Limitation of liability clauses: cost and likely outcome depend on how the cap is drafted, what category of loss it excludes, and whether the breach reaches gross negligence, which no clause can reach under Swedish law. Clauses that fail in practice are usually built around a market template rather than the deal's actual risk allocation, and the gap between the two surfaces only once a real claim is made.
Who this concerns
This concerns any commercial party negotiating a contract where the downside of getting it wrong outweighs the value of the deal itself: acquirers and sellers in a share or asset transaction, suppliers signing long-term framework agreements, and service providers taking on obligations with an uncapped exposure buried somewhere in the small print. The question rarely comes up while the contract is being negotiated. It comes up later, when a counterparty claims a loss the cap was supposed to cover and the wording turns out to say something narrower than either side assumed at signing.
This sits squarely inside Lodline's contracts and transactions practice, because a limitation clause is negotiated, drafted, and eventually litigated within the same contract lifecycle those matters cover. The situation is asymmetric by design. The party drafting the clause writes it to its own advantage; the party accepting it rarely reads past the cap figure to the carve-outs sitting underneath. Both problems surface at the same moment, usually during a dispute, when it is too late to renegotiate the wording and the only question left is how a Swedish court or arbitral tribunal will read what was actually signed.
What the law says
Under Swedish law as it currently stands, freedom of contract governs how a limitation of liability clause is drafted between two commercial parties. There is no statutory cap that a clause must respect, and no requirement that a limitation be expressed as a fixed sum, a multiple of contract value, or a percentage of anything. What the clause says is, in the first instance, what applies.
That freedom has a ceiling. Liability arising from gross negligence or intentional misconduct cannot be excluded or capped by contract, regardless of how the clause is worded or how clearly the parties intended it to apply. A cap drafted to cover "all liability arising under this agreement, however caused" will not reach a breach a court characterises as grossly negligent; the wording simply does not get there. Between businesses of comparable bargaining power, courts are reluctant to override a negotiated cap on the basis that it produces a harsh result for one side, but that reluctance is not absolute. A clause can still be set aside, in whole or in part, under general contract law principles where enforcing it as written would be manifestly unreasonable given how the contract was actually performed and negotiated.
How it works in practice
What the cap actually covers
A limitation clause usually does three things at once: it sets a ceiling in money or as a multiple of a defined value, it defines which categories of loss the ceiling applies to, and it lists what falls outside the ceiling altogether. Disputes rarely concern the ceiling itself. They concern the second and third parts, because that is where the drafting tends to be thinnest and where a claimant looks first.
Direct loss, indirect loss, consequential loss
Most limitation clauses exclude "indirect" or "consequential" loss without defining either term, on the assumption that the distinction is settled. It is not. Lost profit, business interruption and reputational damage get argued by claimants as direct consequences of the breach and by defendants as indirect consequences too remote to recover. Where the contract does not define the boundary itself, the argument turns on the specific facts of what caused what, not on the label the clause happens to use.
Carve-outs that survive any cap
Certain categories are carved out of almost every limitation clause as a matter of course: breach of confidentiality, infringement of intellectual property, fraud, and liability that by its nature cannot be excluded. Where a carve-out list is incomplete, or was copied from an unrelated template, the gap becomes the argument. A claimant with a weak case on the merits will look first at whether the loss can be recharacterised into an uncapped category rather than argued on the facts as pleaded.
Gross negligence and wilful default
Because gross negligence cannot be excluded by contract, it becomes the default argument for any claimant facing a cap that would otherwise defeat the claim. The threshold for gross negligence is fact-specific and turns on how the breaching party actually behaved, not on how the contract describes the standard of care owed. A well-drafted clause anticipates this by addressing what conduct falls short of gross negligence with enough precision that the argument has less room to run.
Interaction with warranty and indemnity provisions
In a share or asset transaction, the limitation clause rarely stands alone. It interacts with separate warranty caps, indemnity baskets, de minimis thresholds and time limits for bringing a claim, often drafted at different points in the negotiation by different advisers. Where these provisions are not read together, a claim can end up governed by two clauses that point in different directions, and the resolution depends on which one a court treats as more specific to the loss actually claimed.
How a dispute over the clause is actually built
A dispute over a limitation clause is built the same way any contract dispute is built: from the wording outward. The starting point is what the clause says, read against the rest of the agreement. The next question is what category the loss falls into and whether that category sits inside or outside the cap. Only once those two points are settled does the argument move to how the parties negotiated the clause and what each side understood it to mean, which matters far less than most clients expect it to.
