Brotherton & Anor. v Aseguradora Colseguros S.A. & Anor.

This appeal arises from a claim by reinsurers to have avoided reinsurances of two Columbian reinsureds for non-disclosure of reports in the Columbian media of allegations of misconduct and related investigations involving the original insured’s business and officers. More particularly, it concerns the question whether the materiality of such reports or investigations, or the validity of reinsurers’ purported avoidance for their non-disclosure, may depend upon the correctness of the allegations and whether there was actual misconduct. Whether there was misconduct should, on the reinsureds’ case, be ascertained by trial (presently fixed to commence on 16^th June 2003).

Moore-Bick J rejected the reinsureds’ case, and on 26^th February 2003 struck out those paragraphs in the defence by which they denied that there had been any actual misconduct. The reinsureds now appeal with his permission. The appeal has been expedited in view of the imminent trial date.

The reinsurers, Peter Malcolm Brotherton and other London insurers, are claimants in the proceedings. They seek declarations that they have validly avoided primary and excess layer reinsurances made in late November 1997 incepting 7^th November 1997 and extended in late November 1998 until 31^st January 1999. The reinsureds, Aseguradora Colseguros S.A. and La Previsora S.A., Compania de Seguros, are Columbian insurance companies. They issued bankers blanket bond and professional indemnity insurance policies to a Columbian state-owned bank, Caja de Credito Agraria Industrial y Minero. (“Caja Agraria”). The policies covered, among other matters, losses caused by dishonest or fraudulent acts of bank employees, subject to qualifications which it is not here necessary to examine. Caja Agraria’s president until his suspension in early November 1997 was S. Benjamin Medina.

Between 28^th January 1997 and late November 1997 seven news bulletins and fifteen newspaper articles (including six in El Tiempo said to be Columbia’s largest circulation newspaper) carried reports of allegations of misconduct and related investigations involving Caja Agraria’s business and S. Medina’s conduct thereof. These culminated with reference to his suspension for 90 days in early November 1997. In January and February 1998 there were further reports of such investigations, including disciplinary charges by then formulated against S. Medina and other bank officers.

Reinsurers plead (in paragraph 6 of the claim) that as a result of such reports both reinsureds, prior to entering into the contracts, knew or ought to have known that serious allegations of corruption and embezzlement of public funds had been made against S. Medina and other serious allegations against other officers; that these included allegations of irregular loans; that such loans were being investigated by the Contraloria General; that S. Medina’s involvement was being investigated by the Procurador General de la Nacion; that the Procurador General had ordered S. Medina’s suspension on or about 4^th November 1997; and that in early November 1997 he had been served with an arrest warrant by the police to comply with a court order to produce evidence in a criminal investigation by the Fiscalia General de la Nacion into corruption or embezzlement of public funds within Caja Agraria. As from February 1998 reinsurers also allege that the reinsureds knew or ought to have known that the Procurador General had brought disciplinary charges against S. Medina and other bank officers. Further, they allege that, at the time of placement (presumably also including the extensions) the reinsureds had no knowledge and no means of proving to reinsurers’ satisfaction that the allegations were untrue, or that claims would not arise out of the allegedly irregular loan transactions.

Reinsurers’ case is that the reports alone, and all the more the reports coupled with the fact of the investigations, were material to be disclosed, firstly as constituting circumstances which might give rise to claims under the reinsurances, and secondly as suggesting moral hazard. For completeness, I note that the lead reinsurance syndicate also asserts that positive misrepresentations were made to it in November 1997 as to the reasons for S. Medina’s non-attendance at a meeting in London and for his suspension.

The reinsureds in their defence admit the reports (although not the accuracy of the summary in paragraph 6 of the claim). They also admit the investigations, including the arrest and charges. It is not their case that the reports were as reports false. They make no admission that they knew or ought to have known of the reports or other matters which reinsurers allege should have been disclosed. But they admit that they could not at placement have demonstrated unequivocally that there was no proper basis for the investigations (a word which I will in this judgment use to include the arrest and later the charges), although they assert that they could have demonstrated their political motivation and that their mere fact was not a reliable basis for any views about S. Medina’s veracity or the regularity or otherwise of the loan transactions. They also assert that the lead reinsurance syndicate (through Mr Satterford) knew or ought to have known of the allegations (although they do not suggest that any such knowledge on its part would fall to be attributed to following reinsurers).

