Follow The Settlements? The Liability Of Insurers And Reinsurers To Contribute To A Settlement Payment

Article by Simon Cooper and Alex Kershaw

The Lumbermens case involved the building of a shopping centre in Glasgow, during the course of which numerous disputes arose between Bovis (the insured contractor) and the developer (Braehead). Bovis issued various claims against Braehead, which in turn counter-claimed for various other sums.

A global settlement was reached, entailing a payment by Braehead to Bovis. However, the settlement agreement did not identify how the payment sum had been calculated and, in particular, did not make it apparent how, if at all, Braehead’s counterclaim had been taken into account in calculating the settlement figure.

Bovis’ insurer, Lumbermens, issued a claim for a declaration of non-liability to Bovis. Bovis counter-claimed against Lumbermens on the basis that Bovis was entitled to claim from its insurers a sum which amounted to the value of Braehead’s valid counterclaim. Bovis argued that this amount was to be calculated by setting off the actual settlement sum paid by Braehead against the damages which Bovis had been claiming from Braehead. While the details of the payment were not in the settlement agreement, Bovis sought to adduce evidence of their calculations of liability in the context of the settlement negotiations to demonstrate their liability.

Colman J rejected this approach. He held that a global settlement, such as the one which had been reached between Bovis and Braehead, did not satisfy the requirement that a loss be sufficiently ascertained if it is to form the basis of a recovery under a liability insurance policy. This was because the global settlement did not impose on the assured (Bovis) any identifiable loss in respect of any identifiable insured eventuality. It is necessary for a party to a settlement to be able to demonstrate by extrinsic evidence that he was under an insured liability, and that what was paid in settlement of that liability was reasonable. If the settlement agreement did not itself adequately identify the loss, the court could not evaluate the insured’s own assessments to demonstrate the validity of the settlement.

Colman J cited various authorities which emphasised the need for the liability to be certain, as it provided the link between the insured event and the payment under the insurance and reinsurance, and referred to his own judgment in the case of King v Brandywine Reinsurance Co (UK) Ltd [2004] EWHC 1033 (Comm), in which he had stated that in order for a settlement payment to create the necessary liability for which an assured could be indemnified under a policy of insurance, that payment should be the consideration for which the claim against the assured was dropped under the terms of the settlement agreement.

A settlement agreement for a global sum which did not allocate the payments or deductions made to the heads of loss and the claims and counterclaims of each party did not achieve that certainty and thus could not give rise to an insured liability.

The decision in Lumbermens may be contrasted with that of the US Second Circuit Court of Appeals in North River Insurance Company v ACE American Reinsurance Corporation (15 March 2004) where the US Second Circuit Court of Appeals held that once a settlement agreement had been validly entered into, how the payment under that settlement agreement should be allocated across a programme of (re)insurance was a decision which the (re)assured could make and (re)insurers would be bound by that decision, subject to it being made in good faith and each (re)insurer’s liability not exceeding the policy limits or extending to liabilities not indemnified under the policy.

North River had allocated all of a $332 million settlement payment within the first and second layers of a five layer insurance programme; the second layer had limits of $345 million. The reinsurers on the second layer disputed this allocation, arguing that the payment should be apportioned across all layers of the programme as the settlement had freed them all from risk of loss. The Court found that North River’s ‘rising bathtub’ (i.e. bottom up) allocation was entirely reasonable and in good faith, notwithstanding that they had modelled exposure to higher layers. Further, the Court found that the ‘risk of loss’, from which the higher layers had been freed, was not insurable.

As a matter of English law, a settlement agreement, in order to give rise to a right of recovery under a policy of (re)insurance, needs to be sufficiently tightly drafted so as to create a liability by linking the payment under the settlement agreement to the insured event under the policy. Therefore, if a factual situation similar to that in North River occurred in England, the authorities suggest that the allocation to the different layers of insurance which North River made was the only one which they could possibly have made. This is because the higher layers would not have come under any liability under the settlement agreement, in the same way that if North River had come under a liability after court proceedings to pay $332m in damages, that would be have fallen within the first two layers of the insurance programme (up to $345m), and only its reinsurers for those two layers could be expected to indemnify it.

The practical application of the Lumbermens case is that it is vital to ensure that a settlement agreement is tightly drafted and specifically describes and allocates any payment which is being made or taken into account under it, in respect of which the payer may wish to claim an indemnity from his (re)insurers; ie the payment obligation reflected in the settlement agreement needs to be identifiable as an insured liability. Further, as a matter of English law, in a situation such as that in North River, the allocation of a settlement payment to layers of (re)insurance should reflect the liability to pay and cannot impose liability where otherwise there would have been none.