Limitation of liability clauses are often the subject of extensive negotiations between business to business parties of an IT contract. It is difficult to escape the conflict between the competing priorities of customer and supplier - the customer aims to secure the maximum protection available against future losses but the supplier wants a level of liability to match the perceived value of the project to it. How wide can the supplier go to limit its liability and what does a customer need to do to ensure fairness while maintaining some certainty as to the losses potentially recoverable?

Limitation of liability clauses are an important tool for balancing the risk between the parties and limiting that exposure. The significance of these clauses and the importance of getting them right in this context are underlined by the fact that much of the leading case law in this area has involved software development contracts. The courts are often asked to determine the reasonableness and enforceability of limitation of liability clauses in the context of an IT contract dispute.

When things go wrong on an IT project (often as a result of inadequate deliverables, functionality issues and defects, delays and scope creep), limitation of liability clauses become crucial. At that point it is often the IT supplier who needs to reduce its exposure to the customer – either by relying on certain clauses in the contract excluding types of loss (usually special or indirect losses although note that loss of profits can be a direct loss where it is a natural result of the breach) or by imposing a financial cap on the overall liability. The damage suffered by a customer if an IT project fails can often exceed the cost of the software.

The approach of the courts

One of the leading cases in this area, Watford Electronics v Sanderson CFL (2001) involved the supply of software products alleged to have been delivered late and with a number of serious defects. Watford Electronics claimed damages in excess of £5.5m. Sanderson had sought to limit its liability and the Court of Appeal was reluctant to interfere with what had been agreed between two sophisticated parties with equal bargaining power, holding that unless one party had taken advantage of the other, or the term was so unreasonable as not to have been properly understood, the court should not interfere. This was seen as a move away from the interventionist approach seen in Peglar v Wang a year earlier, when the court noted that the burden of proof was on the defendant to show that the claims fell within the exclusion of liability condition on its true construction. A more recent decision of the Court of Appeal in Persimmon Homes v Ove Arup & Partners (2017) supports the hands off approach. The court noted here the growing recognition that the parties should be free to allocate risk as they see fit and, where the clauses are clear, said the courts should give effect to their meaning rather than approaching them with a default mind-set to cut them down. That said, the concept of reasonableness and satisfying the test under the Unfair Contract Terms Act 1977 (UCTA) remains important (see below).

Drafting limitation and exclusion clauses

It is essential that these clauses are drafted clearly, without ambiguity and with proper consideration as to what is permitted by law and/or what is reasonable if they are to be effective and enforceable. Remember:

  • Liability for death or personal injury resulting from negligence cannot be excluded or restricted.
  • Implied terms relating to the quality or fitness for purpose of the supplied goods or services can only be excluded or limited if reasonable - what is reasonable will depend on a number of factors including the availability of insurance, the strength of the parties' bargaining positions and the circumstances which were or ought reasonably to have been known or in the contemplation of the parties at the time the contract was made.
  • Financial caps on liability will be considered by reference to a party's availability to meet the foreseeable liability and the level of insurance cover available. The position adopted by the court in St Albans v ICL remains good law. Here the authority sued ICL for damages in excess of £1m caused by the faulty software and ICL sought to rely on its standard limitation of liability cap of £100k. The limit was held to be unreasonable under UCTA. The parties did not have equal bargaining power and ICL had insurance cover of up to £50m which clearly bore no relevance to the cap of £100k. Further, St Albans could not obtain better terms from a different supplier as other suppliers were offering similar caps on liability. A widely drafted clause with a low financial cap is unlikely to be enforced leaving the software developer exposed to potentially unlimited losses. A well drafted clause following a proper analysis of the risk/reward available to the parties under the contract should go some way to achieving an element of certainty on the potential financial exposure in the event an IT project fails.

Exclusion clauses need to be drafted with care. The context of the agreement as a whole should be considered and the cap needs to be tailored to the perceived risks of the transaction as well as the financial reward on offer. Parties quite often agree on a financial cap which is linked to the cost of the goods or services and limit aggregate liability as a multiple of the overall cost. The supplier's limit in its insurance cover can also be used as a measure although care should be taken to make any financial cap on liability an express term.

So be careful not to adopt a one size fits all approach when drafting and negotiating a limitation of liability clause. It is also not always about what should be excluded. If certain heads of loss are clearly expressed to be included with reference to a properly considered financial cap - the parties can achieve a level of certainty from the beginning as to those losses which are potentially recoverable if the IT project fails.