The reinsureds’ defence denies that such reports or other matters were material for disclosure. Their defence contains these further passages which were either struck out by Moore-Bick J or were drafted subsequently for inclusion only if his judgment is set aside:

“5(1) It is denied that there was any proper basis for the investigation, arrest of and charges against Snr Medina; on the contrary:

(i)

they were part of a political campaign by the opponents of the then government of President Ernesto Samper (which opponents included El Tiempo) to smear its supporters and friends, including Snr Medina, in order to discredit the government in the run up to the 1998 elections;

(ii)

63 of the 65 criminal investigations of Snr Medina have been concluded in his favour; the remaining two are still pending; one of them relates to the use of the aeroplane for private purposes (a matter disclosed to Reinsurers) and the other relates to an alleged infringement of public tendering regulations, and is unconnected with the matters reported in El Tiempo as pleaded and /or which are in any event immaterial;

(iii)

out of 17 investigations opened by Procuraduria, 16 have been closed and only 1 is live, which is currently being challenged before the Colombian courts; and the 1 Controlaria investigation into Snr Medina is still pending; ….

(1A) Subject to the appeal from the Order of Mr Justice Moore-Bick dated 26 February 2003, it is denied that there was any proper basis for the investigation, or formulation of disciplinary charges against Pablo Jose Ramirez, Patricia Cruz Ortiz, and Guillermo Leno Ocampo; on the contrary:

(i)

they were part of the political campaign referred to in sub-paragraph (1)(i) above to smear government supporters in order to discredit the government in the run-up to the 1998 elections;

(ii)

in the course of the Procuraduria’s disciplinary investigation, Dr Ocampo was acquitted, and the investigation of Dr Cruz and Dr Ramirez is being judicially reviewed by the Contentious Administrative Tribunal at Cundinamarca;

(iii)

the investigation before Contraloria General de la Nacion is still live.

…..

11A. …. the Defendants will rely upon:

(i)

the fact that the allegations ..… (subject to the appeal of the order of Mr Justice Moore-Bick dated 26th February 2003) were baseless …..” (underlining omitted).

In response to these passages (assuming them to stand), the reinsurers join issue with the allegations in paragraph 5(1) and (1A) that there was no proper basis for the investigations with respect to S. Medina and other officers, and put the reinsureds to proof of the matters pleaded in paragraph 5(1) of the defence.

The forthcoming trial will on any view have to address the following issues:

The critical issues on this appeal are whether (a) the materiality of the admitted reports (or the reports coupled with the investigations) or (b) the validity of reinsurers’ purported avoidance for their non-disclosure depends or may depend upon whether the allegations were correct and there was actual misconduct justifying the allegations and investigations. In his submissions Mr Millett QC for the reinsureds put these forward either as facets of one and the same point, or, if that is not right, as two independent points. His submissions were inspired by Colman J’s judgment in Strive Shipping Corp. v. Hellenic Mutual War Risks Association (The Grecia Express) [2002] EWHC 203; 2 Ll.R. 88, but he found it necessary to suggest what he described as a small gloss on the judge’s reasoning in that case. This gloss was, he submitted, supported by the reasoning of the House of Lords in Pan Atlantic Insurance Co. Ltd. v Pine Top Insurance Co. Ltd. [1995] 1 AC 501, by the rationale of the duty of disclosure and by the mutual duty of good faith in insurance contracts.

Mr Millett further submits that these critical issues can only be resolved at trial, and that Moore-Bick J’s decision was unprecedented in depriving an insured (or here reinsured) of the opportunity of questioning expert and factual witnesses about the materiality of matters allegedly withheld and their effect if they had been known. Whether that is correct depends, however, on the nature and scope in law of the duty of disclosure and of the right of avoidance for non-disclosure. Moore-Bick J was right to exclude a complicated factual issue from trial, if it could in law have no relevance.

