Based in Jinan, Shandong, China | Serving clients worldwide
Legal basis: Civil Code of the People's Republic of China (中华人民共和国民法典), in force 1 January 2021, Articles 584, 585, 586, 587, 588, 591 and 594; Judicial Interpretation of the Supreme People's Court on the General Provisions of the Contract Part of the Civil Code (Fa Shi [2023] No. 13, in force 5 December 2023), Articles 64, 65 and 66; United Nations Convention on Contracts for the International Sale of Goods (CISG), Articles 6, 74, 75 and 77. Chinese text is quoted in translation prepared for this article; the Chinese text governs.
A supplier breaches, and the contract says 30 per cent. Buyers tend to assume one of two things: that the figure is enforceable as written, or that a Chinese court will treat it as an unenforceable penalty. Both assumptions are wrong, and the gap between them is where most of the money sits.
Chinese law does not recognise the common-law distinction between a genuine pre-estimate of loss and a penalty. Every agreed sum is enforceable in principle, and every agreed sum is reviewable on request. What decides the outcome is not how the clause is labelled, but whether you can document a loss, and whether the supplier's conduct was bad enough to shut down its own request for a reduction.
Three features of the regime are consistently under-read outside China, and they run in the buyer's favour more often than not. The 30 per cent benchmark is measured against the loss, not against the contract price. The burden of proving that the agreed sum is excessive sits on the supplier, not on you. And a supplier that breached in bad faith is generally refused any reduction at all, whatever the arithmetic shows.
In England, in Singapore, in most US states, the first question about a liquidated damages clause is whether it is a genuine pre-estimate of loss or a penalty in disguise. If it is a penalty, it is void, and the claimant falls back on proving actual loss. Draftsmen spend considerable effort on recitals designed to show that the figure was a real forecast.
That exercise is wasted on a contract governed by PRC law. Article 585 of the Civil Code provides that the parties may agree a sum payable on breach, or a formula for calculating the loss. It then gives both sides a right of adjustment: if the agreed sum is lower than the loss, the court or arbitral tribunal may increase it on request; if it is excessively higher than the loss, the court or tribunal may reduce it appropriately on request.
There is no category of void penalty. A clause calling itself a penalty is not struck down for that reason, and a clause calling itself a genuine pre-estimate earns no deference for that reason either. Chinese courts look at the commercial function and the relationship between the figure and the loss. The label is close to irrelevant.
The practical consequence is a shift in the whole dispute. In a penalty jurisdiction, the argument is about characterisation at the drafting stage. In China, the argument is about quantum at the proof stage. A buyer who spends its budget on contractual recitals and none on evidence has optimised for the wrong hearing.
Article 65 of the 2023 judicial interpretation supplies the operating test. Its first paragraph requires the court to take the loss defined in Article 584 as the base, and then weigh the contracting parties, the transaction type, the state of performance, the degree of fault and the background to performance, applying fairness and good faith. Its second paragraph supplies the number everyone quotes.
Article 65(2): Where the agreed liquidated damages exceed the loss caused by 30 per cent, the people's court may generally determine that they are "excessively higher than the loss caused".
Article 65(3): Where a party in bad faith breaches the contract and requests a reduction of liquidated damages, the people's court shall generally not support the request.
Read precisely, the threshold is not "30 per cent of the contract price". It is 130 per cent of the loss. If your provable loss is USD 100,000, a clause yielding USD 130,000 sits at the line, and a clause yielding USD 130,001 is presumptively excessive. If your provable loss is USD 10,000, a clause yielding USD 300,000 is not "30 per cent" of anything relevant; it is thirty times the loss, and it will be cut.
This is why the same clause produces wildly different outcomes across cases. The clause did not change. The evidence did.
Article 64 of the interpretation reverses the intuition most foreign counsel bring to the table. A breaching party that asks for a reduction carries the burden of proving that the agreed sum is excessively high. The non-breaching party that asserts the sum is reasonable must also provide evidence, but it is not required to go first.
Two procedural points follow, and both catch buyers out. First, the request for adjustment can be made by counterclaim or by defence, not only by a separate action, so the supplier does not need to sue you to get the figure reduced. Second, Article 66 requires the court to explain to a supplier that has argued the contract is invalid or that it did not breach, but has not asked for a reduction, that it may do so. The court is instructed to raise the point rather than let the supplier lose the argument by omission.
The Court goes further. Where the first-instance court found for the supplier without giving that explanation, and the appellate court decides that damages are payable, the appellate court may give the explanation itself and then reduce the sum after full argument. A defendant that did not appear at first instance for objective reasons and turns up on appeal asking for a reduction may still obtain one. The practical lesson is that a default judgment is not the end of the quantum argument in China, and a supplier that ignored the first round has not forfeited the second.
