Duty to comply with claims cooperation clause where insurer has denied liability/proving quantum


Weekly Update 19/12 reported the earlier instance decision in this case, in which Eder J held, as a preliminary issue, that the insurance policy in question did cover theft by employees, despite the insured’s non-selection of a “Theft by Employees Section” (and so the insurer couldn’t reject the claim on that basis). The insured therefore proceeded to bring a claim for business interruption losses. This was an unusual BI claim, because such claims are normally based on a single adverse event, such as a fire or flood or major burglary which is immediately known to the insured and which causes major losses of profit clearly reflected in the insured’s accounts. By contrast, here the alleged theft took place more than 500 times over more than 5 years (and spanning some 5 policy years) before discovery in December 2008, and the allegations of loss were not founded on accounts but instead on calculations based on various assumptions. Although the case is fact-specific, some general points of interest arose:

(1) Complying with a Claims Cooperation Clause (“CCC”). The policy contained a claims condition that (inter alia), “in the event of a claim being made”, the insured would provide particulars of its claim within 30 days after expiry of the Indemnity Period, and deliver accounts and other “documents proofs information explanation and other evidence as may be reasonably required” by the insurer to investigate the claim. It was also provided that “if the terms of this condition have not been complied with, no claims under this Section shall be payable” (and the term was accepted to be a condition precedent).

The insured argued that no claim could be made here until after discovery of the theft in December 2008, and hence there was no obligation to deliver particulars of the claim before that time. That argument was rejected by the judge. It did not matter that the insured did not discover the thefts until 30 days after the indemnity periods had expired under earlier policies. Particulars had not been sent until 17 February 2009 and so the insured had breached this requirement in respect of all thefts prior to 18 January 2008. However, although the judge accepted that insurers could have “pulled the shutters down” in relation to those thefts, he found that they had not done so on the facts. 

Eder J also approved textbook commentary to the effect that “full particulars” means “the best particulars the assured can reasonably give”, and (unless the policy states otherwise) further particulars can be supplied later on.

The judge did not decide the “very difficult issue of law which has never properly been considered by an English Court” as to whether an insured must comply with claims conditions if an insurer wrongly rejects a claim (and if that constitutes a repudiation of the insurance policy). Instead, he focused attention on the meaning of what was “reasonably” required? 

The judge was prepared to find that the list of various documents required by the insurers had been “reasonably required” “in an abstract sense”. However, he went on to hold that “it does not seem to me that this is necessarily so in circumstances where insurers in effect (wrongly) deny liability in principle or even (wrongly) refuse to admit that “employee theft” was an insured peril, as they did in the present case. I should emphasise that I fully accept that, generally speaking, it may be perfectly “reasonable” for insurers to reserve their position pending receipt of further documents/information and that a requirement by insurers that the insured should deliver such documents/ information may be entirely “reasonable” because a review of such documents/information by insurers is necessary in order to decide, for example, whether cover exists or not. In my view, that is a statement of the obvious but it is not this case. I should also make plain that I fully accept as a matter of the general law of contract that a repudiation is a thing “writ in water” and that, unless it is accepted by the innocent party, the latter will generally continue to be bound by the contract and, in particular, continue to be bound to perform that party’s continuing obligations under the contract. That is, indeed, trite law. However, here the focus is one of construction of the contract and, in particular, what information may be “reasonably required” by insurers in the particular circumstances of the present case; and unless and until the defendants were prepared to confirm at the very least that “employee theft” was an insured peril, the requirement to deliver the other categories of documents …. was, in my judgment, not “reasonable” having regard, in particular, to the time and expense that would have to be incurred by [the insured] in complying with such requirement”.

The judge did, however, find that the insured had breached the claims cooperation clause by failing to provide profit and loss and management accounts from 2005 to 2008, it being common ground that a request for such accounts is routine for commercial claims and hence were “reasonably required”. Although certain issues had been “parked” pending agreement on the question of liability, no additional costs would have had to be incurred in order for the insured to send the accounts.

Furthermore, there had not been unequivocal representation from the insurers that the insured was not required to deliver the accounts (and hence no waiver by estoppel). It did not matter that the insured thought or assumed that they did not have to deliver the accounts – that did not result from any agreement by the insurers. Accordingly, the claim failed because a breach of the CCC. 

Eder J also said that it was “difficult” to say the insurer had breached its duty of good faith because it ought to have warned the insured that it was committing a breach of a condition precedent. The judge agreed that that the insurer was under no duty positively to warn the insured that it needed to comply with policy terms, and there was also no evidence to support the suggestion that it should be inferred that the insurer’s silence was deliberate.

(2) Proving quantum. An insured must establish on a balance of probabilities that a relevant event has been caused by an insured peril. Here, the insured faced difficulty in reaching that burden and proving that thefts had even taken place, and the judge accepted that “there may be different ways of satisfying the legal burden and standard of proof other than by direct evidence”. However, here, although certain thefts (below the policy excess) had been proven, there was no evidence that an assumption made by the insured about the employee’s thieving activities during the relevant period was correct. Indeed, the employee would most likely have tried to keep the number of incidents of theft to a minimum in order to avoid the risk of detection.

COMMENT: As mentioned above, this case does not resolve the issue of whether an insured must comply with a policy condition even if the insurer has (wrongfully) rejected a claim. In Diab v Regent Insurance (see Weekly Update 16/06) the Privy Council held that a claims provision remained binding even where insurers had told the insured they would not pay the claim (and the case was followed in Lexington v Multinacional (see Weekly Update 22/08)), but there is textbook commentary doubting that approach in non-liability or reinsurance cases (see Colinvaux’s Law of Insurance, 9th edn, para 9-018) and Eder J was not strictly bound to follow either Diab or Lexington. Instead, this case focuses on what an insurer can “reasonably” require from the insured following the rejection of a claim, if the policy contains that wording (and even if the clause does not contain that wording, Napier v UNUM Ltd [1996] is authority for the view that the courts will imply a reasonableness test into any requirement for the insured to provide information to the insurer about its claim).

However, it might be said that by focusing the inquiry on the policy wording, the judge did implicitly accept that the insured was still bound to comply with a claims provision even though insurers had rejected the claim (even if he did not expressly say that was his conclusion). It is also noteworthy that the judge linked the issue of “reasonableness” in the policy condition to the insurer’s conduct regarding the claim in general: in short, where a claim has not been accepted, an insurer can’t reasonably request information which it will be costly or timely to produce (and query how much time and expense will result in a request not being reasonable).