and 32X of the FTA and s 12BG of the ASIC Act. At the relevant time, s 32W of the FTA provided: "A term in a consumer contract is to be regarded as unfair if, in all the circumstances, it causes a significant imbalance in the parties' rights and obligations arising under the contract to the detriment of the consumer." Section 32X of the FTA set out matters which could be taken into account by the court in determining whether a term of a consumer contract was unfair. The Full Court held that the late payment fees were neither "unjust" nor "unfair". Having already considered all of the circumstances at the time of Mr Paciocco's entry into the credit card contracts, and the effects of the terms of those contracts, the reasons of the Full Court in this respect were unsurprisingly brief. Allsop CJ said: "Considering the terms of s 32W of the [FTA], at the time of entry into the arrangements, did the provisions in question cause an imbalance in the parties' rights and obligations to the detriment of the consumer? It is difficult to see why this would be so by reference to the matters in s 32X or otherwise. The provisions were clearly disclosed. In most instances, the fees could be avoided. No trickery took place. Although set by the bank in contracts of adhesion, the contracts were terminable at the will of the customer; and the fee could be avoided by the conduct of the customer that was not unreasonable - keeping to her or his contractual limits." His Honour concluded that: "Neither the relevant provisions of the [FTA] nor of the National Credit Code exhibit the intention that the Court should assume the role of a price regulator. It is unjustness or unfairness of transactions or terms that is required to be demonstrated. Price may affect such an evaluation but it does not determine it." That conclusion is correct: to conclude otherwise would be to give the consideration of price a decisive effect which it is not given by the legislation. In this Court, the appellants argued that the late payment fees caused a significant imbalance in the parties' rights and obligations to the detriment of the customer. The requirement of s 32W that the term be "to the detriment of the consumer" was not satisfied because the late payment fee was not a detriment to Mr Paciocco. As has been seen, it was an expense which he chose to risk as more convenient to him than paying his account on time. Conclusion Both appeals should be dismissed with costs.

NETTLE J. These are appeals from a judgment of the Full Court of the Federal Court of Australia (Allsop CJ, Besanko and Middleton JJ). Essentially, they raise two questions. The first is whether a credit card facility late payment fee imposed by the respondent ("the Bank") was a penalty and, therefore, unenforceable. The primary judge (Gordon J) held that it was. On appeal, the Full Court held that it was not established that the late payment fee was a penalty because it was not shown that it was "extravagant and unconscionable in amount in comparison with the greatest loss that could conceivably be proved" to have followed from late payment. The second question is whether the imposition of the late payment fee was "unconscionable conduct" within the meaning of s 12CB of the Australian Securities and Investments Commission Act 2001 (Cth) ("the ASIC Act"), an "unjust transaction" within the meaning of s 76 of the National Credit Code in Sched 1 to the National Consumer Credit Protection Act 2009 (Cth) or an "unfair" contractual term within the meaning of Pt 2B of the Fair Trading Act 1999 (Vic) ("the FTA"). The primary judge did not deal with that question but the Full Court held that none of those provisions applied. The credit card facilities The first appellant, Mr Paciocco, had two credit card facilities. The first was a Low Rate MasterCard Account which was opened in June 2006 with a limit of $15,000 that was increased to $18,000 in November 2009. The second was a Low Rate MasterCard Account which was opened in July 2009 with a credit limit of $4,000. Each credit card facility was entered into on terms set out in the Bank's standard form terms of credit card facility and the Bank's bargaining power was such that Mr Paciocco had no opportunity to negotiate the terms. In effect, it was a case of take it or leave it, although there was no suggestion of the Bank imposing any unfair pressure, undue influence or unfair tactics to cause Mr Paciocco to enter into the facilities, and it was accepted that Mr Paciocco understood the nature of the terms and conditions of the facilities and had a real choice as to whether to enter into the facilities and to make use of them. The terms of each credit card facility included terms as to minimum monthly payments and late payment fees that were comprised in a Letter of Offer, an "ANZ Credit Card Conditions of Use" document and an "ANZ Personal Banking Account Fees and Charges" booklet. In substance, they provided that a late payment fee would be charged upon failure to pay a minimum monthly payment and any amount due immediately (together, "the Monthly Payments") by a specified due date. For example, in June 2006, the Letter of Offer relevantly provided: "Minimum Repayment Each month you are required to pay: • the greater of $10 or 2% of the monthly Closing Balance shown on your statement of account, rounded up to the nearest dollar • or, if the Closing Balance of your statement of account is less than $10, the full closing balance by the Due Date shown on that statement of account. In addition you will need to pay any Amount Due Immediately on receipt of your statement of account. Credit Fees and Charges ... Late Payment Fee: A fee of $35 will be charged to your credit card account if the 'Monthly Payment' plus any 'Amount Due Immediately' shown on the statement of account is not paid within 28 days of the end of the 'Statement Period' shown on that statement." In December 2009, the quantum of the late payment fee was reduced from $35 to $20 but, in substance, the terms that provided for the late payment fee in the relevant 2009 contractual documents remained otherwise the same. Twenty-six late payment fees were charged to Mr Paciocco's accounts: eight in the period up to December 2009 and a further 18 after that. The Monthly Payments that were not paid on time and so resulted in late payment fees ranged between $10 and $750. The delay between the due date and the date on which those Monthly Payments were made ranged from five days to just over 30 days. The proceedings below At trial, Mr Paciocco and the Bank each called an expert witness to quantify the Bank's losses arising from Mr Paciocco's failure to