Liquidated damages and penalties in Australia

That single divergence is the reason Australian penalties law cannot be read off an English or North American textbook. It is High Court authority and it does not vary by state, contract form, or industry. It applies to a delay liquidated damages clause in an AS 4000 subcontract exactly as it applies to a bank fee. The doctrine has two stages: a gateway that decides whether the penalty rule is engaged at all, and a ceiling that decides whether the stipulated sum survives. Australia's gateway is wider than England's. Its ceiling is, after a decade of apparent convergence, close to identical.

The practical stakes are ordinary drafting decisions, not exotic ones: what triggers the LD clause, how the rate was calculated, whether the clause says it is the exhaustive remedy, and whether the extension-of-time mechanism sitting next to it is actually being used. Get those four things right and the Australian authorities so far are on the drafter's side. Get the trigger wrong. Particularly by copying an English drafting fix that assumes breach is a precondition, and the clause is exposed to a doctrine England no longer runs at all.

Stage one: the Andrews gateway does not require a breach

Andrews v Australia and New Zealand Banking Group Ltd [2012] HCA 30 restored the older equitable penalty doctrine that England's Supreme Court later abandoned in Cavendish Square Holding BV v Talal El Makdessi. The Australian gateway asks, as a matter of substance and not form, whether the stipulation is:

1. collateral or accessory to a primary stipulation operating in favour of the other party
2. triggered by the failure of that primary stipulation. Breach is sufficient but not necessary, and
3. such as to impose an additional detriment on the first party, for the benefit of the second, in the nature of security for, and in terrorem of, performance of the primary stipulation.

There is a drafter's escape route built into the gateway itself. The doctrine does not apply where the payment is not a collateral detriment at all, but the agreed price of an additional right, accommodation or alternative performance conferred on the payer. The conditional-benefit analysis. A payment structured as the cost of an option the payer is free to take up sits outside the doctrine altogether. A payment structured as the cost of the payer's default does not.

Stage two: the Ringrow/Paciocco proportionality ceiling

Once the gateway is engaged, the question becomes proportionality. Paciocco v Australia and New Zealand Banking Group Ltd [2016] HCA 28, building on Ringrow Pty Ltd v BP Australia Pty Ltd, asks. Judged objectively and ex ante, at the time of contracting. Whether the stipulated detriment is "out of all proportion" to the legitimate interests the provision exists to protect: "extravagant, exorbitant or unconscionable", or showing "a degree of disproportion sufficient to point to oppressiveness".

The width of "legitimate interests" is the part of Paciocco most likely to surprise a foreign drafter. Interests are not confined to damages recoverable for breach. They extend beyond Hadley v Baxendale remoteness, are not limited by common-law causation, need not have been foreseeable to the counterparty, may be non-monetary, and may embrace the protected party's enterprise-level financial position (Kiefel J at [29] and [65], Gageler J at [161]–[162], and Keane J at [271]–[273]). The benchmark is the greatest loss that could conceivably be proved to follow. Not the loss actually suffered, and not the loss most likely to be suffered. Mere disproportion is not enough, and the onus of proving the penalty sits on the party alleging it.

Paciocco moved the ceiling, not the gateway: the point most commentary gets wrong

Paciocco is widely described as realigning Australia with England. That description conflates two different axes, and getting them backwards is the most common mistake a foreign-trained practitioner makes on an Australian file.

The safe formulation is: Australia kept the gateway England abandoned, and then adopted England's ceiling. Genuine pre-estimate of loss survives, but only as a safe harbour: a clause that is a genuine pre-estimate will not be struck. Not as the operative test for whether a clause is a penalty at all.

Stage three: relief is severance, not voidness

Where a stipulation fails the ceiling, the consequence in Australia is narrower than in England. The clause is unenforceable only to the extent of the excess over the loss actually provable. It is not wholly void, and the protected party may still recover its proved loss. A penalty finding in Australia does not hand the paying party a windfall. It puts the protected party to proof of what it actually lost.

