I.          Executive summary

Parties are generally free to contract as they wish, through exchanging promises and imposing obligations on one another that are enforceable in a court of law. However, there are specific limitations that the court has set on contracts, whether for policy or practical reasons.

One such limitation is the ability of a party to stipulate a sum to be paid in the event of a breach of contract, through what is called a liquidated damages clause. This is different from common law damages, which are awarded by the court to an innocent party as compensation for the actual loss suffered after a breach of contract. Liquidated damages clauses are often used in situations where it would be difficult to determine the appropriate compensation through common law damages, e.g. loss of profits from onward sales or delay in processes.

The key principles on contractual penalties were articulated more than a century ago in Dunlop Pneumatic Tyre Company, Ltd v New Garage and Motor Company, Limited [1915] AC 79 (“Dunlop”): a liquidated damages clause is only enforceable if it is compensatory in nature, meaning that it provides a “genuine pre-estimate” of the likely loss caused by the breach. This penalty rule has been followed in almost all the Commonwealth jurisdictions, including Singapore.

In the past decade, however, two developments occurred that the Singapore courts have not – until recently – had the opportunity to consider. These are contained in Andrews v Australia and New Zealand Banking Group Limited (2012) 247 CLR 205 (“Andrews”) and Cavendish Square Holding BV v Makdessi [2016] AC 1172 (“Cavendish”). In Andrews, the High Court of Australia held that the penalty rule was not limited to liquidated damages clauses on breach of contract, but could also apply to clauses similarly stipulating payment contingent on an event that was not a breach of contract. And in Cavendish, the UK Supreme Court held that liquidated damages clauses were not restricted to compensating for loss. They would be enforceable so long as they protected the “legitimate interests” of the party seeking to enforce the clause, and the detriment imposed was not out of proportion to those interests. While the Singapore Court of Appeal (“CA”) recently acknowledged these developments in Leiman Ricardo v Noble Resources Ltd [2020] 2 SLR 386, it left open the question of whether Singapore would follow the new developments or retain the traditional Dunlop position.

That question has now been answered. In Denka Advantech Pte Ltd v Seraya Energy Pte Ltd [2021] 1 SLR 631, the CA affirmed the old rule articulated by Lord Dunedin in Dunlop, and declined to follow the new developments in Andrews and Cavendish. The appeal concerned alleged breaches of three electricity retail agreements (“ERAs”) between the electricity retailers and their customers. The court held that the customers were in breach of all three ERAs and that the liquidated damages clauses therein were enforceable by the retailers. In reaching that conclusion, the court applied the “genuine pre-estimate” Dunlop test, and rejected the “legitimate interests” Cavendish test.

The decision has established that Singapore will continue following the Dunlop test: a clause that provides for a sum which is more than a genuine pre-estimate of loss is considered penal (not compensatory) and is hence unenforceable, even if it seeks to protect a party’s commercial interest. However, this rule cannot apply to stipulated payments which are contingent on events that are not breaches of contract. In rejecting both new developments, the CA also placed heavy emphasis on the importance of not infringing parties’ freedom of contract.

II.        Material facts

Electricity generation companies participate in the wholesale market, where they sell electricity to the National Electricity Market of Singapore (“NEMS”). The companies submit selling bids and, based on these bids, a fixed Uniform Singapore Energy Price (“USEP”) is generated. All successful bidders are paid the USEP. NEMS then sells electricity to energy retailers. Electricity is sold to these retailers at the Pool Price which is largely derived by adding an Hourly Energy Uplift Charge (“HEUC”) to the USEP.

Seraya was an energy retailer and the wholly owned subsidiary of YTL PowerSeraya Pte Ltd (“YTL”), an electricity generation company. Seraya and YTL had a Contract for Differences[1] (“the CFD”) to enable the whole group to mitigate its exposure to the fluctuations of the USEP and Pool Price, which changed every half hour. As the contract price between Seraya and its customers was fixed, the relevant difference was that between the contract price and the fluctuating Pool Price. Under the CFD, YTL bore most of the risk from Seraya’s contracts with the latter’s customers. If Seraya suffered losses due to its fixed contract price being lower than the fluctuating Pool Price, YTL would pay that difference to Seraya. Conversely, if Seraya made a profit from its contract price being higher than the Pool Price, a portion of that profit would be shared with YTL.

