Skip to content
Perspectives

StandpointInternational Business Law

Consequential damages in international contracts

Almost every international contract excludes consequential damages. Almost nobody can say what that excludes. Where the term comes from, why it keeps surprising both sides, and what to draft instead.

Peter Poleacov (updated )13 min

Aerial view straight down onto a dam, water spilling over its crest.

Gloucester, England. At the City Flour Mills, the crankshaft of the steam engine breaks, and the mill stands still. The proprietors, Joseph and Jonah Hadley, have no spare part. The broken shaft must travel to the manufacturer as a model for the replacement. They hire the carrier firm of Joseph Baxendale, which promises to deliver the shaft on the second day. It delivers days late. The mill stays idle, the Hadleys sue for the profits lost in the extra days of standstill, and a jury awards them more than 20 times the carriage fee. The Court of Exchequer sets the jury’s verdict aside, and in doing so writes one of the most famous judgments in the common-law world.1

Every waiver of “consequential damages” in use today descends from that broken crankshaft. That makes it all the more remarkable that after 170 years of case law across the common-law world, nobody can say with confidence what the term covers. After two decades of negotiating these clauses, I consider the bare term “consequential damages” a liability in itself. Whoever writes it into a contract gets a dispute about words instead of an allocation of risk.

One case, two limbs

The rule of Hadley v. Baxendale is a limitation of liability built into the common law itself. A breaching party is not liable for every consequence of the breach; damages are limited to what is reasonably foreseeable, on one of two grounds: (a) the loss flows naturally from such a breach “in the great multitude of such cases,” or (b) it arises from special circumstances that were communicated to the breaching party when the contract was made.2 The Hadleys failed to recover their lost profits because, in the court’s view, they had not communicated to Baxendale that the entire mill stood still without the shaft.

The first limb became known as “direct” or “general” damages, the second as “special” or “consequential” damages – even though none of those four terms appears anywhere in the judgment.3 Two consequences follow, and both are routinely overlooked.

First, damages that are remote, speculative, or unforeseeable are recoverable under neither limb. That holds by law, waiver or no waiver.

Second, lost profits are not automatically consequential. Profits that any party would obviously lose from this kind of breach are direct damages under the first limb, and as a rule survive a consequential damages waiver.

The gap between the two marks is the dispute the clause creates instead of preventing.

What almost everyone gets wrong

Glenn West, who has studied these waivers more thoroughly than anyone, calls the exclusion “alien vomit,” after the nickname of an invasive sea squirt that overruns whole colonies. The boilerplate attaches itself to contracts on both sides of the Atlantic and crowds out the provisions next to it that someone actually thought about.3 The misconception catalog is long. Deal professionals equate consequential damages with remote losses (wrong: never recoverable anyway), with lost profits (wrong: see above), with “indirect” damages, with everything beyond replacement cost, and there is case law loosely supporting each version, which is precisely the problem.

Two decisions show what the confusion costs.

In 2 Entertain Video v. Sony DADC, rioters set fire to a warehouse during the London unrest of August 2011. It held the entire home-entertainment stock of a BBC Studios subsidiary. The storage contract excluded “indirect or consequential loss,” including loss of profits – but only, said the clause, to the extent such losses were indirect or consequential to begin with. The court held that the lost profits and business interruption were the direct and natural result of the fire, first limb, not excluded. The waiver failed at the very moment it was written for.4

In New York, the manufacturer Conor and its distributor Biotronik had excluded “consequential damages” in their exclusive distribution agreement. Conor assumed Biotronik’s lost resale profits were thus gone. The Court of Appeals held that the profits Biotronik claimed were general damages, recoverable despite the waiver. In West’s assessment, the decision took some New York lawyers by surprise.5

Ken Adams collected what such clauses actually list in practice and arrived at 31 items: consequential, special, direct, general, indirect, incidental, punitive, exemplary, collateral, and delay damages; plus loss of profits, loss of anticipated revenues, loss of savings, loss of goodwill, loss of data, loss of use, loss of production, loss of contracts, loss of business reputation, lost opportunities, cost of capital, increased operating costs, downtime costs, loss of use of equipment, damage to equipment, personal injury, property damage, substitute goods, substitute services, substitute facilities, and claims of customers.6

The problem is not the length of these lists but their construction. In them, items describing how a loss is characterized (consequential damages) stand next to items describing what the loss arose from (damage to equipment). They contain synonyms, because consequential and special damages are the same thing. They are contradictory, because excluding both direct and indirect damages excludes everything. And some of their categories do not exist at all. One example is “collateral damages.”

