Late-delivery LD clauses work under Chinese law — but they are adjustable, and they cannot be stacked on the deposit penalty. How Article 585 works in practice, and how to draft a number that survives judicial review.
Nearly every purchase contract with a Chinese supplier carries a late-delivery LD clause. Nearly none of them was drafted with a Chinese courtroom in mind. The buyer's template says something like "0.5% of the contract value per day of delay" — an aggressive number designed by a buyer who will never collect it — and the supplier signs it because both sides know it will never be tested. Then delay happens, the number gets invoked, and the supplier's first serious response is: have a Chinese court reduce it.
The supplier is usually right that a court can. Whether it will — and by how much — depends on how the clause was built. Here is the framework.
Start with the good news, because it matters: liquidated damages clauses are fully recognized in Chinese law. Article 585 of the Civil Code of the PRC (中华人民共和国民法典) provides that where the parties agree that a breaching party shall pay liquidated damages of a certain amount in case of breach, or agree on the method of calculating damages arising from the breach, the breaching party pays accordingly.
But Article 585 then does something many common-law drafters find surprising: it builds in judicial adjustment. If the agreed liquidated damages are excessively higher than the loss actually caused, the court or arbitration institution may, on the party's request, appropriately reduce them; if they are lower than the loss, the party may request an increase. The clause is a starting point, not a ceiling — Chinese law treats agreed damages as a reasonable pre-estimate that the court can recalibrate, not as a fixed price of breach that stands however far it drifts from reality.
Three practical consequences follow. First, the clause will not be voided for being a "penalty" — the common-law anxiety simply does not apply. Second, the number can be cut, and the requesting party has an incentive to ask. Third — and this is the drafting lesson — a clause anchored to plausible, provable loss is far harder to reduce than one floating free. Which brings us to the anchor.
The starting point for sizing an LD clause should not be "what number scares the supplier." It should be the question a court will actually ask: what loss would this delay plausibly cause, and is the clause proportionate to it?
For late delivery, the buyer's real losses are usually concrete and documentable:
If the LD rate you write can be justified as a rough proxy for these, the clause reads to a court as a sensible pre-estimate. If the rate is an order of magnitude above anything the relationship could produce, the clause reads as leverage — and Article 585 exists precisely to trim leverage dressed as damages.
One drafting device that helps enormously: put the foreseeability on the record in the contract itself. A recital stating that the buyer has disclosed its resale commitments, seasonality, or project deadlines — and that the parties agree delay losses of the kind covered by the LD clause were reasonably foreseeable — will not guarantee anything, but it deprives the supplier of the "I couldn't have known" narrative that usually accompanies a reduction request.
The market-standard structure in China-related supply contracts is also the structure that ages best in court:
Late Delivery. If Supplier fails to deliver the Goods by the Delivery Date, Supplier shall pay liquidated damages equal to [0.5]% of the value of the delayed Goods per week of delay, up to a maximum of [5]% of the value of the delayed Goods. Payment of liquidated damages shall not relieve Supplier of its obligation to perform. After the maximum is reached, Buyer may terminate this Order as to the undelivered Goods and claim damages.
Why this shape works:
Honestly: there is no bright line, and any article that quotes one should be read with care. In judicial practice, adjustment is a fact-driven discretion exercised against the actual loss, the parties' performance of the contract, the party's fault, and expected interests — and the burden of showing that the agreed amount is excessive sits, as a practical matter, with the party requesting reduction. Courts will look at real loss evidence; suppliers requesting reduction must come with more than indignation, and buyers defending a clause should be ready with their own loss arithmetic.
What experience suggests, without pretending to a formula: aggregated LD obligations in the low single digits as a percentage of the delayed value are rarely disturbed when the delivery was late and the losses are ordinary; rates that accrue daily without a cap, or totals that dwarf any documented loss, are the ones that get cut. The strategy is not to find the maximum survivable number. It is to write a number you would not be embarrassed to defend with invoices — which, conveniently, is also the number a supplier's manager will treat as real during negotiation, which is where most LD clauses actually do their work.