pay the Monthly Payments by the due date. Each expert was, however, asked a different question. Mr Paciocco engaged Mr Regan, a certified public accountant, to identify the amounts needed to restore the Bank to the position in which it would have been if Mr Paciocco had not failed to pay the Monthly Payments by the due date. The Bank engaged Mr Inglis, a chartered accountant, to conjecture an array of costs which were not actually incurred by the Bank but which the Bank contended could have been conceived of at the time of entry into each credit card facility as potentially capable of arising from a failure to pay the Monthly Payments by the due date. Unsurprisingly, the calculations of each expert produced different results. Mr Regan found that the Bank incurred costs of no more than $3.94 on each occasion that Mr Paciocco failed to pay the Monthly Payments by the due date. In contrast, Mr Inglis conjectured that the maximum amount of costs which could have been conceived of at the time of entry into each credit card facility as potentially capable of arising from Mr Paciocco's failure to pay the Monthly Payments by the due date ranged between $5 and $147. The principal reason for that difference was that Mr Inglis included what he described as estimates of the following three categories of costs: (1) increase in provision for bad or doubtful debts; (2) increase in regulatory capital; and (3) operational costs including certain costs associated with collecting amounts owed by customers ("collection costs"). The primary judge accepted Mr Regan's figures and rejected Mr Inglis' estimates as irrelevant. Her Honour held that each late payment fee was payable upon breach of a contractual obligation to pay the Monthly Payments by the due date and that, because the amount of the late payment fee very substantially exceeded the amount of damages that could be awarded for the breach of the obligation to pay the Monthly Payments by the due date, the late payment fee was extravagant and unconscionable in comparison with the greatest loss that could reasonably be proved. It followed, as her Honour held, that the late payment fee was a penalty. The Full Court adopted a test of whether the agreed sum was commensurate with the interest protected by the bargain. Allsop CJ, with whom the other members of the Court agreed, said that, although that test is "not ... unconnected with recoverable damages", the question is to be determined by reference to "the obligee's interest in the due performance of the obligation" and that that "is different from working out what damage can be proved from a particular breach after the event". His Honour reasoned that it was "open ... to take into account the fact that one stipulation could be broken in countless ways, many (and likely most) trivial in consequence and some serious". It followed, in his Honour's view, that, despite the small amount of costs actually incurred by the Bank as a result of late payment by Mr Paciocco, and the very small amount that would have been recoverable as damages for breach of the Monthly Payments obligation, the range of costs which it was conceivable at the time of entry into the credit card facilities could be incurred as a result of a late payment was very much broader than the costs actually incurred. On that basis, the Full Court rejected Mr Regan's figures as irrelevant and accepted Mr Inglis' projections as demonstrating that, as at the time of the establishment of each credit card facility, it was objectively conceivable that failure by Mr Paciocco to pay the Monthly Payments by the due date could result in losses of up to $147 by way of an increase in provision for bad or doubtful debts, an increase in regulatory capital and collection costs. Thus, the Full Court held that it was not demonstrated that the late payment fee was a penalty. Tests for the identification of a penalty In Dunlop Pneumatic Tyre Co Ltd v New Garage and Motor Co Ltd, Lord Dunedin identified four tests (of which the fourth is comprised of four parts) ("the Dunlop tests") which his Lordship said may assist in deciding, and in some cases may determine, whether an obligation to pay a sum of money upon a breach of a contract is penal. In substance, those tests are: (1) whether the contract describes the payment as a penalty or liquidated damages is not decisive; (2) the essence of a penalty is a payment "in terrorem" (which in this context means to deter the offending party from committing the breach), whereas the essence of liquidated damages is a genuine pre‑estimate of damage; (3) the question is one of "construction" (more accurately, of characterisation) of the terms of the contract having regard to the inherent circumstances of the contract at the time the contract was made; (4)(a) the agreed sum will be held to be a penalty if it is extravagant and unconscionable in amount in comparison with the greatest loss that could conceivably be proved to have followed from the breach; (4)(b) the agreed sum may be held to be a penalty where the breach consists only in not paying a sum of money and the stipulated sum is greater than the sum which ought to have been paid; (4)(c) there is a presumption that a single lump sum is a penalty if it is payable on the occurrence of one or more of several events of which some may occasion serious damage and others do not; and (4)(d) where the consequences of breach make the precise pre‑estimate of damage almost impossible, it is more probable that an agreed sum is a penalty than where the damage is capable of precise estimation. As Lord Dunedin made plain in Dunlop, the Dunlop tests are not rules of law and they are not exhaustive. Rather, they represent the distillation of various processes of reasoning that were applied in the several authorities from which the tests were derived and hence are conceived of as likely to prove useful and sometimes determinative in similar cases. Nevertheless, because of the broad range of cases from which the Dunlop tests were derived, and consequently the broad spread of cases to which they are capable of application, they have proved to be of widespread and enduring utility. The Dunlop tests have been applied by this Court on a number of occasions spanning many years. Recently, in Ringrow Pty Ltd v BP Australia Pty Ltd, this Court re‑articulated the orthodox application of the Dunlop tests, that a penalty is "out of all proportion" to the "greatest loss that could conceivably be proved to have followed from the breach". By contrast, in Andrews v Australia and New Zealand Banking Group Ltd, it was observed that the critical issue determined in Dunlop was broader: "whether the sum agreed was commensurate with the interest protected by the bargain". Similarly, in Cavendish Square Holding BV v Makdessi, the Supreme Court of the United Kingdom posed the question in terms of "whether [the sum agreed] is ... out of all proportion to any legitimate interest of the innocent party in the enforcement of the primary obligation". At first sight, the formulations propounded in Andrews and Cavendish might appear to have departed from the Dunlop conception of being "out of all proportion" to the greatest conceivable loss flowing from the breach. Closer examination shows, however, that the Andrews and Cavendish formulations accord with Dunlop. Asking whether the sum agreed is out of all proportion to any legitimate interest of the innocent party in the enforcement of the primary obligation reprises the test formulated by Lord Robertson in Clydebank Engineering and Shipbuilding Company v Yzquierdo y Castaneda (on which the Dunlop tests were in part based) for application to a case where the damage suffered by the innocent party as a result of a breach is incapable of precise or even approximate quantification. The Andrews description of Dunlop, as being concerned with whether the sum agreed was commensurate with the interest protected by the bargain, was part of the Court's consideration of cases in which damage is incapable of even approximate quantification. Nothing said in Andrews runs counter to the approach adopted in Ringrow that "[i]n typical penalty cases, the court compares what would be recoverable as unliquidated damages with the sum of money stipulated as payable on breach". Similarly, as Lords Neuberger and Sumption observed in Cavendish, to pose the test in terms of whether the sum agreed is commensurate with the interest protected by the bargain reflects the way in which a number of English judges in recent times have approached cases involving more complex considerations than arise out of straightforward damages clauses in consumer contracts. But, as their Lordships emphasised, ordinarily the only legitimate interest of an innocent party in the performance of a primary obligation is in its performance or in some appropriate alternative to performance. Hence, in the case of a "straightforward damages clause" the innocent party's interest will rarely extend beyond compensation for the breach and, therefore, in such a case, it is to be expected that the Dunlop tests will usually be "perfectly adequate" to determine whether a provision is a penalty. Are the Dunlop tests adequate in this case? As will be apparent from the differences between the approach of the primary judge and the reasoning of the Full Court, the outcome of the penalty appeal turns to a large extent on whether this case should be regarded as one of the straightforward kind in which the Dunlop tests are "perfectly adequate" to resolve the issues or whether it should be seen as one of the more complex types of cases referred to in Cavendish which necessitate considerations beyond a comparison of the agreed sum and the amount of recoverable damages. In particular, should the Bank be conceived of as having had a legitimate interest in Mr Paciocco's performance of the Monthly Payments obligation which extended beyond its performance or compensation for breach of its performance? Prima facie, where a bargain is for the provision of a credit card facility made available at an agreed rate of interest, the lender's only legitimate interest in the enforcement of the primary obligation is repayment of the facility with interest at the agreed rate plus adequate recoupment of any costs imposed on the lender as a result of the customer's failure to adhere to the terms of the facility. Other things being equal, the lender's interest to be protected by the bargain does not extend to the payment of a liquidated sum that is disproportionate to any amount of additional costs imposed on the lender by reason of the breach or which could have been conceived of at the time of entry into the contract as a cost which would be incurred as a result of the breach. Of course, there are exceptions. In some cases, there may be evidence or other indications of some broader interest to be protected. Cases such as Clydebank, Dunlop and, to some degree, ParkingEye Ltd v Beavis serve as examples. But, for reasons which will be explained, in this case, there is no evidence or other indication of any interest to be protected by the timeous performance of the Monthly Payments obligation apart from the avoidance of costs. In Dunlop, the evidence given by the appellant's manager established that the payment in question was calculated to protect Dunlop's pricing structure, and thus goodwill, albeit that it was impossible to quantify the damage to its pricing structure or goodwill other than to say that it was potentially very substantial. In Clydebank, the losses caused by delay were not susceptible to monetary quantification given that the nature of the contract was to construct warships for Spain. In ParkingEye, it was an agreed fact that the object of an increased parking fee which was charged once a car remained parked for longer than a specified time was to ensure that other members of the public would be able to use the available car spaces. It was accepted that the fee charged for customers for staying longer than a specified time (although being several times the standard parking rate) was no higher than was necessary to achieve the objective of enabling other members of the public to use the available car parking spaces. In contrast here, as has been seen, the only evidence offered in support of the late payment fee was Mr Inglis' projections as to what he termed the greatest amount of costs which could conceivably have been, but which were not in fact, incurred. On the available evidence, including any inferences which might legitimately be drawn from the fact that the late payment fee was imposed by a bank, and that there were a very large number of customers of the Bank with similar credit card facilities, the only interest which it might reasonably be supposed the Bank had in the timeous performance of the Monthly Payments obligation was the avoidance of the costs in fact incurred or the costs which the parties could reasonably have conceived at the time of entry into the contract would flow from late payment. Counsel for the Bank contended that it could be assumed that the Bank had an interest in keeping its cost of funds to a minimum and therefore that the late payment fee was calculated to ensure a degree of compliance with the Monthly Payments obligation resulting in a risk profile of the