What's the first thing lawyers turn to when reviewing a commercial contract, and invariably one of the last issues to get resolved? The exclusion of liability clause.

Important as these provisions are, it is all too easy to get swept along in the "custom and practice" of exclusion clauses and to lose sight of what the relevant law actually is and how the courts are currently interpreting this area of law.

The starting point

The 'rule' established in Hadley v Baxendale (1854), and subsequently split into two limbs, provides that damages can be recovered for breach of contract where either they:

  1. arise naturally from the breach: those reasonably foreseeable by an objective bystander - "direct losses"; or
  2. may reasonably be supposed to have been in the actual contemplation of both parties at the time the contract was made - "indirect losses".

All other losses are too remote and cannot be recovered.

Traditionally contract drafters have sought to exclude liability for all indirect or consequential loss on the basis that it would not be commercially acceptable for a party to be liable for losses which it could not reasonably foresee. Often the concept of "remoteness" is forgotten as parties feel that if they don't exclude "indirect" losses, their liability would soar.

An assumption of responsibility?

In 2008, the rules suddenly appeared to change! In the case of Transfield v Mercator Shipping (2008) (The Achilleas) a new concept of "assumption of responsibility" arose. The House of Lords looked at what liabilities the parties had chosen to assume (rather than just what they foresaw), and dealt accordingly.

A chartered ship was delivered back three days late. During the charter, market charter rates had doubled, and the owners had pre-let the vessel on a charter to start straight after its return at more lucrative rates. When the vessel was returned late, this charter was renegotiated at a lower rate, losing the owners $8,000 a day for the whole period of the next charter.

The House of Lords held Transfield was not liable for that loss. In the charterparty market, practice was that damages for late delivery were the difference between market and charter rate for the period of the delay only, and nothing more. Even though it was foreseeable that if delivery was late Mercator would lose money on a subsequent contract (i.e. it would appear to be a direct loss), the judge said those losses could not be recovered because Transfield had not "assumed responsibility" for them.

Fortunately, in Sylvia Shipping Co. Ltd v Progress Bulk Carrier Ltd (2010) the court held that any departure from traditional remoteness rules will be exceptional. Sylvia chartered a ship to Progress, who sub-chartered it to a third party, and lost out on a lucrative arrangement because Sylvia had failed to repair the ship properly in accordance with its contractual obligations.

The judge reviewed Transfield and held that there is a principle of "assumption of responsibility" bound up in the Hadley remoteness test which sometimes will apply but only in exceptional circumstances i.e. where the market, or custom and practice, dictated.

No assumption of responsibility applied in this case because there was nothing unpredictable about the losses or the extent of them, and damages on the basis of Hadley were awarded for the loss of the sub-charter. In summary, this means that the Hadley (foreseeability) rule is alive and well and remains the starting point for determining whether losses are recoverable or not.

A practical approach

So we still have Hadley. And GB Gas Holdings (Centrica) v Accenture (2010) now illustrates the practical approach that the courts take in determining whether particular heads of loss are direct or indirect. Accenture was engaged to provide and maintain a new IT system which included an automated billing system. Problems arose and customers were either billed incorrectly or not at all.

The Court of Appeal determined whether certain items of claimed loss were irrecoverable pursuant to a fairly standard clause excluding "consequential" and similar losses, such as loss of profit.

There were five heads of loss amounting to almost £30 million. They were gas distribution charges, compensation paid to customers, additional borrowing charges, the cost of chasing debts not due, and additional stationery.

Accenture argued that all these losses were indirect losses not being specifically within the parties' contemplation at the time the contract was made and so excluded. The Court of Appeal disagreed, holding that all of the losses arose naturally from the breach and were therefore direct losses.

Points to consider when drafting exclusion clauses

  • The courts will generally take a common sense approach to liability and direct losses assessing what they consider to be fair and equitable.
  • The distinction between direct and indirect losses is not always clear. Based on Hadley and GB Gas, indirect losses are a narrow and limited category of loss. It did not exclude loss of profit claims in GB Gas. The mere exclusion of indirect and consequential losses can therefore provide a false sense of security.
  • Remember that "loss of profits and revenue" are broad and generic terms and some losses, like "compensation paid to customers", can be interpreted as an expense or a loss of profit. The claimant can categorise it as it sees fit and plead it accordingly to limit the impact of an exclusion clause.
  • In large or risky contracts there may be a role for using the "assumption of responsibility" mechanism to include/exclude certain categories of liability. For example, if GB Gas Holdings had said "Supplier is not assuming any responsibility for any compensation you pay to your customers as a result of a breach of this contract by the supplier", then that would have been a much clearer indication of intent and harder for a judge to avoid.