The customers in the ERAs here were Denka Advantech Pte Ltd (“DAPL”) and Denka Singapore Pte Ltd (“DSPL”) (collectively, “Denka”). ERAs 99 and 101 contained the following liquidated damages clause:

In the event that the Contract Duration is terminated, … [Denka] shall immediately pay [Seraya]

(if the Contract Duration is terminated [pursuant to a contractual right to terminate if, among others, Denka breaches any obligation] …) liquidated damages in an amount computed as follows:

Amt = A x B x 40%


Amt is the amount of the liquidated damages payable to Seraya Energy pursuant to [the liquidated damages clause];

A is [the number of remaining months in the contract duration]; and 

B is [the average of the amount payable by Denka in a stipulated period prior to termination].

The clause in ERA 100 was a more general clause which was triggered if the contract was terminated “for any reason whatsoever”.

About two years into the ERAs, Denka decided that it wanted to “get out” of the ERAs as the price was disadvantageous to Denka. In the negotiations that ensued between the parties, DSPL wrote a letter to YTL stating that “the supply of steam and electricity shall cease”. Seraya regarded this as Denka’s repudiation of (i.e. renouncing their obligations under) the ERAs, and further clarified that DAPL, too, wished to cease the supply of electricity under the ERA. DAPL did not reply. As a result, Seraya terminated all three ERAs on the basis that Denka did not perform its obligations under the ERAs. Seraya then sued for liquidated damages against Denka.

III.       Issues

Though the CA considered a number of issues in its judgment, this summary focuses on the issues relating to enforceability of the liquidated damages clauses in the ERAs. It was in this context that the court had to consider whether to follow the developments in the penalty rule in Andrews and Cavendish.

A.        Did Denka breach the ERAs?

An antecedent question was whether Denka was even in breach of the ERAs, and thus liable for wrongful termination of the same. The court answered this question in the affirmative. Denka’s letter purporting to unilaterally cease the purchase of electricity amounted to repudiation of the ERAs. In response, Seraya then validly terminated the ERAs, relying on either its common law or contractual right of termination.

B.        Were the liquidated damages clauses in the ERAs enforceable, or were they penalties and therefore unenforceable?

The CA first considered the applicable legal principles, before applying them to the facts of the case.

(1)        The scope of the penalty rule

In light of Andrews and Cavendish, the CA had to consider the initial question of whether the penalty rule should apply even where there was no breach of contract.

The High Court of Australia in Andrews held that the rule’s scope of application was not limited to clauses purporting to take effect only upon a breach of contract. That is, if a clause provides for a sum to be paid if a party fails to do X – even where X is not a contractual obligation – the clause may nevertheless be unenforceable for violating the rule against penalties. The UK Supreme Court in Cavendish,however, disagreed with Andrews and maintained that the scope of the rule, in English law, was limited to situations involving a breach of contract. Thus, the CA in this case had two different positions to consider.

Ultimately, the CA agreed with Cavendish insofar as the scope of the rule was concerned. First, the CA did not agree with the historical approach adopted by the court in Andrews.[2] The penalty rule was developed in relation to the very specific situation of enforcement of penal bonds;[3] there was no sound reason why it should apply in the same manner to all modern contracts. Instead, the CA noted that there were persuasive reasons in logic and principle to prefer the requirement of a breach of contract. Most crucially, the requirement of breach cohered with the concept of freedom of contract, because the penalty rule would be confined to the sphere of secondary obligations, without interfering with the primary obligations between the contracting parties.[4] While the strict threshold of a breach of contract may be thought of as being too easy to circumvent by clever drafting, this did not justify removing the prerequisite altogether. Such a problem of drafting could be dealt with under other contract law principles such as construction of contract, and in particular, prioritising substance over form.

