Reinsureds in the London Market Excess of Loss spiral did not have to prove losses fell within the cover of all underlying contracts, only their inwards contracts and their reinsurance contracts. Recoverable losses could be proved on the balance of probabilities by using actuarial models.

Equitas Limited v R&Q Reinsurance Company (UK) Limited; Equitas Limited v Ace European Group Limited

  • [2009] EWHC 2787 (Comm)

Help with citations

Facts

This test case concerned 26 excess of loss (XL) reinsurance contracts caught up in the London Market Excess of Loss (LMX) spiral.

XL reinsurance cover is triggered when the ultimate net loss incurred by the reinsured exceeds a specified amount (known as the attachment point).

Losses originating from the grounding of the Exxon Valdez oil tanker in 1989 and the Iraq invasion of Kuwait in 1990, however, included losses that were not properly recoverable under the policies (in the case of Exxon) and losses wrongly aggregated together (in the case of aircraft stranded at Kuwait airport at the time of the invasion).

It was some years before these problems came to light. In the meantime, significant losses had already been claimed and paid throughout the market. LMX claims still in the market, however, have been in lockdown until a solution can be found.

It is generally accepted that it would now be impossible to reconstruct the LMX spiral so as to work out whether, once the wrongly aggregated and irrecoverable elements are removed, the relevant attachment point under each individual contract had been reached, or to determine the proper amount now due.

The question was whether Equitas (as the assignee of the rights of the reinsured Lloyd's syndicates) could recover under these reinsurance contracts by using actuarial modelling to prove, on the balance of probabilities, the syndicates' recoverable losses.

R&Q said that, as a matter of law, unless Equitas could prove that the sums claimed were properly due, syndicate by syndicate and contract by contract, it could recover nothing at all. Estimation and guesswork would not do. In any event, the actuarial models used by Equitas were flawed and did not achieve their purpose.

Equitas argued that how it proved its claims was an evidential question of fact, not law. It was entitled to use the best evidence available in the circumstances, which in this case meant the actuarial models.

The LMX spiral

The LMX spiral has been described as "a multiple game of pass the parcel" (Deeny v Gooda Walker Limited [1994]).

Many syndicates who wrote excess of loss reinsurance in the London market passed on their liabilities, over and above the level of their individual retention, by taking out XL reinsurance. Those reinsurers took out their own XL reinsurance cover with reinsurers who, in turn, obtained their own XL reinsurance, and so on.

Within the LMX market, a smaller group of participants reinsured each other at multiple levels, creating a complicated network of mutual reinsurance known as the spiral.

Claims entered the spiral when a major loss occurred and the primary insurers claimed on their XL reinsurance. Liability would then be passed from reinsurer to reinsurer until it eventually reached those participants who had used up all their XL reinsurance protection.

The effect of a loss repeatedly "hitting" successive layers of reinsurance as it circulated the spiral was to magnify the original loss many times over. 

The Exxon and KAC claims

On 24th March 1989, the oil tanker Exxon Valdez ran aground off the coast of Alaska, causing a major oil spillage. Exxon's claims against its insurers were finally settled in 1997. Reinsurance claims relating to Exxon Valdez entered the London market from the early 1990s and began to be passed round the LMX spiral under the market code "Cat 89G".

Many reinsurers, however, argued that claims for the cost of pollution clean-up fell outside the terms of the underlying insurance and, therefore, could not be covered under their reinsurance contracts.

In 2005, the Court of Appeal confirmed that the insurance policy did not provide cover for clean-up operations (King v Brandywine Reinsurance Co [2005]). This meant that some of the losses grouped together under the Cat 89G code were irrecoverable and should not have been included in claims entering the LMX spiral.

The other claim concerned the seizure of 16 aircraft during the invasion of Kuwait on 2nd August 1990. Fifteen aircraft belonging to Kuwait Airways (KAC) and spares worth about US$150 million were flown to Iraq over a period of time. One aircraft belonging to British Airways, however, was left at the airport and was badly damaged by bombing in 1991.

The London market gave the KAC/BA losses a single market code "Cat 90V". For about five years, Cat 90V claims were made and paid by insurers and reinsurers within the LMX spiral on the basis that they arose out of one event, the invasion of Kuwait.

But in 2003, the Court of Appeal held the loss of the KAC aircraft and spares and the loss of the BA aircraft arose out of two separate events and should not have been aggregated together (Scott v Copenhagen Re [2003]).