Article 584 defines the measure: compensation equals the loss caused by the breach, including the benefit the claimant would have obtained had the contract been performed, but not exceeding what the breaching party foresaw or ought to have foreseen at the time of contracting. Article 591 adds the duty to mitigate, and allows the cost of reasonable mitigation steps to be recovered.
For a buyer of goods, the loss is rarely just the difference between what you paid and what you got. The categories that survive scrutiny in reported cases are consistent:
What does not survive is a round number with no supporting documents. A penalty clause is not a substitute for a damages case; it is a shortcut through one, and the shortcut closes the moment the supplier puts the loss in issue.
Article 591 matters here in a way buyers underestimate. If you could have covered the shortfall cheaply and did not, the difference is deducted. If you sat on defective goods for eight months before notifying, the deterioration is on you. Mitigation is not a courtesy; it is a deduction the supplier is entitled to argue, and in practice it is the argument that does the most damage to an otherwise strong claim.
Article 65(3) is short and, for buyers, unusually valuable: a party in bad faith that breaches and then asks for a reduction will generally not get one. The provision does not define bad faith, and the reported decisions build it from conduct rather than from intention.
The patterns that have supported the finding include taking payment when the supplier knew there was no capacity to perform, accepting a deposit against goods already committed elsewhere, continuing to accept instalments while the delivery date had already become impossible, stripping assets or transferring the business to a connected company once the dispute began, and repeated breaches of the same obligation after written assurances. What these share is that the supplier's conduct at the time of contracting or during performance, not merely the fact of breach, is what the court reacts to.
The strategic implication is straightforward. If you have material of that kind, plead it as a distinct factual case with documents, not as a rhetorical flourish in the damages section. Once the finding is made, the 30 per cent arithmetic stops being the battleground. And if you have nothing of that kind, do not spend credibility asserting it; the court will read the file.
Most disputes that reach a Chinese court or tribunal are not about non-delivery. They are about late delivery or late payment, where the only real loss is the cost of money. In that setting, the agreed rate is tested against what the money would have cost, and the outcome is fairly predictable in range, though not in any single case.
Reported decisions show a consistent compression. A rate of 0.2 per cent per day has been reduced to 0.05 per cent per day. A rate of 0.2 per cent per day plus a lump sum has been reduced to 0.05 per cent per day with the lump sum disallowed as duplicative. A rate of 0.2 per cent per day has been converted to 24 per cent per annum. A rate of 0.05 per cent per day has been reduced to 0.02 per cent per day where the only loss was the use of the money. In each case the court started from the funding cost and worked outward, rather than from the contract.
Broadly, a clause in the range of 0.03 to 0.05 per cent per day tends to survive, because it approximates commercial funding cost plus a modest margin. A clause at 0.1 per cent per day is often reduced but not in every case. A clause at 0.2 per cent per day or above is exposed, and a clause at 0.3 per cent per day or above is close to a drafting error, because it invites a reduction that will be granted. These are observations about reported practice, not statutory bands, and there is no rule that fixes them.
Structuring the clause as a rate rather than a lump sum helps for a second reason. A rate that runs only until payment is easier to justify than a fixed sum that looks punitive on its face, and it keeps accruing while the dispute is pending, which is when the pressure on the supplier is greatest.
Article 588 allows a party to choose between the penalty clause and the deposit rules where both are stipulated. It does not allow both. Article 586 caps a deposit at 20 per cent of the contract value, and anything above that cap does not operate as a deposit. Article 587 then supplies the sanction: a recipient who fails to perform returns double the deposit.
Two drafting traps follow. First, a contract that labels a prepayment as a deposit, and provides a separate penalty for the same failure, has created an election the buyer did not intend to make. If the deposit is 20 per cent and the penalty is 30 per cent, the buyer may find that the deposit route returns less than the penalty route, but that having taken the deposit it cannot then claim the penalty. Second, if the deposit is genuinely insufficient, Article 588(2) permits a claim for the shortfall, but that claim is a damages claim and must be proved like any other.
The sensible arrangement, where the commercial position allows it, is to use the deposit as security for performance and keep the penalty clause for delay and quality, with the two covering different obligations rather than the same one.
Contracts between a Chinese seller and a buyer in another contracting state are governed by the CISG by operation of the Convention. A clause choosing PRC law does not, in mainstream Chinese practice, displace it: China's declaration limits only the Article 1(1)(b) route, and the Convention applies where both parties are in contracting states. Buyers routinely assume that a PRC law clause puts them in front of the Civil Code. Often it does not.