The CA also rejected a possible alternative approach. There had been debate in Australia as to whether the broad doctrine of unconscionability[5] could have been utilised instead of extending the penalty rule. The CA observed that Singapore had only adopted the narrow doctrine of unconscionability, which could not accommodate or substitute for the extended penalty rule. Indeed, the narrow doctrine of unconscionability was consistent with the limited scope of the penalty rule that the CA favoured over Andrews, in that both reflected a judicial approach of minimal intervention in contractual bargains.

(2)        The content of the penalty rule

As regards the substantive criteria of the penalty rule, the CA had to decide between endorsing the traditional statement of principles by Lord Dunedin in Dunlop or incorporating the development in Cavendish.In Dunlop,the determinative question was whether the liquidated damages clause provided a genuine pre-estimate of the likely loss, such that the focus of the clause was on compensating the innocent party. However, in Cavendish, the UK Supreme Court held that there could be “legitimate interests” (that go beyond compensation) which can justify imposing a sum over and beyond the likely loss caused by the breach of contract.

The CA declined to follow Cavendish, and instead endorsed the principles in Dunlop. In the CA’s view, the Dunlop requirement that a stipulated sum be a genuine pre-estimate of the likely loss was preferable, because it was wholly consistent with the fact that secondary obligations are about compensating the plaintiff. Clauses designed to go beyond compensation of the likely loss suffered by the innocent party, even if commercially justifiable (i.e. the Cavendish approach), were necessarily penal in nature.

Thus, when considering the enforceability of a liquidated damages clause, the question was simply whether the clause provided a genuine pre-estimate of the likely loss. There would nevertheless be “legitimate interests” relevant to the determination of this question, to the extent that the “legitimate interest” was that of compensation. For instance, the loss of the usage of the subject-matter of the contract because of a failure to deliver it, or the damage caused to a trade system because of a failure to abide by certain provisions meant to protect branding and reputation, were indeed held to be losses that must be compensated for. Such interests remained relevant in determining whether the amount was a genuine pre-estimate of the likely loss. But all other “legitimate interests” (or commercial interests) had no role to play, as they did not relate to compensation of the likely loss.

There were additional reasons for the court’s decision to prefer Dunlop over Cavendish. First, the approach adopted was consistent with the CA’s earlier decision in PH Hydraulics & Engineering Pte Ltd v Airtrust (Hongkong) Ltd [2017] 2 SLR 129. In that case, the court held that punitive damages generally cannot be awarded for breach of contract. Secondly, there was a practical concern that the concept of “legitimate interest” (as advanced in Cavendish)was too broad a concept. There was potential for it to be used too flexibly, which would lead to too much uncertainty – both prior to entry into contracts as well as in the results arrived at by the courts.

The CA clarified that the rejection of the legitimate interest test did not mean that the other principles in Cavendish were entirely rejected. Instead, the other elements discussed in Cavendish had to be viewed in light of whether the clause is a genuine pre-estimate of the likely loss. For instance, the fact that parties were of equal bargaining power, with the necessary legal representation, featured prominently in Cavendish. The court agreed that the penalty rule must be exercised sparingly in such instances so as not to undermine the general principle of freedom of contract; this element remained open for a party to rely upon it in argument.

Another potential consideration was the purpose of the underlying transaction and the particular primary obligation that had been breached. That is, in determining whether the clause was a genuine pre-estimate of the likely loss, the court may have regard to the nature and importance of the primary obligation as well as the seriousness of the consequences of breach.

(3)        The liquidated damages clauses were secondary obligations and hence subject to the penalty rule

The CA then applied the above principles to the facts of the case. In doing so, the CA had to determine whether the liquidated damages clauses were secondary obligations because, as decided by the CA, the rule only applied in the event of breach. It was not disputed that the liquidated damages clauses in ERAs 99 and 101 were secondary obligations, triggered only upon the breach of a primary contractualobligation. The one in contention between the parties was that contained in ERA 100 which stipulated payment to be made if the contract was terminated “for any reason whatsoever”.

While the liquidated damages clause in ERA 100 could be triggered by a breach of contract, it could also be triggered in other circumstances such as Denka terminating the ERA with written notice. There was thus some difficulty in deciding whether to classify the clause as a primary or secondary obligation, which would determine whether the clause was subject to the penalty rule.

