The limitation of liability clause and contractual interpretation: case update

February 14, 2018
by Lyons Davidson

In this article, we revisit the case of The Royal Devon and Exeter NHS Foundation Trust v ATOS IT Services UK Limited [2017] EWCA Civ 2196, following the appeal of a judgment on a preliminary issue: namely, whether there was a ‘cap’ on the level of damages that might accrue under a breach of contract and the enforceability of the limitation of liability clause.

At first instance, the court found that one cap applied, contingent on when the cause of action accrued. The trust appealed the decision (on different grounds from the original submissions that the clause was not enforceable) and the Court of Appeal disagreed with the lower court’s decision about the meaning of seemingly inconsistent contractual provisions. This article focuses on the issue of whether there was a cap on the defendant’s liability and, if so, how it was to be determined.

Exclusion clause

Although this case does not blaze a trail of innovation, it does highlight the importance of an accurately drafted exclusion clause and provides a useful illustration of how the court interprets clauses within a contract.

The court will be concerned with the intentions of the parties when interpreting a contract and will look at what a reasonable person with the knowledge of the parties would have understood the contract to be by looking at the meaning of the words in their context. The same test was applied in both cases but different decisions were reached – it is a prime example of how the court’s interpretation of contractual provisions can differ.

Liability cap

The claimant, a hospital trust, contracted with Atos to provide an electronic management system for just shy of £5 million. Atos provided an inadequate system and the trust terminated the contract, seeking damages of £7.9 million for wasted expenditure incurred in reliance on Atos’ performing the contract.

Atos argued that damages should be capped by virtue of a clause in the contract that sought to cap its liability to the trust, which is broadly set out as follows:

Liability for any one default shall not exceed the figure in Schedule G;
The aggregate liability for all defaults shall not exceed the amount in Schedule G;
Schedule G provided that the total charges would not exceed:

for any claim arising in the first 12 months of the term of the contract, the total contract price; or
for claims arising after the first 12 months of the contract, the total contract charges paid in the 12 months prior to the date of that claim. [Our emphasis]

Atos argued that the above amounted to a valid limitation of liability and was, consequently, valid and enforceable, and that the clause would make sense if the words at paragraph c(ii) ‘for claims arising’ were substituted for ‘any claim arising’ which was consistent with the paragraph c(i).

Atos submitted that either:

There was one cap applying to all defects regardless of when they occurred; or
There were two caps, depending on whether the defaults occurred within the first 12 months of the contract or after.

The trust argued that the clause was not capable of being construed as a limitation of liability, as references in the clause to “claims” and “that claim” were contradictory, and it is not clear whether there is a single cap or a cap for each claim that might arise. In effect, the clause sought to define terms within itself that were themselves not defined, and it was therefore unenforceable through uncertainty.

Commercial contracts

In commercial contracts, the court will generally give effect to a clause that limits a party’s liability, as long as the clause is clear and unambiguous or can at least be construed as such. There is no presumption against parties having agreed to give up or limit their contractual remedies and, provided that clear words had been used, the court would give effect to the commercial allocation of risk in a contract.

Where there is more than one interpretation of a clause, the court will apply the one it deems more appropriate by looking at the intention of the parties and the wording of the contract itself. It is open to the court to prefer the interpretation that makes commercial sense but that is not to say that the court will redraft badly worded contractual provisions or correct a ‘bad bargain’ with the benefit of hindsight.

First decision: one cap

At first instance, the court accepted that paragraph c(ii) could be rectified by reading all references to claims plural in the singular.

The court then looked at how that paragraph would be read with the preceding paragraph and whether it provides for one, two or multiple caps when read together. Ultimately, the court agreed with Atos’ primary argument that there was one cap. The court’s reasoning was:

The use of the word “or” between the paragraphs and the words “aggregate liability […] shall not exceed” at paragraph b suggest that one cap should apply depending on which paragraph the relevant default fell within; and,
If paragraph c(ii) were to provide for multiple caps then paragraph c(i) would have to be read the same. This could result in damages far exceeding the contract value and would render the purpose of the cap useless.

Ultimately, the court found the clause to be valid and enforceable, and preferred the interpretation that made commercial sense: the clause provided one cap for all defaults with the level of cap depending on whether the first default occurred within the first 12 months of the contract or after.

Appeal decision

The trust sought permission to appeal and succeeded, abandoning its argument that the clause was so unclear as to make it unenforceable. Instead, it argued that the clause imposed two caps and not one.

The Court of Appeal, disagreeing with the lower court, decided that the clause provided for two caps:

The use of the word “aggregate” at paragraph b could mean that the aggregate liability totals the amounts in paragraphs c(i) and c(ii);
The use of the word “or” can be construed disjunctively or conjunctively and in this case the court construed it conjunctively. The two caps related to distinct periods of time which did not overlap;
It made commercial sense that defaults occurring within the first year would attract a higher cap than works in years two to five, as the work carried out within the first year was of higher value than work done in subsequent years;
If a default occurred in the first year that was equal to exceeded the cap in paragraph c(i), then there is no reason why the defendant should not be liable for further defaults occurring after the first year;
The appeal judge also commented that the analysis of the judge at first instance ran into difficulties in situations where there are defaults in the first year and subsequent years: which cap would apply? It would not be sufficient to determine which cap applied by reference to the date of the default, as suggested by the first judge, because paragraph c(ii) of the clause does not make provision for that.

Ultimately, the appeal court found that the natural meaning that produces the least-bizarre outcomes and accords with business common sense is that the clause imposed two caps.

This was decided as a preliminary issue and should narrow the issues in dispute, making it more likely that broad terms of settlement can be agreed. However, costly litigation could have been avoided had the contractual provisions been drafted with more precision. It is inevitable that parties turn their attention to what a contract actually means once a dispute crystallises but it is also worth pointing out that it would still have been open to the parties to agree what the exclusion provisions meant, which would have avoided the need for court intervention.

For more information on any of the issues raised in this article, please contact Nichola Board in the Commercial Litigation team by emailing [email protected] or calling 0117 394 5034.