Pan Atlantic is the most recent and authoritative exposition of the duty of disclosure. It is not suggested that there is any relevant difference between the marine reinsurance context and the present non-marine reinsurance context. Lord Mustill examined two authorities (the first relating to over-valuation of the vessel insured, the second related to under-valuation of cargo insured), in which the concept of the “risk” which is the subject of the duty to disclose was clarified. At pp. 534C-535B he said this with regard to Ionides v. Pender (1874) LR 9 QB 531:

“The facts of the case were simple. Commissions and other ancillary interests in goods were insured by the plaintiffs under a marine voyage policy at greatly excessive values. The vessel sank in circumstances which gave rise to suspicion. At the trial the jury was unable to reach a conclusion on whether the ship was cast away and whether the overvaluations were fraudulent, but did find that it was material to the underwriter to know of the overvaluation and that the fact of it was concealed. On these facts the sole issue was whether the non-disclosure was material. This was the culmination of a long-standing controversy about the nature of the "risk" to which the duty of disclosure was related. Was the duty of disclosure confined, as Duer had argued, at p. 390, to those facts which affected the intrinsic nature of the risks - i.e. those which affected the probability that the subject matter would be lost or damaged by a peril insured against? Or were Phillips and Arnould right to say that "risk" should be given a wider meaning so that the duty extended to anything which would probably influence the insurer's ultimate decision, including what later came to be called the "moral hazard?" The Court of Queen's Bench decided in favour of the latter view. This was certainly very important, but the decision was concerned with the kind of risk which was the subject of the duty to disclose, and not with either the standard imposed by the duty or the identity of the person by reference to whom the extent of the duty was to be ascertained; and there is nothing in either the texts or the decisions cited in argument which bore on these questions at all. The nearest that can be found in the entire report is the following passage from the judgment of the court, delivered by Blackburn J., at p. 539:

 "We agree that it would be too much to put on the assured the duty of disclosing everything which might influence the mind of an underwriter. Business could hardly be carried on if this was required. But the rule laid down in [Parsons, A Treatise on the Law of Marine Insurance and General Average, vol. I], p. 495, that all should be disclosed which would affect the judgment of a rational underwriter governing himself by the principles and calculations on which underwriters do in practice act, seems to us a sound one."

As I read this passage its purpose was the following. The law postulated by Duer was comparatively simple to apply. One simply looked at the intrinsic nature of the perils insured against and inquired whether they were enhanced by the non-disclosure. This was a question which was capable of reasonably objective ascertainment. The problem with the wider test, preferred by Parsons and in the event by the court itself, was that if the duty of disclosure was widened to include all the additional matters which might influence what an underwriter might think or do the duty would be at the same time impractically wide and impossible of ascertainment. The court therefore answered Duer's objection by emphasising that it was matters which might influence the mind not of "an underwriter" (Blackburn J.'s words) but of a hypothetical reasonable underwriter, whose standards of materiality would be at once bounded and possible to fix. This qualification was an essential part of the court's decision on the controversy which was the only issue before it …..” (emphasis added in the three italicised passages)

He addressed the second authority at p.538D-F:

“Before coming to the cases decided after 1906 I should mention Rivaz v. Gerussi Brothers & Co. (1880) 6 Q.B.D. 222, which was relied on to suggest that a test expressed in terms of the underwriter's assessment of the risk is contrary to authority. I disagree. As clearly appears from the argument for the defendants (reported at pp. 225-226) the case was concerned with the same controversy between the views of Duer and Parsons on whether the duty of disclosure extended beyond matters directly affecting the probability of a loss by perils insured against, as had previously been resolved in favour of the wider view in Ionides v. Pender, L.R. 9 Q.B. 531. That the word "risk" must be understood in the wider sense is now beyond dispute, but this has no bearing on the principle which I derive from the authoritative texts, and indeed from both the letter and the spirit of Carter v. Boehm, 3 Burr. 1905, that it is the relevance to the underwriter's intellectual process when assessing the risk which determines the scope of disclosure.” (emphasis again added in the italicised passage)

These authorities are reflected in ss.18 and 20(1) and (2) of the Marine Insurance Act 1906, codifying the previous law:

“18.