The CA observed that such clauses had not received consistent treatment although they were often used in hire-purchase contracts. The controversy arose because an extravagant liquidated damages clause would be unenforceable if it was triggered by an accepted repudiation, but the same clause could be enforceable if it was instead triggered by a lawful termination. This was previously described as an “absurd paradox” by a prior UK court.

The CA then held that there could be a contractual clause which is a “hybrid” obligation: the clause could be either a primary or secondary obligation, depending on the event that triggers the termination of the contract. This is because, as previously held, the penalty rule depended on breach of contract, and this was merely the outcome of applying that rule. The rationale in allowing the clause to be enforceable where there was no breach was that the amount was simply the price (albeit a high one) for one’s exercise of its termination right, which parties are free to agree to without the court’s interference.

Here, the liquidated damages clause in ERA 100 was indeed such a “hybrid” obligation. If, for instance, Denka had exercised its right to terminate with notice, then Denka would have had to pay the sum stipulated in the clause (as a primary obligation). But as the event giving rise to termination here was a repudiatory breach of contract, the liquidated damages clause imposed a secondary obligation on Denka. It was thus subject to the penalty rule, similar to the liquidated damages clauses in ERAs 99 and 101.

On this basis, the CA proceeded to the next question of whether all the liquidated damages clauses were penalties and hence unenforceable. The CA answered this question in the negative, as discussed in the next subsection.

(4)        The clauses were genuine pre-estimates of the likely losses and were enforceable

One of the principles espoused by Lord Dunedin was the greatest loss principle, i.e. whether the liquidated damages formula was an extravagant sum to impose on Denka considering the greatest loss that Seraya might suffer from breaches that engaged the liquidated damages clause. The liquidated damages formula here was 40% x (the number of remaining months in the contract duration) x (the average of the amount payable by Denka in a stipulated period prior to termination).

Denka argued that because Seraya’s contractual profit had to be shared with YTL under the CFD, that had to be taken into account in accurately assessing Seraya’s nett profit and hence loss from the breach. However, the CA held that because the CFD was a purely internal arrangement that was not strictly enforced, it was irrelevant in determining whether the liquidated damages clauses were a genuine pre-estimate of the likely loss.

The CA then compared the liquidated damages clauses with common law damages. As regards ERA 100, Seraya would be entitled to claim damages for the loss of the whole contract because Seraya had accepted Denka’s repudiation. As regards ERAs 99 and 101, the measure of damages to which the terminating party would be entitled depended on the event that triggered the contractual right of termination. Here, the trigger for the contractual right to termination – Denka’s repudiation – concurrently gave rise to the common law right of termination, such that Seraya would be similarly entitled to damages for the loss of the whole contract.

The CA also discussed how these common law principles of damages related specifically to the context of analysing a liquidated damages clause. The assessment of the genuineness of such a clause must be decided at the time of contract formation, not breach. Thus, hindsight reasoning should be avoided. The true question was whether the liquidated damages clause was extravagant in comparison to the greatest loss reasonably anticipated at the time of contracting, which was answered through construction and interpretation of the relevant clause. Here, the clauses that gave rise to the contractual right of termination were drafted widely enough that they implicitly contemplated serious breaches – which would simultaneously give rise to the common law right of termination.

Further, the CA held that the 40% multiplier in the liquidated damages formula was justified. Based on the evidence of Seraya’s expert, the 40% multiplier not only fell within the range of estimated annualised losses, but was also on the lower end of the range. Denka was unable to dispute this. In particular, the expert evidence rested on cogent and reasoned computations and assumptions, and only reflected Seraya’s (not YTL’s) loss. The estimated annualised losses were arrived at taking into account the amount Seraya could earn by selling to an alternative party, and Denka’s contractual right to hedge the electricity price and Seraya’s need to unwind any hedges in the event of termination.