Bank's credit card account assets which optimised the Bank's cost of funds. In short, it was contended that riskier credit self-evidently is more costly credit. That contention is not persuasive. There was no evidence of how much greater the cost of funds might be and, in the absence of evidence, as Mr Regan observed, it is equally probable that the likelihood of default in performance of the obligation to make Monthly Payments was already priced into the relatively high rates of interest charged on credit card facilities. There is, moreover, no apparent correlation between the amount of the late payment fee and the amount and duration of lateness of the Monthly Payments; and, in that critical respect, the late payment fee is the antithesis of the kind of interest rate increase which Colman J concluded in Lordsvale Finance Plc v Bank of Zambia could be justified on the basis that riskier credit is more costly credit. Counsel for the Bank also submitted that the late payment fee could be viewed as a "non-punitive" financial incentive, set by the Bank in the context of a competitive banking market, of which the objective purpose was to encourage customers to comply with the terms of their facilities as opposed to punishing them for failing to comply with those terms. That submission is equally unpersuasive. Whether the late payment fee is called an incentive or a punishment is beside the point. In effect, every penalty (or punishment) is an incentive even though not every incentive is a penalty. As the law of penalties has evolved, however, it is not concerned with whether an obligation is properly to be conceived of as a punishment as such. It is concerned with whether an obligation to make a payment on breach of a contractual or other principal obligation is of an amount which is grossly disproportionate to the foreseeable consequences of the breach. In effect, the gross disproportion is in itself sufficient to render the obligation "unconscionable" and therefore unenforceable. Accordingly, in England, and hitherto in this country, the main purpose of the law of penalties has been as a rule of law to prevent a party recovering a sum of money in respect of a breach of contract which bears little or no relationship to the loss actually suffered by that party. And, similarly, in equity, as Mason and Wilson JJ observed in AMEV‑UDC Finance Ltd v Austin (and as was later re‑affirmed in Andrews), "there seems to have been no instance of equity awarding compensation over and above the amount awarded as common law damages, other than cases in which equity would not relieve against the penalty". For the same reason, the fact that the late payment fee may have been set by reference to what other banks were charging in the context of a competitive banking market is essentially irrelevant. As Lords Neuberger and Sumption observed in Cavendish, although the penalty rule originated out of equity's concern to prevent exploitation at a time when credit was scarce and borrowers were particularly vulnerable, the modern rule is substantive, not procedural. It does not normally depend for its operation on establishing that advantage was taken of one party. An obligation to pay a fee does not cease to be penal just because the obligee's competitors impose similar penalties on their customers. Where there is an incentive constituted of an obligation to pay a sum of money conditioned on a breach of contract of which the amount is wholly disproportionate to the greatest costs which would have been conceived of at the time of entry into the contract, the obligation will be regarded as penal unless there be some aspect of the contract which makes it possible to say that the amount of the obligation is not wholly disproportionate to the interest protected by the bargain. Nor does the fact that the late payment fee may have been set by reference to fees being charged by the Bank's competitors establish that the Bank had an interest in the timeous performance of the Monthly Payments obligation other than its performance or adequate compensation for its breach. Of itself, it suggests no more than that the extent of competition, or more likely lack of it, led the Bank to conclude that the level at which it set the fee was unlikely to result in an unacceptable loss of custom to other banks. That says nothing about whether this fee or those fees were penal in the relevant sense. In the result, although in some circumstances the test of whether a fee is a penalty may go beyond the quantitative assessment of disproportion between the fee and the greatest loss that could conceivably be proved, and if so may require a qualitative assessment of disproportion between the fee and a more imprecise conception of the interest protected by the bargain, in this case it is not possible to identify any interest to be protected by the bargain other than payment of the amount which was due together with compensation for such costs incurred as a result of late payment as would be recoverable as damages for breach of contract. That being so, to accept that the Bank was entitled to impose a late payment fee wholly disproportionate to the greatest loss that could conceivably be proved would be in effect to abandon a large part of the existing law relating to penalties. None of the authorities and nothing which was advanced in argument provides an acceptable basis for taking that step. There was and is no reason why the matter should not be determined in accordance with the Dunlop tests. Which Dunlop tests apply? Mr Paciocco claimed that the Full Court erred by applying test 4(a) of the Dunlop tests rather than test 4(b). Counsel for Mr Paciocco submitted that it was apparent from the terms of the Dunlop tests and a number of the cases in which they have since been applied that tests 4(a) and 4(b) are mutually exclusive in the sense that, where a breach of obligation consists only in failing to pay a sum of money and the stipulated sum is greater than the sum which ought to have been paid, test 4(b) applies to the exclusion of test 4(a). That argument should be rejected. As Lord Dunedin said in Dunlop, although test 4(b) embodies one of the most ancient indicia of a penalty, it is but a corollary of test 4(a) and owes its existence to a time when the only available remedy for breach of obligation to pay a specified sum was recovery of the specified sum and perhaps interest. At that point in the law's development, it was not open to recover unliquidated damages for the consequences of failure to pay a specified sum of money or to pay it timeously. It was for that reason, as Lord Parmoor observed in Dunlop, that: "Since