Disclosure by assured

(1)

Subject to the provisions of this section, the assured must disclose to the insurer, before the contract is concluded, 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. If the assured fails to make such disclosure, the insurer may avoid the contract.

(2)

Every circumstance is material which would influence the judgment of a prudent insurer in fixing the premium, or determining whether he will take the risk.

(3)

In the absence of inquiry the following circumstances need not be disclosed, namely

(a)

Any circumstance which diminishes the risk;

(b)

Any circumstance which is known or presumed to be known to the insurer. The insurer is presumed to know matters of common notoriety or knowledge, and matters which an insurer in the ordinary course of his business, as such, ought to know;

(c)

Any circumstances as to which information is waived by the insurer;

(d)

Any circumstance which it is superfluous to disclose by reason of any express or implied warranty.

(4)

Whether any particular circumstance, which is not disclosed, be material or not is, in each case, a question of fact

(5)

The term 'circumstance' includes any communication made to, or information received by, the assured.

….

20.

Representations pending negotiation of contract

(1)

Every material representation made by the assured or his agent to the insurer during the negotiations for the contract, and before the contract is concluded, must be true. If it be untrue the insurer may avoid the contract.

(2)

A representation is material which would influence the judgment of a prudent insurer in fixing the premium, or determining whether he will take the risk.”

In Pan Atlantic itself the House of Lords affirmed that the right to avoid depends not simply upon the failure to disclose material information, but in addition upon whether knowledge of the information withheld would have induced the actual underwriter to act differently, either by refusing to write the risk at all or by writing it only on different terms. It was in that latter sense, referring to the requirement of inducement, that Lord Mustill spoke of the need for an insurer to show “harm” at p.549C-D.

Mr Millett accepts that the materiality of a given circumstance has to be tested at the time of the placing of the risk and by reference to the impact that it would then have on the mind of a prudent underwriter. The question whether knowledge of it would have induced the actual underwriter to act differently must be judged at the same time. But Mr Millett argues that the only circumstances requiring disclosure are those which actually exist, and that allegations or investigations with respect to possible misconduct do not have to be disclosed, if there was in fact no misconduct, even if there was at the time of placement no way of knowing or showing this.

There is long-standing authority that a person proposing for insurance may be obliged to disclose intelligence (as distinct from “loose” or “idle” rumours) about which he knows prior to placing: see e.g. De Costa v. Scandret (1823) 2 EQ. Ca. Ab 636 (Lord Macclesfield LC); Seaman v. Fonereau (1743) 2 Stra. 1183 (Lee CJ); Lynch v. Hamilton (1810) 3 Taunt. 15 (on a rule nisi to set aside a jury verdict before the Court of Common Pleas); Lynch v. Dunsford (1811) 14 East. 494 (proceedings by writ in error from the Common Pleas to the King’s Bench); Durrell v. Bederley (1816) Holt 283; and Morison v. The Universal Marine Insurance Company (1872-3) LR 8 Exch. 40 and 197. S.18(5) of the Marine Insurance Act 1906 further provides:

“The term “circumstance” includes any communication made to, or information received by, the assured”.

In his notes to s.18 in his Digest of the Law relating to Marine Insurance (1901) Sir Mackenzie Chalmers, as he became, stated:

“An apparently well-founded rumour, though it turns out to be incorrect, must be disclosed (Arnould, Ed. 6, p.574)”.

The relevant text in Arnould reflects the effect of the authorities mentioned earlier in this paragraph. It was unsettled in 1906 whether the evidence of expert underwriters would be admitted on the issue of materiality (although Blackburn J’s judgment in Ionides v. Pender pointed in that direction). Now it is well-established that such evidence is admissible. But it will on these authorities be directed to whatever information was available at the time of placing.