What to check now
Before relying on a limitation of liability clause, or before accepting one, the following should be checked against the actual wording, not against what the clause is assumed to say:
- Whether indirect or consequential loss is defined, or left to the label alone
- Whether the carve-out list matches the risks specific to this contract, not a generic template
- Whether the cap interacts with a separate warranty or indemnity cap in the same agreement
- Whether the time limit for bringing a claim under the clause is shorter than the general limitation period
- Whether the clause distinguishes liability in contract from liability in tort for the same conduct
- Whether the governing law and forum clause sit in the same document or point elsewhere entirely
Does a limitation of liability clause cover losses identified during due diligence?
No, not automatically. A limitation clause governs liability for breach of the contract it sits in; a loss identified during due diligence is usually addressed separately, through price adjustment, a specific indemnity, or an exclusion from warranty cover. Where the two overlap, how due diligence findings are typically handled before signing sets out how a finding gets addressed before completion rather than argued about afterward, once positions have hardened.
Can a limitation clause be challenged once a dispute reaches court?
Yes, though the bar is high. A Swedish court will generally enforce a negotiated cap between commercial parties as written. A challenge succeeds only where the clause fails to reach the loss on its own wording, the loss falls into gross negligence, or enforcing the clause as drafted would be manifestly unreasonable given how the contract was performed. How district and appeal courts decide contract disputes sets out which forum actually hears that argument first.
Does a limitation clause work the same way in a share deal and an asset deal?
Not quite. In an asset deal, liability attaches to specific assets and undertakings transferred, and the cap has to be read against what was actually assumed rather than against the target company as a whole. How asset deals allocate transfer undertakings sets out where that allocation is usually drafted too loosely for a cap negotiated at company level to apply cleanly to the assets actually changing hands.
The numbers
There is no fixed figure to attach to the cost of a limitation of liability clause going wrong, and any number offered without seeing the actual wording and the loss claimed would be invented rather than informative. What can be said is how the cost moves. It rises with the number of separate provisions that touch the same loss, the number of carve-outs argued to apply, and whether expert evidence is needed to characterise the loss as direct or indirect. It falls where the clause is short, the carve-out list is closed, and the loss claimed sits cleanly inside or outside the cap without room for argument on the label alone.
Time behaves the same way. A dispute confined to interpreting the clause moves faster than one that also has to establish what the breaching party actually knew or intended, because that is where a gross negligence argument gets built, and where evidence rather than wording becomes the deciding factor. Neither figure can be given in advance; both depend on the specific clause and the specific facts once a claim is on the table.
Where it usually goes wrong
A limitation clause usually goes wrong at the point where it was drafted to close a negotiation rather than to answer a dispute. Two patterns repeat. The first is a cap set as a round figure or a multiple of contract value with no relationship to the loss categories actually excluded underneath it, so that the ceiling and the exclusions were negotiated separately and never reconciled against each other. The second is a carve-out list copied from a different transaction, covering risks that do not exist in this contract while missing the ones that do.
Where the counterparty, its parent company, or the assets a claim would need to attach to sit outside Sweden, the analysis changes in a way that is easy to miss while drafting. A cap enforceable in a Swedish court means little if the counterparty's assets sit in a jurisdiction that does not recognise the judgment, or if the parent company that would actually pay is not a party to the contract at all. Recognition and enforcement abroad, and whether the contracting entity is the entity with anything to lose, need checking before the clause is relied on, not after a dispute has already started.
The clause also stops being the last word once a claimant recharacterises the loss into an uncapped category. A cap drafted around "breach of this agreement" does very little against a claim framed as fraud, misrepresentation, or breach of a separate statutory duty, even where the underlying facts are the same conduct the cap was meant to cover. Where that recharacterisation is plausible on the facts, the cap is a starting position for the negotiation, not the ceiling on the outcome.
What to do next
Reading the clause is not the same as knowing how it will be read once a real claim is on the table. Where the review stops being something to do alone is the point where the wording has to be tested against an actual set of facts, a specific counterparty, and what its assets would look like if a claim had to be enforced. A finding surfaced during diligence gets priced into the deal rather than left as an open question in a clause nobody has stress-tested. Get in touch to have the clause reviewed against the facts of the specific deal.