the damage for the breach of covenant is in such cases by English law capable of exact definition, the substitution of a larger sum as liquidated damages is regarded, not as a pre-estimate of damage, but as a penalty in the nature of a penal payment." Since then, the position has changed. Now it is possible to recover unliquidated damages for breach of an obligation to pay a specified sum and, accordingly, the amount recoverable for breach of such an obligation is no longer necessarily capable of exact pre‑estimation. There is no longer any reason in principle or policy why test 4(a) should be regarded as inapplicable to a failure to make a specified payment. As Lord Parmoor said, each case depends on its own facts and circumstances. The better view of the operation of test 4(b) in contemporary circumstances is that it represents a possible, but not always necessary, application of the broader principle expressed in test 4(a) to facts of the kind identified in test 4(b). In any event, strictly speaking, this case is not of the type described in test 4(b). The late payment fee was payable on failure to make the Monthly Payments on time. The breach or failure of performance on which it was conditioned was lateness in payment as opposed to a failure to pay. Properly understood, as the primary judge held, this case fell within the general principle reflected in test 4(a) and therefore the applicable test was whether the late payment fee was exorbitant or extravagant (or, in other words, "out of all proportion") in comparison with the greatest loss that could conceivably be proved to have followed from the breach. Additionally, as the primary judge also held, this case engaged test 4(c) of the Dunlop tests. The late payment fee was of the same amount regardless of the magnitude of the Monthly Payments required to be made and of the extent of lateness in payment, and thus, according to test 4(c), there arose an evidentiary presumption that the payment was penal. It then fell to the Bank to show why the late payment fee was not penal. The greatest recoverable loss that could conceivably be proved Mr Paciocco further contended that, if test 4(a) of the Dunlop tests were applicable, the primary judge was correct to assess the greatest loss that could conceivably be proved to have followed from the breach by reference to what would be recoverable as unliquidated damages. For that reason, it was submitted, the Full Court erred in taking into account forms of projected losses which might be conceived of as bearing some possible relationship to the breach but which at law are regarded as too remote to be recoverable. That contention should be accepted. As Mason and Wilson JJ observed in AMEV-UDC, during the first half-century following Dunlop the concept of an agreed sum being extravagant in comparison with the greatest amount of damage that could conceivably have been contemplated by the parties became, in effect, a test of whether the agreed sum was greater than the amount of damages that could be awarded for the breach of contract. That change was "consistent with an underlying policy of restricting the parties, in case of breach of contract, to the recovery of an amount of damages no greater than that for which the law provides". Hence, as stated in Chitty on Contracts, the word "damage" in this context means "net loss" after taking account of the claimant's expected ability to mitigate. In AMEV-UDC, Mason and Wilson JJ posited that it would be in the interests of freedom of contract, and therefore desirable, for the courts to return to the idea that an agreed sum should only be characterised as a penalty if it is "out of all proportion" to the damage likely to be suffered as a result of the breach. They proposed that the test be one of degree that depends on a number of circumstances including the degree of disproportion between the stipulated sum and the likely loss to be suffered (a factor which they said would be relevant to the oppressiveness of the term to the party required to pay) and the nature of the relationship between the contracting parties (a factor which they said would be relevant to the unconscionability of the party's conduct in seeking to enforce the term). It may be accepted that the law has now reached that stage of development. But, even so, as was stated in Ringrow, in assessing the degree of disproportion in typical penalty cases the sum of money stipulated as payable on breach is to be compared with what would be recoverable as unliquidated damages for the breach. For the reasons already stated, this case is a typical penalty case of the kind referred to in Ringrow. In those circumstances, the primary judge was correct to consider whether the late payment fee was out of all proportion to the amount recoverable as unliquidated damages. Perspective for the assessment of loss The Full Court criticised the primary judge's approach to comparing actual damage suffered by the Bank with the late payment fee as an ex post determination of whether the prima facie penal character of the late payment fee was rebutted. According to the Full Court, what was required was an ex ante analysis (as at the date of entry into the credit card facility) to assess whether the fee was extravagant and unconscionable in amount in comparison with the greatest loss that could conceivably be proved to follow from the breach. It followed, the Full Court held, that the primary judge was wrong to have regard to Mr Regan's evidence of costs actually incurred by the Bank on late payment by Mr Paciocco and wrong to reject Mr Inglis' projections as irrelevant. Contrary to the Full Court's reasoning, the primary judge did not take an ex post approach to the identification of conceivable loss. As her Honour's reasons make clear, she approached the task ex ante in accordance with Dunlop tests 4(a) and 4(c). By that means, she discerned that, because the late payment fee was of a fixed amount regardless of the magnitude and duration of the late payment, there was no ex facie relationship between the amount of the fee and any loss resulting from the lateness of payment. That gave rise to an evidentiary presumption that the late payment fee was penal. Then, in order to determine whether that presumption was rebutted, her Honour embarked upon a comparison of the amount of the late payment fee with the amount which would be recoverable as unliquidated damages for breach of contract. The primary judge was also not in error in having regard to Mr Regan's evidence of actual damage. Although the exercise is one of characterisation and therefore, as the Full Court said, looks forward ex ante from the