However, only in the litigation involving Lynch was the intelligence (credible as it might have seemed at the time of the placement) incorrect in fact. So it is the decisions involving Lynch that are most significant. It is worth quoting the two judgments on the writ in error in the King’s Bench as reported in full:

“Lord Ellenborough, C.J. The question is, whether the assured’s agent was bound to communicate to the underwriters a material fact within his own knowledge, as coupled with the report made relating to the supposed risk they were about to insure, which report afterwards turned out not to be true. Now the duty of the assured or his agent in making such communications of material circumstances within their knowledge must attach at the time of effecting the insurance, and cannot depend upon the subsequent event. There is no case perhaps exactly like this in species, but others have been decided involving the same principles, that the assured is bound to communicate to the underwriters every thing material to the risk within his knowledge at the time. Here, coupling the peculiar knowledge which the agent had of the name of the ship, on board of which the goods were loaded, with the information contained in the paper stuck up at Lloyd’s, it cannot be said that the fact was not material to be communicated to the underwriters. With the knowledge of such a fact kept back from them, can they be said to have contracted upon equal terms? The intelligence announced in the paper at Lloyd’s was nothing to the underwriters, unless they had the means of applying it to the particular ship or goods in which the assured were interested. If the underwriters had had the knowledge possessed by the assured, it might have been a question with them whether they would have insured at all; or if they did, whether they would not have required an enhanced premium.

Bayley, J. The assured’s agent is blameable, not for not communicating the rumour, but for not communicating to the underwriters a fact material with reference to that rumour, which fact was within his knowledge, so as to enable them to apply it to the rumour, and exercise their judgment accordingly. As to the assured taking the chance of the event upon himself; he did not tell the underwriters of the fact within his knowledge, and that he was willing to take that chance upon himself; but he took the chance of their finding out his knowledge of the fact, if it afterwards turned out to be true.”

The legal test of materiality established by authority and by statute is on the face of it clear. A matter is material if it would influence the mind of a rational underwriter governing himself by the principles and practices on which underwriters do in practice act - cf per Blackburn J in Ionides v. Pender - or would (using the more modern terminology) influence the judgment of a prudent insurer in fixing the premium, or determining whether to take the risk (including, of course, whether to take it on the same terms) - cf ss. 18(2) and 20(2) of the 1906 Act and per Lord Mustill in Pan Atlantic. Materiality falls to be considered as at the date of the placing, by reference to the circumstances (which may include no more than intelligence) within an insured’s knowledge at that date. Whether a material non-disclosure induced the actual underwriter to act to his prejudice depends likewise upon whether the circumstances withheld would, if known, have caused him to act differently, either by not writing the insurance at all or by only writing it on different terms. To ask an expert underwriter or a court to consider either materiality or inducement in the light of circumstances not within either the insured or the insurers’ knowledge at the time of placement is inadmissible. An insured can only disclose what lies within his knowledge, and an underwriter cannot complain about non-disclosure of other matters. But the insured must at least disclose what is within his knowledge, provided that it is material in the above sense.

Mr Millett submitted without opposition on the point that Lynch v. Dunsford does not bind us (Footnote: ¹), and that the decisions in both the Common Pleas and King’s Bench are wrong. He submits (contrary to Lord Ellenborough’s statement) that the true rationale of the duty of disclosure is not to put the parties in an equal bargaining position, but to give the underwriter a “fair opportunity” to understand “the actual risk”. In judging the actual risk, the court should not blind itself to “objectively ascertainable facts”. There should be a trial to ascertain those facts, i.e. whether there was any actual misconduct justifying the allegations and investigations regarding S. Medina and others. S. Medina should be allowed to give the evidence which the reinsureds propose to call from him in that respect. One virtue of Pan Atlantic was, Mr Millett submits, to focus attention on the need for actual prejudice to the actual underwriters. If there was no actual misconduct, then the reinsurers will have written the reinsurances on the same basis as that on which they (in ignorance of the allegations and investigations) thought that they were underwriting them. Reinsurers will have suffered no “harm”, and, since the law regarding disclosure does not have any disciplinary element, the matters not disclosed cannot be regarded as material and cannot justify reinsurers’ purported avoidance.

