There will have been a sigh of relief in reinsurance circles this week following the decision of the House of Lords in Lexington Insurance Company v AGF Insurance Ltd [2009] UKHL 40.

Lexington is a US insurer which had covered pollution liability of Alcoa at 58 sites (45 in the USA and 13 outside it) dating back to the 1940s. Specifically Lexington wrote insurance cover for the period 1 July 1977 to 1 July 1980 and likewise London reinsurers of Lexington (AGF and Wasa) wrote reinsurance cover for that risk. Notwithstanding the limited period so stated, the Supreme Court of Washington found that, as a matter of the proper construction of the direct insurance policy, Lexington was liable for all loss or damage manifesting itself during the policy period, even damage which started before policy inception. In other words, Lexington found itself liable for pollution whether it occurred before or during the policy period, and therefore liable to a much greater extent than it presumably had in mind when its underwriters wrote the risk some 30 years ago.

In England, the Court of Appeal in turn found that the London reinsurers had likewise incurred this extended liability. The Court of Appeal focused on the proposition that a term in the underlying insurance was to be construed as operating in exactly the same way in the reinsurance cover.
The House of Lords unanimously thought differently and reversed the Court of Appeal’s decision. In so doing, the House distinguished its own previous decision in Forsikringaktieselskapet Vesta v Butcher [1989] AC 852. In Vesta, it had held that the legal effect of a term in the underlying insurance had to be the same in the corresponding reinsurance, even where the governing laws of the two contracts were not the same and that of the underlying insurance (Norwegian in that case) gave a different effect to that which the law of the reinsurance (English in that case) would give.
In Vesta, the effect of Norwegian law was well known at the time when the reinsurance was written. By contrast, in Lexington, the House of Lords found that even the question of what the governing law of the underlying insurance was could not have been predicted with any certainty by the reinsurers writing cover back in 1977. It just happened to turn out to be Pennsylvania due to circumstances which could not have been known at the time.
The uncertainty in turn meant that the overall outcome was itself uncertain. Even now, there is a plurality among US states in that some apply the approach taken by the Supreme Court of Washington in Lexington, while others take a pro rata approach, dividing liability among insurers according to the time periods for which they provided cover, when compared with the wider period in which loss or damage occurred.
As Lord Collins of Mapesbury put it, there was “no principled basis” for interpreting the effect of the policy period in the reinsurance contract in the same way as had been done in Washington for the direct insurance.
In effect, this was a step too far for the Vesta principle and the meaning of the policy period definition in the reinsurance contract had to be its ordinary meaning as used in the London reinsurance market. Otherwise, absurd commercial consequences could result, such as that the liability of a reinsurer would be the same whether it had written cover for the full period of Lexington’s insurance, or only for 3 months of it.
If the Court of Appeal’s judgment had been allowed to stand, the reinsurance market could still have found ways of drafting round the problem in the long term, but the fall out in the meantime would have been considerable. For the time being at least, while some US states may take the wider approach, in England reinsurers can take some comfort that a policy period will mean what they expect it to mean, cover for damage occurring in the policy period.