The cause is not malice but self-protection. Almost nobody drafts such a clause from scratch. It is carried over from the last contract, from the trade-association template, from the other party’s draft. Everyone who works on it adds one more item, because one more item cannot hurt and nobody will ever criticize them for it. The drafters want to protect themselves. The result is the opposite. The longer the list, the more synonyms and contradictions a court can use to construe the whole clause, or strike it. Adams calls it negotiation theater: precision work on the wrong front. I suspect that most of the people writing the waiver into their contracts today have never heard of Hadley v. Baxendale.

Drafting that works

Three rules cover most of the ground.

Name the losses, don’t label them

If you want to exclude loss of profit, loss of production, or loss of data, exclude them by name, as standalone exclusions – not as illustrations of “consequential damages.” The moment the list becomes a subset of the label, a court reads the label. It is then once again open which of the two limbs the court assigns the loss to. That is exactly where the clause in 2 Entertain Video failed.

Cap first, plus named exclusions

A monetary cap is the one limitation instrument whose meaning no court has to guess. Adams long argued for stopping there – and retracted that in 2022. The bare cap is a cop-out, and the concern behind the exclusion is legitimate. His current answer is the more interesting one. Instead of anchoring on the effect (“consequential damages”), he anchors on the cause, the special circumstances communicated to the obligor at the time of contracting. That is simply Hadley’s second limb, spelled out rather than labeled. Practitioners in plant engineering and software argue along the same lines. A cap alone is too blunt where a single defective component can trigger revenue losses far above the price, and it fails if the cap itself is held invalid. In practice, a cap and named exclusions therefore belong together. Each does its own work.6

Mind the carve-outs

In an NDA, never waive consequential damages. The loss from a leaked trade secret almost by definition consists of consequential damages. So the waiver takes away the very protection for which the parties enter into the agreement. The market standard is the opposite carve-out. In M&A, where remedies run through indemnification, say expressly that recoverable losses are limited to what contract law would allow under the foreseeability rule. Otherwise, with a broad “Losses” definition, this limit can quietly fall away.7

Assembled, this looks like the following clause, which I use in export contracts. I wrote it from the exporter’s side and generalized it. It covers only the exclusions and the exceptions to them. The cap belongs in a separate provision with the same exceptions.

The critical losses in paragraph 2 are excluded by name, each standing on its own; “consequential damages” appears only as a residual category at the end, where it can do no harm to the rest of the list. Paragraphs 1 and 3 exempt intent, gross negligence, personal injury, and mandatory law from the exclusions. In this clause, the label alone is deliberately not decisive.

The German twist

Under German law without the United Nations Convention on Contracts for the International Sale of Goods (CISG), this clause, like virtually any liability limitation of economic weight, is invalid the moment it qualifies as a standard term. In practice, it almost always does. Why it fails is set out in the article “Why your limitation of liability in B2B contracts fails”. An exclusion of liability for gross fault, and any exclusion or limitation for injury to life, body, and health, fail in standard terms even between businesses. Between businesses, Section 309 No. 7 of the German Civil Code (Bürgerliches Gesetzbuch, BGB) does not apply directly. But such a clause is measured against Section 307 BGB (Section 310(1) sentence 2 BGB). And for a simple negligent breach of an essential contractual obligation (a cardinal duty), the case law requires liability for foreseeable, contract-typical damages to remain intact.8 That is precisely the category the parties want to exclude with a consequential damages waiver. In commercial dealings, lost profits and production losses are the typical, foreseeable damages.9 The savings clause in paragraph 3 rescues nothing. German law does not reduce an overreaching standard term to the permissible maximum but strikes it, as the article “To the extent permitted by law? Not a good idea” shows.