Mr Millett also draws attention to the line of authority on moral hazard, particularly March Cabaret Club & Casino Ltd. v. The London Assurance [1975] 1 Ll.R. 169; Reynolds v. Phoenix Assurance Co. Ltd. [1978] 2 Ll.R. 440; Inversiones Manria SA v. Sphere Drake Insurance Co. plc (The Dora) [1989] 1 Ll.R. 69; and The Grecia Express. The subject also arose in Insurance Company of the Channel Islands v. The Royal Hotel Ltd. [1998] LRIR 151, where I introduced it with these words, referring to the passages quoted above from Lord Mustill:

“It is important to realize what is embraced by “risk”. It is not simply the peril or possibility of loss or damage occurring within the scope of the policy. It embraces other matters which would, if known, be likely to influence a prudent underwriter’s decision. It includes what is known as “moral hazard”, which may merely increase the likelihood of it being made to appear (falsely) that loss or damage has occurred falling within the scope of the policy: see per Lord Mustill in Pan Atlantic at pp 534D-E and 538E-F.”

The present case is only in part concerned with moral hazard. The reports of improper loans are also relied on as directly material to the perils insured (as was in fact the position in each of the cases cited in paragraph 16 above). But, since Mr Millett submits that materiality, at least in cases of moral hazard, must depend on the known existence of actual moral hazard, rather than the possession of information suggesting the possibility of moral hazard, it is necessary to examine the position. As a matter of principle, and in the light of the authorities already examined, it is difficult to see any reason why, if the evidence satisfies the court that a prudent underwriter would have regarded information suggesting the possibility of moral hazard as material in the sense identified by Blackburn J and Lord Mustill, that should not suffice. In my view that is the basic legal position.

However, the authorities grapple understandably with some hard cases. Firstly, what if at the time of placing the insured himself is under investigation for, or has been charged with, an offence which he knows that he did not commit, and of which he is subsequent to the placing indeed acquitted? Forbes J in Reynolds and Fisher J in Gate v. Sun Alliance Insurance Ltd. [1995] LRLR 385 (High Court of New Zealand) thought that this could not be material. May J in March Cabaret and Phillips J as he was in The Dora held, after hearing underwriting evidence, that it could be, on the basis, as Phillips J put it, that:

“When accepting a risk underwriters are properly influenced not merely by facts which, with hindsight, can be shown to have actually affected the risk but with facts that raise doubts about the risk”.

I add however that, in this situation, the issues of both materiality and inducement would in all likelihood fall to be judged on the basis that, if there had been disclosure, it would have embraced all aspects of the insured’s knowledge, including his own statement of his innocence and such independent evidence as he had to support that by the time of placing. This might itself throw a different light on the answer to one or both of the issues of materiality and inducement. That would of course be a matter of fact and evidence.

Secondly, May J in March Cabaret and Colman J in The Grecia Express addressed the different position of an insured who prior to a placing has both been charged and acquitted of an offence, but was in reality guilty of it. In that situation, can the fact of guilt be material for disclosure? May J in March Cabaret, with Colman J agreeing in The Grecia Express, held that the fact of guilt could remain material and disclosable. But, thirdly, Colman J disagreed with May J’s further suggestion that non-disclosure of a mere conviction could justify avoidance, even if the assured maintained that it was wrong and would (if allowed) be able to prove this. Colman J thought that this was hardly consistent with the previous proposition that an insurer could prove that an acquittal prior to placement was wrong. The logic, he thought, was that it should “be open to the insured to disprove his guilt and thereby to disentitle the insurers to avoid the policy” (p.130(2)). I do not consider that the suggested inconsistency exists. Since what is material depends upon what would influence the judgment of a prudent insurer at the time of the placing, both the (known) fact of guilt, in the case of an acquittal, and the (known) fact of a conviction, in a case where the insured himself knows that he is innocent, may be capable of being material to a prudent insurer. Once again, however, in the latter case, the consequences may be mitigated by the consideration that an insured can and will presumably disclose not merely the conviction, but all matters supporting his statement that he was wrongly convicted. I add that caution is required with respect to Colman J’s suggestion (at p.130(2)) that the trial in March Cabaret proceeded on the basis that it would have been open to the insured to disentitle the insurers from avoiding, by disproving his guilt at trial. The insured in fact withdrew his plea of innocence prior to the trial, so that it was never relevant to consider either the materiality of the conviction alone, or the admissibility of the plea: see [1975] 1 Ll.R., 169, 174(1).