Finally, the CA considered the “single lump sum” principle, another principle elaborated on by Lord Dunedin in Dunlop. The CA emphasised that a large sum payable only gives rise to a presumption that it is a penalty. Ultimately, the greatest loss test is of overarching importance. If the court finds that the clause is not extravagant or out of all proportion to the greatest loss test, this should lead the court to the conclusion that the clause is a genuine pre-estimate of loss. Thus, while the stipulated sums here were large and appeared to violate the single lump sum test, the presumption was rebutted as the CA had already found the clauses not to be extravagant or disproportionate.

As genuine pre-estimates of Seraya’s likely losses, the liquidated damages clauses were enforceable. The sums stipulated under the clauses were thus awarded to Seraya.

IV.       Others

In light of the liquidated damages awarded, it was not necessary for the CA to determine the common law damages payable. Finally, the court also found against Denka for the following:

  • Denka’s offer to carry on with the ERAs despite its own repudiation was not a reasonable mitigation offer, as no additional benefit was offered to Seraya. Seraya was also entitled to reject that repudiation and require Denka to perform its obligations, as it did.
  • Seraya’s conduct in waiting to accept Denka’s repudiation was reasonable in the circumstances. It had repeatedly sought confirmation on DSPL and DAPL’s position during that period and thus the ERAs subsisted up till Seraya’s termination. Denka must pay the contract price for the electricity during that period, and not the lower market price.

V.        Conclusion

In coming to its final decision to award the full liquidated damages of over $30 million, the CA made the following key findings. First, Denka had breached the ERAs when it repudiated its obligations. Secondly, pursuant to that repudiation, Seraya then terminated the contracts, and the liquidated damages clauses imposed a secondary obligation on Denka. Thirdly, Seraya was entitled to the full liquidated damages payable, because they were a genuine pre-estimate of the likely loss Seraya would suffer.

VI.       Lessons learnt

The judgment clarifies the Singapore courts’ position with respect to the penalty rule. Parties should therefore draft their new contracts in line with this development and be mindful of it when assessing their rights under their existing contracts. This author suggests three key takeaways:

(a)        Primary obligations are clearly not subject to the penalty rule. Parties should thus be careful in agreeing to a contract term which provides that a sum must be paid if an event occurs that is not the non-performance of a contractual obligation. Such clauses may be enforceable by the other party, even if the sum is extravagant or disproportionate.

(b)        Parties cannot circumvent the penalty rule by drafting wide liquidated damages clauses that apply in any situation where the contract is terminated. The court may find these clauses to be “hybrid” obligations and ultimately have recourse to the underlying reason for the termination. Further, where a party finds that they have to terminate a contract, it may well be the case that a repudiation of the contract is more advantageous than a lawful exercise of a termination right. Parties should seek legal advice on this before deciding which course to take.

(c)        When a party seeks to enforce the liquidated damages clause, it must be able to show that the stipulated sum is a genuine pre-estimate of the losses that it is likely to incur. In this regard, the only “legitimate interest” is that of compensation. However, compensation should not be understood too literally or mechanistically. Indeed, it continues to include both pecuniary and non-pecuniary interests.

– – – For a PDF version of this article, click here. – – –

Written by: Jeremy Chai Zee Peng (3rd-Year LLB Student), Yong Pung How School of Law, Singapore Management University.
Edited by: Nicholas Liu (Lecturer), Yong Pung How School of Law, Singapore Management University.

[1] A contract for differences is one in which a party pays another the difference between the price of the same good, at the start and at the end of the contract.

[2] In Andrews, the court relied on the historical fact that the penalty rule originated in equity, to justify extending the scope of the rule to include situations not relating to breach of contract.

[3] The penal bond was a written document which recorded one party’s unilateral promise to do something by a set date, unless an accompanying condition was satisfied. The fulfilment of such condition would then render the bond void.

[4] In general, primary obligations are obligations imposed by contracting parties on each other, whereas secondary obligations are obligations that arise if the primary obligations are not fulfilled. For instance, A promises to do an act for B, and also agrees to pay damages to B if A fails to complete the act on time. The completion of the act by A is considered the primary obligation, while the obligation by A to pay damages is the secondary obligation.

[5] In general, the doctrine of unconscionability allows for a contract to be set aside if it was procured by unconscionable means: see the leading Singapore case of BOM v BOK [2019] 1 SLR 349.