time of entry into the contract, evidence of what happens after a contract is entered into is capable of providing a sound basis from which to infer what the parties would have reasonably expected to be the loss. That is why, in the typical kinds of penalty cases to which this Court referred in Ringrow, courts generally follow the approach of comparing what would be recoverable as unliquidated damages with the amount of the stipulated payment. For that purpose, courts may admit evidence to prove losses which may fairly and reasonably be considered to have arisen naturally according to the usual course of things so as to come within the first limb of Hadley v Baxendale, and losses which, because of particular facts or circumstances known to the parties at the time of entry into the contract, may reasonably be conceived of as having been within the contemplation of both parties as the probable result of breach so as to come within the second limb. As has been seen, it has only been in more complex cases of the kind considered in Clydebank, Dunlop and Cavendish, where the damage likely to result from a breach is not capable of precise pre‑assessment, or cases like Cavendish and ParkingEye, where the interest of the innocent party in having the contract performed can be seen to extend beyond compensation for breach, that evidence has been admitted to prove what conceivably might have occurred even though it did not in fact occur. Late payment fee is out of all proportion to recoverable damages The question then is whether the primary judge was correct in concluding that the amount of the late payment fee was extravagant and unconscionable or out of all proportion to the amount which would be recoverable as unliquidated damages for breach of the Monthly Payments provision. As was previously mentioned, Dunlop test 4(c) recognises that the fact that a single lump sum is payable on the occurrence of one or more of several events of which some may occasion serious damage and others do not suggests that the obligation is a penalty. Similarly, the fact that the late payment fee in this case is a fixed fee regardless of whether the late payment is serious or trivial with respect to time or amount suggests that the late payment fee is a penalty. Hence, applying the Ringrow approach, the issue becomes one of whether there is any evidence or other compelling considerations sufficient to rebut the presumption to which that gives rise. As already noticed, the only evidence which the Bank put forward to support its contention that the late payment fee was not penal was Mr Inglis' projections of the costs of increase in provision for bad or doubtful debts, increased costs of regulatory capital and collection costs. The primary judge rejected those projections as irrelevant because, to a very large extent, no such costs were incurred. For the reasons which follow, her Honour was correct to do so. (i) Increase in provision for bad or doubtful debts The Bank uses provisioning to estimate the impairment of its financial assets. It is required to hold provisions for losses under both international and Australian accounting standards. One such standard, Australian Accounting Standards Board 139, was concerned with the measurement of a present or current loss, as opposed to estimates of future losses that might be expected or anticipated, in order to ensure that the financial statements present a fair value of what is likely to be collected from the Bank's receivables. Consistently with those standards, when accounting for a provision, two changes were reflected in the Bank's accounts. First, the amount owed by customers as recognised in the Bank's balance sheet was reduced to the level that the Bank expected to recover. Secondly, a loss was recorded in the Bank's profit and loss account as an expense or cost representing the impairment in the asset value. Given the nature of those entries, the primary judge held that the costs of increase in bad or doubtful debts which Mr Inglis attributed to each late payment were in effect no more than an estimate of possible future losses and so were not recoverable as damages for breach of contract. That conclusion was correct. A provision for bad or doubtful debts is an estimate of future loss, not an incurred loss. Perforce of the applicable accounting standards, the provision must be brought to account in the Bank's balance sheet as a credit against (and so in reduction of) the value of loans and other credit facilities the subject of provision. It must also be reflected in the Bank's profit and loss account as a credit against (and thus in reduction of) reported annual income. But, if the debt is recovered in a future year of income, the provision is or should be reduced accordingly and the reduction in provision should be carried to the profit and loss account as an accretion to the latter year's income. Most importantly, although the amount of an increase in provision for bad or doubtful debts is recorded as a once and for all expense against annual income for the year in which the provision is raised, it does not mean that the Bank has in fact received any less by way of income in that year or that it has had to pay away any of its income in satisfaction of a diminution in capital in that year. As the primary judge said, it is simply an estimate of what might or might not one day prove to be the case and therefore is not recoverable as damages. Conceivably, the Bank might suffer some detrimental consequence as a result of the reported reduction in asset value or annual income in the relevant year of income. At least in theory, either the reduction in asset value or the reduction in annual reported income could reduce the Bank's ability to borrow or lend money as part of its income producing activities. But the words "in theory" are stressed because, as Mr Regan suggested, in reality the Bank's generation of income from credit card facilities (and therefore at the very high rate of return which they generate) is as much dependent on lending to customers who it may be assumed will be late in making Monthly Payments as it is upon the state of its balance sheet and annual reported income. And, even if the Bank did suffer some reduction in its ability to borrow or lend money as a result of the increase in provision, the loss thereby occasioned to the Bank would certainly not be the amount of the increase in provision. It would be the present discounted value of the estimated reduction in income flowing from the restriction in borrowing or lending activities as a result of the increase in provision. Self-evidently, any such amount would be vastly less