Reverting to the first case, considered in paragraph 22 above, Colman J in The Grecia Express agreed with Phillips J in The Dora. Colman J. pointed out that the insurer might well have preferred either not to trust the word of the insured or to conduct his own investigation: see pp.131(2) and 133(2). He also addressed a fourth, more general case of

“circumstances involving [an insured] or his business or property which reasonably suggest that the magnitude of the proposed risk may be greater than what it would have been without those circumstances” (p.131(2)).

Colman J accepted that, if an insured knows of such circumstances, then

“the known facts do not cease to be material because it may ultimately be demonstrated that the suggested facts do not exist” (p.132(1)).

Mr Millett submits that Colman J erred in accepting this. I disagree. Colman J was in this respect clearly right. The sound philosophical basis of the duty of disclosure in an insurance context is that a true and fair agreement for the transfer of risk on an appropriate basis depends on equality of information. This was stated by Lord Ellenborough and was also encapsulated, partly in Latin, by Lord Mansfield in that locus classicus of insurance law from the Age of the Enlightenment, Carter v. Boehm (1766) 3 Burr 1905, 1910:

“Good faith forbids either party by concealing what he privately knows, to draw the other into a bargain, from his ignorance of that fact, and his believing the contrary.

…..

There are many matters, as to which the insured may be innocently silent – he need not mention what the under-writer knows – Scientia utrinque par pares contrahentes facit. [i.e. Equal knowledge on each side makes equal contracting parties]”

However, Mr Millett relies strongly on a supplementary passage in which Colman J stated that

“…. for reasons which were not argued or considered in either [Reynolds or The Dora], I do not consider that failure to disclose allegations which on the evidence before the Court are proved to have been false entitles underwriters to avoid the policy”.

It is appropriate to set out Colman J’s reasoning in full:

“ ….. one should not lose sight of the fact that were the evidence before the court, upon which underwriters rely to avoid the policy, to establish that, although the known facts were not disclosed, the suggested facts did not in truth exist, underwriters would be seeking to avoid liability in respect of a risk which, had they been in possession of the true facts, as distinct from the allegations of suggested facts, they would have written without hesitation. In so doing they would, in effect, be utilising loss of the opportunity of forming an unfounded suspicion of non-existent facts in order to avoid paying a loss under a policy which, had the truth been made known to them when they wrote the risk, they would not have hesitated to underwrite. To persist in such a course in the face of evidence before the court that the suggested facts never existed would, in my judgment, be quite contrary to their duty of the utmost good faith. Such a course would be so starkly unjust that I would hold that in such a case it would be unconscionable for the court to permit the insurers to avoid the policy on the grounds of non-disclosure. Having regard to the equitable origin of the jurisdiction to avoid a policy for breach by the assured of the duty of the utmost good faith, the court should not be inhibited from giving effect by appropriate orders to the insurers’ countervailing duty of the utmost good faith to the assured. The breach of that duty by the insurers would be so unconscionable as to disentitle the insurers from invoking the equitable jurisdiction of the court to avoid the contract on the grounds of non-disclosure by the assured.

The procedural and evidential consequences which flow from this conclusion are, in my judgment, as follows:

(1)

In the field of moral hazard, a failure by the assured to disclose an existing allegation against him of dishonesty or relevant criminal conduct or a criminal charge will normally be non-disclosure of a material fact which prima facie entitles the insurer to avoid the policy.

(2)

If, in proceedings in which the insurer seeks to avoid the policy for such non-disclosure, the assured proves that the allegation or charge was unfounded and that there has been no dishonesty or criminal conduct on his part, the insurers will not normally be entitled to avoid the policy. For example, where the assured has been charged with a criminal offence and subsequently acquitted at a trial, he can deploy his acquittal as some evidence, but not conclusive evidence, of his innocence. Similarly, if he has been charged but not yet convicted, he can prove his innocence in order to displace the entitlement of the insurers to avoid for his failure to disclose the charge against him.

