Discovery Of Losses -A Cautionary Tale For Bankers Blanket Bond Underwriters

The case concerned the Bankers Blanket Bond (BBB) coverage carried by a Peruvian bank (BSP), which was insured locally and reinsured into the London market. Both the insurance and reinsurance were written on an amended DHP 84 wording - widely used in the London market. Like most bond wordings, this provided cover on a discovery basis, ie in respect of losses which the insured might " sustain or discover that they have sustained" during the policy period.

The relevant fidelity insuring clause covered losses sustained by reason of a "dishonest or fraudulent act of [an employee] committed with the intention of making improper personal gain for themselves." There was an exclusion for any claim "…arising out of any circumstance or occurrence known to the insured prior to the inception hereof and not disclosed to Insurers at inception," and a discovery limitation clause precluding liability in respect of claims arising out of such circumstances.

The dispute between reinsurers and the Peruvian cedants centred upon whether BSP had discovered a fidelity loss before or after inception of the policy. The transaction which caused the loss (the removal of a quantity of currency notes from BSP’s vault for lending to another bank in financial difficulties) was discovered one month prior to inception. The withdrawal was acknowledged at that time to have been unauthorised and ill-judged, and BSP’s internal auditor recognised that the employees involved might have received kick-backs, and recommended that they should be disciplined. However, BSP contended that it was not until after inception that they knew that the transaction involved employee dishonesty and could found a fidelity claim under the policy.

Only 10 days after inception, a "possible dishonesty" claim was notified by BSP following growing suspicion on the part of the auditor. A further unauthorised withdrawal was discovered some three months later and it subsequently emerged that the BSP employees involved in both transactions had personally received large sums.

Reinsurers avoided the reinsurance and contended that the underlying policy was void on the basis that the circumstances of the first transaction had been known to senior officials of BSP before inception but had not been disclosed. However, the claim under the underlying policy was later settled by the cedants, who commenced Commercial Court proceedings against reinsurers to recover their outlay.

Raymond Jack J held that for "discovery" to take place there had to be actual, subjective knowledge on the part of BSP that the necessary ingredients for a fidelity claim were present. Mere suspicion of dishonesty was insufficient. The judge accepted on the evidence that, at inception, BSP’s senior officials did not recognise that the relevant employees had acted dishonestly. Consequently the claim would attach under the policy only after inception, when the involvement of dishonest employees was clearly apparent, and BSP had not been guilty of non-disclosure.

Reinsurers argued that there should be an objective test for discovery which would apply if the insured discovered facts from which a reasonable insured would deduce that a potential fidelity claim arose. The judge rejected this argument in the absence of an express clause to that effect.

He also held, despite BSP’s knowledge of the transaction before inception, that neither the relevant exclusion nor the discovery limitation clause afforded reinsurers any protection because, again, these would only bite if BSP had actual, subjective knowledge at that stage that the necessary ingredients for a claim were present. The judge rejected reinsurers’ argument that "circumstance" in the context of those provisions should be construed more widely (although he conceded that this argument was supported by the natural meaning of the wording), because this could lead to a gap in cover for the insured. Such a gap would arise if a loss was notified under one policy but its dishonest aspects only emerged in the following policy period. The insured would then fall between two stools in that the first insurers could, in the judge’s view, reject the claim on the basis that dishonesty had not been established, and the second insurers could do likewise on the basis that the loss had been discovered during the prior policy period.

Reinsurers’ alternative argument was that an objective test should apply, and that any circumstance which would cause a reasonable person to consider that they had suffered a loss which could lead to a claim under the policy should be disclosed for these purposes. Again, the judge held that the wording provided no justification for the reading reinsurers suggested.

The reasoning which underpinned the judge’s conclusions was derived almost exclusively from BSP’s predicament as an insured facing a possible lacuna in coverage between the expiring and renewing policies. This consideration seems to have played a disproportionate part in the decision.

In fact, it seems probable that, if the circumstances had been notified under the expiring policy, the underwriters then on risk would have been bound to accept that any covered loss arising from those circumstances would have attached under the expiring policy. Although decided in the context of liability insurance, we would expect a court to apply analogous principles to those adopted in J Rothschild Assurance v Collyear ¹, namely that later events (for example the confirmation of initial suspicions of dishonesty) could be taken into account in determining whether the notification operated to attach the loss. In practice therefore, BSP would have been very unlikely to suffer any disadvantage had they notified the circumstances of the matter under the expiring policy which would clearly have been prudent.

The facts which had emerged before inception of the renewing policy, even though they gave rise to a suspicion of dishonesty at the lower end of the spectrum, were facts which most Bankers Bond underwriters would have regarded as material. The judgment appears to discount that materiality. The application of the subjective test of the insured’s state of knowledge, which was clearly treated as the determining factor in this particular case, therefore had the effect of significantly restricting the renewing underwriters’ ordinary rights of avoidance.

The result is that, because all the evidence of knowledge is in the insured’s possession, there would be nothing to prevent dilatory or unscrupulous insureds from simply failing to disclose suspicions about potential losses. This might even provide an incentive not to investigate such suspicions too closely, until BBB coverage is safely in place, or has been renewed.

In order to avoid the same predicament as the reinsurers in this case, underwriters should consider introducing an expressly objective element into attachment and other policy provisions, as is increasingly common in current wordings, for example: "Discovery occurs when the insured becomes aware of facts which would cause a reasonable person to assume that a loss has been or will be incurred."

Underwriters could also consider modifying the disclosure provisions in BBB proposal forms to impose on the insured an obligation to disclose circumstances which bjectively give rise to the suspicion that a covered loss has, or may have, occurred.

Underwriters may also wish to cater for the difficulty contemplated by the judge, as to the possible gap in cover. One answer might be to include an express deeming provision within the policy so that, if a covered loss does develop from a circumstance notified within a policy period, it will be deemed to attach under the policy then in force. Such a clause would be consistent with the principles applied in the J Rothschild Assurance v Collyear decision and likely to reflect the understanding of most insurers as to the effect of notification of circumstances.

1 [1999] 1 Lloyd's Rep IR 6