than the amount of the increase in provision. Furthermore, assuming that an increase in provision for bad or doubtful debts were recoverable as damages for breach of contract, it is apparent that the amount of the "costs" of increased provision that Mr Inglis attributed to late payment by Mr Paciocco was out of all proportion to any increase in provision which may have resulted from late payment. As appears from Mr Inglis' report, the Bank did not calculate provisions individually for each customer. Instead, it obtained approval from the Australian Prudential Regulation Authority to use an Internal Ratings-Based approach to credit risk which resulted in a Collective Provision for currently identified losses on pools of loans and other credit facilities that the Bank assessed to be at similar risk of defaulting in repayment. A provision was then calculated for the whole of each pool by means of the following formula: "Expected losses = PD [Predicted Default] x EAD [Exposure at Default] x LGD [Loss Given Default] x PCE [Potential Credit Exposure]". The Predicted Default figure was a statistical probability of an account in a pool of customer accounts with similar payment behaviours going into default over the following 12 months. That figure was multiplied by the Exposure at Default, which was a statistical estimate based on all current consumer credit accounts of the amount outstanding in the event a customer defaults. That figure was then multiplied by the Loss Given Default, the expected loss in the event of default, which was a statistic modelled across all consumer credit accounts. In turn, that amount was multiplied by the Potential Credit Exposure, which was the greater of the account limit or account balance where that amount might be assessed on an individual account basis but, for the purpose of at least some of Mr Inglis' calculations, where an "average exposure" was assumed. Given that, on the evidence, there were more than two million customer accounts in the pool, with credit limits ranging up to $500,000 per account, and therefore a total exposure of more than $19.4 million, and given that Mr Paciocco had only two facilities, of which the greater had a credit limit of less than $19,000, any amount by which late payment by Mr Paciocco caused the provision to be increased was necessarily minuscule in comparison to the total. In what was said to be an endeavour to provide an estimate of the portion of the increase in provision actually caused by each late payment by Mr Paciocco, Mr Inglis split the total population between customers who had made a late payment and those who had not. He then calculated a difference in "average" cost of provision for bad or doubtful debts per account (calculated across each group using the formula already described), of $23 per account (on his most conservative estimate). According to that analysis, the $23 was an "average" cost of increase in provision per late payment per account for each customer within the group of customers who made late payments, which was the cost to the Bank of each of Mr Paciocco's late payments. The problem with that, however, is that, due to the nature of the formula already described, and the fact that the "average" cost per late payment was calculated across the group of customers regardless of the amount of the individual facility limit or potential default of each customer, the $23 "average" cost per late payment was equally disproportionate to the extent to which each of Mr Paciocco's late payments added to the Bank's costs of increase in provision for bad or doubtful debts, and equally disproportionate to what the Bank and Mr Paciocco might be supposed to have contemplated at the time of entry into the facility as costs which could result from any of Mr Paciocco's late payments. (ii) Increase in regulatory capital As an authorised deposit-taking institution, the Bank is required to maintain adequate capital, known as regulatory capital, to act as a buffer against unexpected losses. The minimum amount of regulatory capital required to be held is determined by measuring the Bank's overall capital base against its holdings of risk weighted assets ("RWA"). The more RWA the Bank holds, the greater the amount of regulatory capital it must put aside to counter the risk of default on those assets. The primary judge concluded that none of the alleged increases in regulatory capital could be directly or indirectly related to any of Mr Paciocco's late payments and, consequently, that the alleged cost of increase in regulatory capital should not be taken into account in the assessment of damages recoverable for breach of contract. That was also correct, although for partly different reasons from those which appealed to her Honour. According to his report, Mr Inglis' estimate of the "cost" of increase in regulatory capital consequent upon late payment was comprised of two elements: (a) the "cost" of increase in RWA; and (b) the cost of additional regulatory capital to replace Core Tier 1 capital the result of increase in RWA. The estimate of the "cost" of increase in regulatory capital consequent upon a late payment was between $9 and $12 per late payment. Mr Regan made a number of criticisms of the methodology of that estimate which, if accepted, would suggest it was significantly inflated. Over and above those criticisms, however, it is apparent that the "cost" of increase in regulatory capital as so estimated is no more than an estimate of increase in provision for RWA. It follows that, like the increase in provision for bad or doubtful debts, it is not a loss or outgoing that would be recoverable as damages for breach of contract. It is merely an estimate of a cost which might one day be, but equally might not be, incurred. In contrast to the "cost" of increase in RWA, it appears that the cost of replacing Core Tier 1 capital might have been a loss or outgoing actually incurred, due to the need to supplement Tier 1 capital relegated to regulatory capital, and so might be recoverable as damages for breach of contract. But the amount of it was de minimis. Mr Inglis' estimation of the "average" costs of additional regulatory capital to replace the Core Tier 1 capital was between only $1 and $2 per late payment. Mr Inglis also proposed an alternative calculation, of between $5 and $6 per late payment, which he said he computed by excluding repayment events (that is, repayments and write-offs). But he did not offer any justification for excluding repayment events. Nor is there any reason in principle why repayment events should have been excluded. To the contrary, as Mr Inglis indeed explained, repayment