So the honest advice under German law leads to three other routes. Negotiate the liability clause individually, genuinely and documentably. Do the limiting work one level earlier, in the statement of work. What is not owed cannot be breached. In its narrow core, the statement of work is the most powerful limitation instrument, because content review does not reach it there. How that works is set out in the article “Why contracts fail on the statement of work”. Or change the benchmark altogether. Under the CISG, limitation clauses are measured against the Convention’s flexible framework rather than the strict BGB model,10 which is one of the strongest reasons for German exporters to keep the CISG, as the article “CISG: advantages and disadvantages for German exporters” shows; to be safe, combine it with a supplementary law such as Swiss law plus an arbitration clause providing for a seat in Switzerland.11

The takeaway

The Hadleys lost out because of what they had not said. Most parties fighting over a consequential damages waiver today lose for the same reason. They wrote a label where they should have written a list, and each side read into the label what it hoped for. Exclude losses by name, cap the total, keep the carve-outs honest, and if you stay with German law without the CISG and use the liability clause as a standard term, accept that the real limitation work happens in the statement of work, not in the liability clause. If that is not enough for you, move up a level and negotiate the choice of law itself.

Notes

  1. Hadley v. Baxendale (1854) 9 Ex. 341, 156 Eng. Rep. 145 ([1854] EWHC Exch J70). Facts as reconstructed in Glenn D. West & Sara G. Duran, Reassessing the “Consequences” of Consequential Damage Waivers in Acquisition Agreements, 63 Bus. Law. 777, 784 f. (2008), and West (fn. 3). ↩

  2. Hadley v. Baxendale (1854) 9 Ex. 341: damages “should be such as may fairly and reasonably be considered either arising naturally … from such breach of contract itself, or such as may reasonably be supposed to have been in the contemplation of both parties, at the time they made the contract, as the probable result of the breach of it.” ↩

  3. Glenn D. West, Do You Really Know What “Consequential Damages” Means?, Weil Global Private Equity Watch, 18 May 2020: “In the thousands of cases considering this rule in the ensuing 166 years across the common-law world, it does not appear that anyone has really nailed this down.” ↩1 2

  4. 2 Entertain Video Ltd v Sony DADC Europe Ltd [2020] EWHC 972 (TCC). ↩

  5. Biotronik A.G. v. Conor Medsystems Ireland, Ltd., 11 N.E.3d 676 (N.Y. 2014); on the reaction in practice, West (fn. 3). ↩

  6. Kenneth A. Adams developed the critique over years: Excluding Consequential Damages Is a Bad Idea, 15 February 2010; Limitation-of-Liability Overkill, 5 March 2020; the reversal and his proposed solution, anchoring on the “special circumstances” rather than on “consequential damages”, in Here’s an Alternative to the Usual “Consequential Damages” Randomness, 20 June 2022, and in the Consequential Damages rider to MSCD5 (as of 7 November 2022). Each including the practitioner counterpoints in the comment sections. ↩1 2

  7. On NDAs: Kenneth A. Adams, Excluding Consequential Damages in a Confidentiality Agreement?, 18 October 2011. On M&A: West & Duran (fn. 1), 786 ff.; Glenn D. West, Consequential Damages Redux: An Updated Study of the Ubiquitous and Problematic “Excluded Losses” Provision in Private Company Acquisition Agreements, 70 Bus. Law. 971 (2015). ↩

  8. BGH, judgment of 18 July 2012 – VIII ZR 337/11. ↩

  9. Leuschner, NJW 2016, 1222 (1223). ↩

  10. OLG Köln, judgment of 9 January 2025 – 8 U 46/23 (on a clause making the delivery date essential): where the content review asks whether a term departs from a statutory model, the CISG’s provisions are the benchmark within its scope. ↩

  11. For a comparative treatment of the construction and validity of exemption clauses under German, Swiss, and English law, see Patrick Ostendorf, RabelsZ Online First 3/2025 (open access); section III covers the effect of the BGH’s cardinal-duty case law on the validity of liability limitations under German law, section IV Swiss law, and section V English law. ↩

Reference: Poleacov, P. (2026). Consequential damages in international contracts. INN.LAW. https://inn.law/en/perspectives/consequential-damages/