A recent High Court decision has potentially allowed losses in the LMX spiral to be paid after a logjam that has lasted for several years. Gross J has held - setting an important precedent - that Equitas can recover reinsurance losses based on modelled calculations of the LMX spiral, rather than being required to prove the outcome of each and every loss at each layer going up through the spiral.

The LMX spiral was created by several excess of loss underwriters taking out excess of loss cover for their own risks, the resulting losses being passed back and forth between those involved. The result was “a complex intertwining network of mutual reinsurance” (Phillips J in Deeny v Gooda Walker Ltd [1994] CLC 1224).

In this case, the High Court was asked to decide whether the fact that paid claims which had entered the LMX spiral including both wrongly aggregated and irrecoverable losses precluded Equitas (the assignee of various rights of Lloyd’s syndicates) from recovering under reinsurance contracts for otherwise recoverable losses “tainted” by the above historic errors.

The underlying claims

The case concerned the treatment of indemnities under retrocession contracts in respect of paid claims originating from the Exxon Valdez oil spillage and the first Gulf War.

  • The Exxon Valdez

The running aground of the Exxon Valdez in 1989 led to significant claims against Exxon, who, in turn sought indemnity from their insurers. The insurers settled Exxon’s claims under a compromise agreement. Arguments followed over liability under various reinsurance policies. The matter subsequently went to the Court of Appeal and - in King v Brandywine Reinsurance Co [2005] EWCA 235 - it was held that the losses were not within the scope of the direct policy and that claims had been paid out wrongly.

  • The First Gulf War - loss of Kuwait Airway Corporation and British Airways planes

During the Iraqi invasion of Kuwait in 1990, 15 Kuwait Airways Corporation (KAC) aircraft were seized by the Iraqi army. A British Airways (BA) plane which was stranded at Kuwait airport was later destroyed in fighting which followed the Allied invasion. In subsequent litigation (Scott v Copenhagen Re Co (UK) Ltd [2003] Lloyd’s Rep IR 696) it was determined that claims relating to the loss of these aircraft had been settled on the basis of an incorrect analysis of aggregation - the BA loss should not have been aggregated with the KAC loss.

As a consequence of both the wrongly paid out Exxon Valdez and incorrectly aggregated KAC/BA claims, the LMX market had been in “lockdown” with no claims being paid as “bad” claims and “good” claims were intertwined, and it was supposedly impossible to determine whether the relevant limits of the underlying policies had been reached so as to trigger the reinsurance coverage.

The arguments

It was Equitas’s case that it was entitled to recover from its retrocessionaire, R&Q, on the basis that it was able to prove its recoverable losses on the balance of probabilities through the application of actuarial modelling. This modelling provided for discounts which stripped out both the wrongly aggregated and irrecoverable losses, leaving the appropriate amounts due under the contracts.

R&Q argued that Equitas was unable to recover anything unless it was able to prove, contract by contract, at every layer of the spiral, precisely how much it was entitled to recover under the policies. The losses of different syndicates could not be proved by the application of an actuarial model which was incapable of replicating the LMX spiral.

The decision

The questions before Gross J were whether as a matter of law the claims fell within the terms of the underlying contracts and, if so, was it necessary to prove the precise loss under each and every contract in the spiral?

The judge held that as a matter of law Equitas was obliged to ensure that the claims in respect of which it sought to recover were within the terms of both the underlying contracts and reinsurance contracts. Once that was established, however (and there was no dispute between the parties that - apart from the “tainted” Exxon Valdez and BA losses - the claims were legitimate in principle), the indemnity recoverable under the retrocessions could be proved as a matter of fact by acceptable evidence.

Gross J went on to find that although actuarial modelling was imperfect, it produced an acceptable basis on which to establish properly recoverable losses. It was not necessary for Equitas to go further and to demonstrate strictly as a matter of law that each syndicate's loss for correctly aggregated KAC settlements and recoverable Exxon Valdez loss settlements exceeded the relevant attachment point of each relevant reinsurance contract and, if so, by how much. It was permissible to lead actuarial evidence to quantify, on the balance of probabilities, the amount of loss recoverable under the retrocessions.