IRB Brasil Resseguros SA v CX Reinsurance Company Ltd

Under a double-proviso follow settlements clause, the reinsured must prove on the balance of probabilities that the compromised claim fell within both the original insurance and the reinsurance. The court examines the factual and legal basis on which the claim was compromised, rather than re-litigating the underlying claim. The evidence required is fact-sensitive and may include reasonable inferences and circumstantial evidence. Allocation of a loss across policy periods may be established by a reasoned or approximate methodology. Whether losses arise from one event depends on the contractual context. An arbitration award is read reasonably and commercially, without an overly technical search for errors.

The claimant challenged an arbitration award arising from excess of loss reinsurance contracts covering casualty risks from 1976 to 1983. The arbitrators had found that compromise settlements concerning product liability, asbestos and environmental pollution claims were recoverable under the reinsurance.

The appeal raised four issues: the standard and factual basis for proving compliance with a double-proviso follow settlements clause; proof that losses occurred during the relevant period of cover; the allocation of losses between policy years; and the meaning of losses arising out of one event.

1. Outcome. The appeal was dismissed. Despite occasional infelicities in the arbitrators’ language, their reasoning and conclusions were legally unchallengeable.
2. Both provisos in the follow settlements clause had to be proved on the balance of probabilities. For the first proviso, the relevant inquiry concerned the factual and legal basis of the claim as compromised. The court did not need to determine whether the original claim would certainly have succeeded if fully litigated. The arbitrators had correctly distinguished proof of the legal requirements from the fact-sensitive evidence needed to satisfy them, applying [1996] 1 AC 1239 and [2009] EWHC 2787.
3. The arbitrators’ references to arguability did not alter the applicable standard. Read fairly and commercially, the award found that the compromised claims probably fell within both the underlying insurance and the reinsurance. The central findings concerning the breast implant and blood-product settlements were factual findings supported by evidence.
4. As to allocation, the arbitrators were entitled to infer that losses fell within the relevant policy years from the available evidence, including contractual allocation methodologies, long-tail liability and a straight-line approach. The burden of proof had not been reversed. Express citation of every relevant authority or policy clause was unnecessary where the reasoning showed that the issue had been considered.
5. Whether losses arose from one event depended on the contractual context. The arbitrators were entitled to treat the determination to engage in asbestos-insulation activities as an aggregating event and, alternatively, to rely on the finding that the claim had been settled as a single loss. Their reference to cause rather than event was an obvious verbal error, not an error of law.
6. The court applied the commercial approach to reading arbitration awards stated in [1985] 2 EGLR 14: awards should not be construed as statutes or subjected to meticulous nit-picking.

The claimant appealed to the Commercial Court from an arbitration award concerning recoverability under excess of loss reinsurance contracts. The High Court dismissed the appeal.