events result in a reduction in the provision for RWA that reduces the need for regulatory capital and thus has an associated cost saving. Hence, to exclude repayment events from the calculation significantly distorts the estimate. (iii) Collection costs Late payment of amounts owing on a credit card account may trigger collections activity by the Bank to recover the amounts due, including by contacting overdue customers by telephone. Mr Inglis put forward two alternative estimates of collection costs caused by late payment of the Monthly Payments: Calculation A and Calculation B. In Calculation A, he conjectured that the "average" collection costs which could result from a late payment were only $4.90 and in Calculation B, he proposed a much higher figure of $15.70 per account. The method of calculation employed in Calculation B differed from the method used for the purposes of Calculation A (and, therefore, from the method used for all other calculations undertaken by Mr Inglis) in that it was made on what Mr Inglis termed a "per account" basis, as opposed to a per late payment basis, as was used for all other calculations. Since the late payment fee was chargeable on a per late payment basis, and since the question for present purposes is whether the late payment fee was out of all proportion to the costs which the Bank might conceivably incur as a result of the late payment, there is no acceptable justification for the adoption of the Calculation B methodology. Mr Inglis also provided a further alternative calculation which he called the "maximum amount of costs that the Bank could conceivably have incurred", which ranged between $44.90 and $99.20 per account. Those estimates were made on the basis of the 95th percentile of the total telephone call duration for each stage of collections activity. But there is even less justification for accepting an estimate of that kind than there is for adopting the per account basis employed in Calculation B. For even though it is open to take into account the fact that a stipulation may be broken in countless ways, and that although the majority of them are likely to be trivial some may be serious, conjectured costs of between 10 and 20 times Mr Inglis' own, on any view, generous estimate of the increase in collection costs likely to be incurred on an "average" basis are not recoverable as damages for breach of contract and cannot otherwise realistically be regarded as anything like a reasonable estimate of the greatest conceivable loss likely to be incurred. As counsel for Mr Paciocco put it, such an estimate is untethered from reality. (iv) Total additional costs It follows from the foregoing that, so far as the evidence went, the maximum amount of additional costs resulting from a late payment that might conceivably have been recoverable as damages for breach of contract was $6.90 per late payment ($2 "average" cost of replacement of regulatory capital and $4.90 for "average" collection costs). And it follows from that that the late payment fee of $35 per late payment (or even $20 per late payment as it later became) was grossly disproportionate to the greatest amount of damages recoverable for breach of the Monthly Payments obligation. The late payment fee is a penalty As Mason and Wilson JJ proposed in AMEV-UDC, the question of whether an obligation to make a specified payment conditioned on a breach of contract is penal is a question of degree which depends on a number of circumstances, including the degree of disproportion between the stipulated sum and the likely loss to be suffered and the nature of the relationship between the contracting parties. In this case, the contract is a standard-form consumer credit contract and the Bank's bargaining power was such that Mr Paciocco had no opportunity to negotiate the terms. That consumer relationship, combined with the fact that the late payment fee of $35 (or $20 as it later became) was extravagant or otherwise out of all proportion to the $6.90 of costs which might conceivably have been recoverable as damages for breach of contract, warranted the primary judge's conclusion that the late payment fee was a penalty. Notice of contention By a notice of contention, the Bank contended that Mr Paciocco's claim to recover a late payment fee incurred on 4 September 2006 is time-barred by the operation of the Limitation of Actions Act 1958 (Vic) ("the Limitation Act"). Given the way in which the majority of this Court has determined that the penalty appeal should be decided, it is perhaps unnecessary to consider whether, if it were recoverable, the late payment fee incurred in 2006 would be time-barred. For completeness, however, it should be observed that, as finally put in argument, the sole basis of the contention was that the primary judge and the Full Court were in error in holding that s 27(c) of the Limitation Act is capable of application to a mistake of law. On that basis, the contention may be rejected. It is plain for the reasons essayed by the primary judge, and more fully in the Full Court by Besanko J, that s 27(c) of the Limitation Act, like s 26(c) of the Limitation Act 1939 (UK), on which it is based, is capable of applying to a mistake of law as much as to a mistake of fact. Statutory causes of action The hearing before the primary judge was conducted on the agreed basis that, if the judge determined that the late payment fee was penal, it would be unnecessary for her Honour to decide whether the imposition of the late payment fee was "unconscionable conduct" within the meaning of s 12CB of the ASIC Act, an "unjust transaction" within the meaning of s 76 of the National Credit Code or an "unfair" contractual term within the meaning of Pt 2B of the FTA. Accordingly, having held that the late payment fee was penal, the primary judge did not go on to consider whether the late payment fee was caught by any of those statutory provisions. Given that I have concluded that the primary judge was correct in holding that the late payment fee was penal, it is unnecessary for me to consider whether the late payment fee attracted any of those provisions. Conclusion and orders In the result, I would order that the appeal in M220 of 2015 be allowed with costs. The orders of the Full Court should be set aside and in lieu thereof it should be ordered that the appeal to the Full Court be dismissed and that the Bank pay the appellants' costs of the appeal. In M219 of 2015, which was in effect dependent on the outcome in M220 of 2015, it should be ordered that the appeal be dismissed without any adjudication upon the merits and that the appellants pay the Bank's costs of that appeal.