Who Knows?

As part of its consultation paper, 'Insurance Contract Law: The Business Insured's Duty of Disclosure and the Law of Warranties', the Law Commission has made some interesting comments on the imputation of knowledge to companies for the purposes of the duty of disclosure under section 18(1) of the Marine Insurance Act 1906.

These comments are worth noting, as they provide a useful and revealing analysis of this complicated area of the law.  They also appear to cast doubt on some of Rix J's reasoning in the well-known case of Arab Bank Plc v Zurich [1999] 1 Lloyd's Rep 262.

Section 18(1) provides: "The assured must disclose every material circumstance… which is known to the assured, and the assured is deemed to know every circumstance which, in the ordinary course of business, ought to be known by him".

There are two separate concepts here – actual knowledge ("which is known to the assured") and constructive knowledge ("the assured is deemed to know every circumstance which, in the ordinary course of business, ought to be known by him").

Actual knowledge

People can actually know things but, by definition, a company cannot.  Therefore, to work out what the company actually knows, one needs first to identify which natural person(s) 'count' as the company for the purpose of section 18(1).  In most circumstances their knowledge will then be attributed to the company.

In Arab Bank the insurers argued that the knowledge of the managing director (MD) of the company must surely count.  Rix J disagreed, for a number of distinct reasons.  One of those reasons was, effectively, that as the policy insured the directors as well as the company, the MD alone could not count as the company; he suggested that only the board as a unit could count. 

This was despite it having been held by Staughton LJ in an earlier Court of Appeal case (PCW Syndicates v PCW Reinsurers [1996] 1 WLR 1136) that, under Section 18(1), each of the directors would count as the company, as well as "employees whose business it was to arrange insurance for the company".  (In the Arab Bank case, it appears the MD was also responsible for arranging the insurance – he had signed the proposal form).

In the paper, the Law Commission argues that Staughton LJ's analysis represents the current law (indeed they go further and say that the current law is right in principle and should not be changed).  Arab Bank is not mentioned at all in the paper.

Constructive knowledge

Constructive knowledge only applies where the company does not have actual knowledge.  The issue is whether, in the ordinary course of the company's business, it ought to have known of the fact.  Surprisingly there is relatively little modern case law on this test; and the existing cases are difficult to rationalise.  The law in this area needs codifying, as the Law Commission suggests.

The prevailing view seems to be that the company will have constructive knowledge of facts that should have been revealed by an internal enquiry that was actually undertaken.  However, there is no duty to undertake a reasonable enquiry – thus the company is not deemed to know facts that would have been discovered had a reasonable enquiry been undertaken.  Yet, if a fact would have been discovered but for the wilful turning of a 'blind eye', the company will be deemed to know.

There is no clarity in the cases but, logically, the natural person who has made the internal enquiry (or caused it to be made), or who has wilfully turned a blind eye, must be a person who counts as the company for the purposes of the first limb of section 18(1) – ie principally a director or if different an employee whose responsibility it is to arrange the insurance.

The Law Commission suggests that the law should be changed, so that a company is under duty to make reasonable enquiries before placing insurance.   The duty would then be for the company to disclose those (material) facts (a) actually known about and (b) that would have been known about had such an enquiry been undertaken.

Innocent non-disclosure clauses

IND clauses are typical in many classes of business, including FI risks.  Essentially they limit the insurer's right to avoid to a fraudulent breach of the duty of disclosure.

Before there can be such a breach, the company must actually know of three things: (a) the fact not disclosed; (b) that it was material; and (c) that it had not been disclosed (eg by the broker).  This rules out constructive knowledge.

There is no reason in principle why the person who counts for the company for the purposes of (a) should not count for the purposes of (b) and (c).  If so, under an IND clause, the question should really be whether a director or an employee responsible for placing the insurance had actual knowledge of all three.

This is subject to what has become known as the Hampshire Land principle.  Under this principle, it seems a company will not have actual (or indeed constructive) knowledge of a fraud of which it is the intended target or victim (Stone v Rolls [2008] EWCA Civ 644 at para. 72, overturned by the House of Lords on other grounds).

If the fact not disclosed was, for example, that the company had faced claims pre-inception, and this was known by the employee who arranged the insurance (or by a director of the company), the Hampshire Land principle will not prevent that person's knowledge from being attributed to the company, as such knowledge is not of any fraud directed at the company.  It might be argued that the Hampshire Land principle prevents the attribution of the necessary further knowledge to the company (ie the relevant person's additional knowledge that the fact was material and had not been disclosed), because effectively that would be attributing to the company knowledge of a fraud (on insurers) of which the company would then be a victim (by reason of insurers being entitled to avoid the policy).  However, since the company would not be the intended victim of the fraud (the intended victim being insurers) and/or because it is insurers rather than the company that is being defrauded, we do not think this argument would be correct.  If it were correct, insurers could never avoid against a company for a fraudulent non-disclosure.