Under the Convention the damages rule is Article 74: compensation equals the loss, including lost profit, subject to foreseeability at the time of contracting. Article 75 supplies the substitute-transaction measure, and Article 77 imposes mitigation. The damages provisions are treated as exhaustive, in the sense that they displace the domestic law of damages.
Article 6 then permits the parties to derogate from the Convention or vary its effect, which is the door through which an agreed-sum clause enters. What is unsettled is how such a clause interacts with Article 74. One reading treats the agreed sum as a pre-estimate that displaces the Article 74 inquiry altogether, subject only to the general obligation of good faith. Another treats it as subject to the Article 74 ceiling and the foreseeability limit, with the result that the agreed figure is really a starting point rather than a fixed entitlement. Chinese tribunals and courts have not settled the question in a way that can be relied on across cases.
What follows for drafting is that a clause intended to operate under the Convention should present itself as an estimate of identified losses rather than as a sanction. Reciting the categories of loss the parties had in mind, and the basis on which the figure was calculated, gives the clause something to stand on under either reading. A bare percentage with no explanation does not.
The sequence that preserves the most value is unglamorous and time-sensitive.
Freeze the evidence first. The replacement quotation, the substitute contract, the invoice, the payment, the freight, the downstream penalty notice, the internal record of the idle line. Loss in China is proved document by document, and documents created after the dispute starts carry less weight than documents created while it was happening.
Write to the supplier with a computation, not a demand. A letter that asserts the clause and stops there invites a reduction argument. A letter that sets out the loss item by item, attaches the documents, and states that the sum claimed is a pre-estimate of those losses sets the frame before the supplier's lawyer does.
Plead in the alternative. Claim the agreed sum as primary relief, and loss plus interest as alternative relief. If the clause is reduced, the alternative claim is what keeps the case alive, and it costs little to plead at the outset.
Consider preservation before you serve. A supplier that is already moving assets will not wait for the judgment. Freezing measures are available at the outset of proceedings and are dealt with separately in the article on locating and reaching Chinese company assets.
Watch the clock. Article 594 gives four years for disputes arising from international sales contracts and technology import and export contracts, which is longer than the general three-year period. Four years is not a reason to wait. Evidence ages, counterparties dissolve, and the enforcement position deteriorates faster than the claim does.
The clause in your contract is neither a formality nor a fixed entitlement. It is a starting position that a Chinese court will test against your evidence, with the testing burden placed on the party that broke the contract. Buyers who lose on this point usually lose for one of three reasons: they claimed a percentage and did not build a loss file, they assumed a PRC law clause excluded the CISG, or they had facts showing bad faith and did not plead them as such.
The first is a preparation failure, the second is a drafting failure, and the third is the one that costs the most, because Article 65(3) is the provision that turns a reduced award into a full one.
If you are negotiating the clause now, the work is in the drafting and costs almost nothing. If you are already in dispute, the work is in the file, and it should start before the supplier's lawyer frames the loss. The general routes are set out under dispute resolution, the negotiation-stage alternative is covered in corporate counsel, and the practical steps for a supplier that has stopped performing are in recovering money from a Chinese supplier that stopped shipping. If the company has already been emptied, the analysis in capital reduction and who pays is the next step. Background checks before you commit are covered under due diligence.
Aaron Li (Li Xinlei) is a partner at Beijing Huicheng (Jinan) Law Firm in Jinan, Shandong, where he acts for foreign companies and individuals in international trade disputes, cross-border enforcement and construction claims. He has five years of international engineering and market development experience in the Middle East and South Asia, and writes on China trade lawyer practice from the buyer's side: what to check, what it costs, and what can actually be recovered.
Source note: the statutory text discussed is the Civil Code of the People's Republic of China (in force 1 January 2021), Articles 584, 585, 586, 587, 588, 591 and 594; the Judicial Interpretation of the Supreme People's Court on the General Provisions of the Contract Part of the Civil Code (Fa Shi [2023] No. 13, in force 5 December 2023), Articles 64, 65 and 66; and the United Nations Convention on Contracts for the International Sale of Goods, Articles 6, 74, 75 and 77. Chinese statutory text is quoted in English translation prepared for this article; the Chinese text governs. Statements about how courts have treated a point describe reported practice, not a rule of binding effect, and where the article says a question is unsettled that reflects the absence of a judicial interpretation on the point as at the date of writing.
Disclaimer: This article provides general legal information and does not constitute legal advice for any specific case. The law and judicial practice change, and the appropriate route depends on the facts of the individual matter, including the governing instrument, the date the claim arose and the evidence available. Please consult and appoint a qualified lawyer for your own case.
Related recommendations
China Full-time Lawyer&Attorney
Protect your china purchase and legal rights
RECENT POSTS
RECENT COMMENTS
Leave a Reply
Your email address will not be published. Required fields are marked *