A trap for the unwary: many contracts contain both an LD clause and a deposit (定金) with the statutory deposit penalty — and buyers sometimes invoke both for the same delay. Article 588 of the Civil Code forecloses that: where the parties agreed both on liquidated damages and on a deposit for the same breach, and one party breaches, the other party may elect to apply either the liquidated damages clause or the deposit clause — not both (with the deposit returned if the LD route is chosen).
For buyers, the takeaway is to know in advance which remedy is worth more in each scenario. The deposit penalty on a 30% deposit is a large, fast, self-executing remedy — the money is already in the supplier's hands, and the statutory rule doubles it: the party receiving the deposit who breaches returns double; the party paying who breaches loses it. The LD route may total less, but preserves the deposit and lets you claim performance. Elect deliberately, in writing, and on advice — the election is consequential and, once made, the route not chosen for that breach is gone.
And note the terminology trap from the other direction: a Chinese supplier's contract that calls the advance payment 定金 in the Chinese text and merely "deposit" in the English translation has quietly armed the penalty against you. We have flagged that mismatch more than once in contract reviews — the English page said "deposit," the Chinese page said 定金, and the two pages were describing different legal animals. This is treated in detail in the deposit rules article.
Set expectations before drafting, because LD clauses get over-sold inside buying organizations too. Liquidated damages compensate; they rarely compel. If your production line stops because the shipment is three weeks late, the LD stream running in the background does not start the machines — and Chinese courts are, as a practical matter, more comfortable awarding money than supervising a continuous supply obligation from a foreign plaintiff. The clause's real power is upstream of litigation: a credible LD regime changes a supplier's dispatching priorities at the exact moment your container is competing with three others for the same production slot.
LD also does not collect itself. A judgment or award for liquidated damages is worth the supplier's solvency, nothing more — which is why the serious remedies discussion (asset preservation before suing, enforcement against the right entity) belongs to the same conversation as the drafting discussion. A perfect clause against a shell company is a perfect zero.
Finally, LD does not fix a broken specification of time. If the contract says "delivery: 45 days" without defining from what event, and the PO and the proforma invoice disagree, expect the delay dispute to be fought about the clock before anyone reaches the rate. Fixing the trigger definition costs one clause; litigating its ambiguity costs a hearing.
| Parameter | Drafting default | Why |
|---|---|---|
| Accrual rate | ~0.5% of delayed-goods value per week (daily only for perishable/seasonal) | Defensible as a pre-estimate; rarely reduced |
| Cap | ~5% of delayed-goods value, then termination right | Caps read as reasonableness; termination is the real teeth |
| Base | Value of the delayed goods, not the whole order | Prevents inflation that invites reduction |
| Excusable delay | Force majeure only, narrowly defined | Keeps "supplier problems" outside the clause |
| Remedy interaction | LD vs deposit penalty: election per Article 588 | No stacking; know your better remedy in advance |
| Performance | LD does not excuse performance | You're buying delivery, not a discount |
An LD clause is one of the few contract terms that behaves exactly as drafted — until the day it is tested, when it behaves exactly as adjusted. Drafting it with Articles 585 and 588 in view, and with the loss evidence architecture behind it, is routine work in a contract and trade document review. And if the delay has already happened, the escalation path — demand, remedy election, preservation, and litigation — is covered in litigation and enforcement and in what to do when a supplier doesn't deliver.
This article is general information, not legal advice, and does not create an attorney–client relationship. Statutory references are to the Civil Code of the People's Republic of China as currently in force; adjustment of liquidated damages is fact-specific and no outcome is assured. Always confirm current law and practice with counsel.
Send us the contract before your next order. We'll check the trigger definitions, the rate and cap, the deposit interaction, and the remedy election — the four places where LD clauses quietly fail.
Review my contract