Unable to untangle the irrecoverable and wrongly aggregated elements from recoverable losses, the market stopped settling claims and the LMX spiral ground to a halt. 

The reinsurance contracts

Of the 26 reinsurance contracts involved in this test case, 14 were said to be "tainted" by the wrongly aggregated KAC/BA claims and 12 by the inclusion of irrecoverable Exxon losses.

All the contracts incorporated Joint Excess Loss Committee (JELC) clauses, under which the reinsurer promised to indemnify the reinsured in settlement of its net loss. "Net loss" was defined as the sum paid by the reinsured in settlement of loss, after deduction of any salvage and recovery, including recoveries from other reinsurance. "Loss" was defined as a loss, damage, liability or expense arising from any one event.

In addition: "It is condition precedent to liability under this contract that settlement by the [reinsured] shall be in accordance with the terms and conditions of the original policies or contracts." 

Many of the contracts also included a standard "settlements clause", which stated: "All loss settlements by the [reinsured]… shall be binding upon the Reinsurers, providing such settlements are within the terms and conditions of the original policies and/or contracts… and within the terms and conditions of this Reinsurance".

R&Q argued that, in order to satisfy the first part of this clause, Equitas needed to represent correctly aggregated and properly recoverable losses upwards through the spiral, showing at each stage that the losses fell within the terms and conditions of the underlying policies.

Equitas accepted that no indemnity could be recovered for those parts of the syndicates' settlements which fell outside the terms and conditions of the immediate underlying contracts written by the syndicates (the "inwards" contracts) or under the reinsurance contracts. But, it argued, the clause did not impose a burden of proof that required the loss to be proved at every underlying stage of the spiral.

The judgment

The judge agreed with Equitas.

He concluded that the reference in the settlements clause to "original policies and/or contracts" referred only to the inwards contracts written by the reinsured syndicates, not to contracts further down the reinsurance chain or to the original insurance contracts.

The settlement clause required proof that the settlements were within the cover provided by the inwards policies. It did not mean that Equitas could only satisfy the burden of proof by re-presenting losses upwards through the spiral. 

The judge drew a distinction between the extent of the cover under the inwards contracts and the reinsurance contracts (which was a question of law) and the evidence required to show the claims fell within that cover (which was a question of fact).

The question was whether, on the balance of probabilities, Equitas could establish that the claim was properly within the scope of the cover and/or the underlying limits had been exhausted. How it attempted to do so was a question of fact or evidence, not law.

In this regard, the judge was unable to accept that that the use of actuarial models was wrong in principle. Equitas was free to deploy what evidence it chose to satisfy the burden of proof.

"Though actuarial models are regrettably expensive and undoubtedly add to the cost and the complexity of a case in which they are used, I can see no proper logical or principled objection into the use of the models here," he said. 

"If the choice facing Equitas was to abandon its claims (because the LMX spiral could not be reconstructed) or to seek to make good its claim by using a model, I can see no reason why it should be precluded from making the attempt."

After a detailed consideration of the actuarial models and the specific objections raised by R&Q, the judge concluded that the models provided a reasonable approximation of reality, in that they produced reinsurance programmes that were reasonably typical of those expected to be found in the actual LMX spiral – at least for participants without extreme or exceptional characteristics.

Choosing the more conservative of the two alternative scenarios created by the models, the judge found on the balance of probabilities that 86.5% of the syndicates' losses relating to KAC/BA and 75% of the losses relating to Exxon Valdez were recoverable under the reinsurance contracts.

Commentary

The decision goes back to the basic point that each reinsurance contract is an independent transaction, separate from the underlying contracts of insurance or reinsurance.

The judge's interpretation of this settlements clause means that the only settlements that fall within its scope are those made under the reinsured's inwards contracts. Even if errors have been introduced lower down the reinsurance chain, the reinsured may still be able to show, on the balance of probabilities, that the attachment point triggering its own reinsurance cover has been reached by properly recoverable losses.

The use of actuarial modelling to achieve this sets an important legal precedent. Although the judge warned that the technique was "complex, expensive, imperfect and, for my part, not ideal", the decision may nevertheless start to unwind the spiral for the numerous LMX claims that have been awaiting the outcome of this test case. The judge clearly saw the models as helping to provide "practical justice", a solution he said was "emphatically preferable to leaving the losses to lie crudely where they fall".

We are processing your request. \n Thank you for your patience. An error occurred. This could be due to inactivity on the page - please try again.