That has a consequence a drafter should plan around rather than only litigate around. Because the fallback is severance and proof of actual loss rather than outright loss of the remedy, a principal facing a penalty challenge is rarely staring at zero recovery. It is staring at the cost and delay of proving up its loss the ordinary way, with the certainty and the cash-flow advantage of the liquidated sum gone. That is a real loss even when the clause is only partly cut down, and it is the reason the drafting checklist below is worth following even on a clause a principal expects would ultimately survive a challenge.

The construction cases: every reported delay LD challenge in this corpus has lost

The doctrine is High Court law, but it is tested in construction disputes against ordinary delay liquidated damages clauses, and the reported record so far runs one way.

Layer: jurisdictional. Australian case law is the least-verified block in this corpus. AustLII, NSW Caselaw, Jade and hcourt.gov.au returned 403 during research. Every citation above should be checked against a primary source before it is relied on in submissions.

What a construction LD clause must do to survive

Two further points sit outside the checklist because they are not yet settled by authority. Grocon had a tiered LD structure, and tiering and capping by severity [INFERENCE] is a plausible way to answer the single-obligation attack, though no decision tests it directly. And where the counterparty is a consumer or small business, UNVERIFIED the unfair contract terms regime should be checked separately. A clause can survive the penalties doctrine and still be void as an unfair term.

"No breach" triggers: a live drafting risk, and an English reflex that backfires

Because Andrews removes the breach requirement, a stipulation triggered by a non-breach event: a milestone that is not itself a contractual obligation, or a service-level abatement regime. Is within the Australian doctrine even though the identical drafting would sit outside the English one entirely.

That produces a trap for anyone trained on English precedent. The standard English response to a penalty risk is to restructure the trigger so that it is not a breach. That takes the clause outside Cavendish altogether. In Australia it does not. Removing breach from the trigger does not remove the stipulation from the Andrews gateway, and it may forfeit the one argument that would otherwise have helped: that the sum is a genuine pre-estimate of loss flowing from a breach. The drafting response that works in London can leave a clause worse off in Sydney.

★ NEGATIVE FINDING No Australian construction decision has yet struck down a delay stipulation on the Andrews no-breach ground. This is the honest limit of the doctrine as applied so far: commentary identifies the risk, and abatement regimes are the named example, but it has not crystallised into a reported construction outcome. The risk is doctrinal and live, not a demonstrated result. And a clause should be drafted defensively against it precisely because no court has yet had to decide the point.

The drafting answer is the conditional-benefit structure, not the no-breach structure. Make the payment the price of an additional right or accommodation conferred on the payer, rather than a collateral detriment securing the payer's performance. That is the escape route the gateway itself provides, and it does not depend on removing breach from the trigger.

How Australia compares: three common-law systems, three different answers

Australia's severance-based relief is a genuine point of difference from both its common-law neighbours, and the comparison is worth stating because it decides what "unenforceable" actually means to a client.

Australia, Canada and the United States have all declined Cavendish, and each for a different reason. Canada runs a two-limb genuine-pre-estimate test with an unconscionability overlay. The United States is anchored in a Restatement provision that does not move because England moved. Australia went the other way entirely: it restored an equitable doctrine England had discarded, then borrowed England's proportionality language for the ceiling it applies once that doctrine is engaged. The result on relief is a third pattern again. Where Australia severs the excess and lets the rest stand, an American court enforces the clause whole or throws it out whole, with no judicial reduction available either way.

For an owner or contractor pricing risk on an Australian job, that last row is the practical takeaway. A penalty finding in Australia is not the windfall it would be if the whole clause fell away. The protected party keeps its proved loss, and the practical exposure for the paying party is the gap between the struck rate and what could actually be proved, not the whole liquidated sum.

None of this changes the primary conclusion for anyone drafting or reviewing a delay LD clause today. The construction cases so far. Grocon, Growthbuilt, Leighton, Abgarus. All went the protected party's way, and the onus sits with the challenger throughout. The doctrinal risk sits at the gateway, not in the reported outcomes: it is the clause built around a non-breach trigger, with no documented rate derivation and no operating extension-of-time mechanism, that would be the first to test the Andrews ground a construction court has not yet had to decide.

About this material Library content is general information about construction claim practice, not legal advice. Entitlement, deadlines and procedure are governed by your own contract and by the law of the place the work is performed. Reviewed September 1, 2026 · Report a correction