(3)

If I am wrong in concluding that an assured is under no duty to disclose facts merely because they are objectively suspicious as to his own wrong-doing when he knows that the suggested facts do not exist, it must by parity of reasoning be open to the assured to displace the underwriters’ entitlement to avoid for non-disclosure of circumstances because they are objectively suspicious by proving that the suspicion was misplaced and that the facts of the existence of which there was suspicion never in truth existed.

(4)

If the facts objectively raise suspicions going to the magnitude of the risk, the assured is under a duty to disclose them but if at the trial he establishes that there was in truth no basis for those suspicions it is not open to the insurers to invoke the court’s equitable jurisdiction to avoid the policy.

I therefore agree with the conclusion as to the duty to disclose allegations of misconduct going to moral hazard arrived at by Phillips J. in The Dora, supra, and I am unable to accept the analysis of Forbes J. in Reynolds, supra. However, for reasons which were not argued or considered in either case, I do not consider that failure to disclose allegations which on the evidence before the court are proved to have been false entitles the underwriters to avoid the policy.”

The reasoning is addressed to cases of moral hazard. But, if correct, it must also apply in a case like Lynch v. Hamilton and Lynch v. Dunsford. Indeed, those decisions would on that basis have had to be decided in the opposite sense, since there was no doubt there about the falseness of the intelligence. Several strands are discernable in Colman J’s reasons. One strand is that the right to avoid is conditional upon the consistency of any such avoidance with good faith or conscience. A second (implicit in the reasoning) is that an insured is, if necessary, entitled to litigate the issue of the truth or falsity of known but undisclosed intelligence, in order to argue that, if it is shown to be incorrect, the insurer would be acting in bad faith or unconscionably in avoiding. The third strand is that the court has a role in permitting (or refusing to permit) insurers to avoid a policy for non-disclosure.

The third strand is relevant in a case like The Grecia Express or the present, where the correctness or falsity of intelligence cannot be determined short of a full trial. It is in my judgment wrong in general principle. It is clear that rescission in the general law of contract is by act of the innocent party operating independently of the court: see Abram Steamship Co. v. Westville Shipping Co. [1923] AC 773 (rescission of a shipbuilding contract for misrepresentation) and Horsler v. Zorro [1975] 1 Ch. 302 (rescission of a contract for sale of land). I see no basis for saying that avoidance of an insurance contract for non-disclosure or misrepresentation is any different. Ss.18(1) and 20(1) of the 1906 Act also indicate that it is not. Moore-Bick J was right to hold accordingly in Drake Insurance plc v. Provident Insurance Co. [2003] EWHC 109 (Comm). Mr Millett accepts this.

Mr Millett nonetheless submits that the first and second strands of Colman J’s reasoning can be supported independently. The court should take into account (and if necessary determine by trial) whether intelligence which the insured failed to disclose was correct. It would, Mr Millett submits, be contrary to good faith for an insurer to give, or to persist in relying on, a notice of avoidance which could at trial be shown to have been based on the materiality of intelligence, which with hindsight could be seen to have been incorrect. English insurance law regarding disclosure could and should be mitigated in this way. He showed us some academic writing welcoming Colman J’s general approach: cf Insurance and moral hazard, Ray Hodgin, NLJ 18/4/03 and also Marine Insurance fraud, good faith and avoiding loss, Prof. R.M. Merkin, (2002) 14 Insurance Law Monthly, although the welcome given in this commentary appears to have been addressed to a case where, by the time of the purported avoidance, apparently material facts had proved to be untrue (cf p.8), which does not correspond with the facts either in The Grecia Express or in the present case. It is true that English insurance law has been criticised in a number of respects (and in the area of private insurance, mitigated by convention and the activity of the Insurance Ombudsman). Lord Mustill in Pan Atlantic identified and considered certain main criticisms at pp.528-530. But they did not in that case, and do not seem to me in this case, to bear on the solution of the present appeal. Courts, which are the ultimate decision-makers on issues with respect to both materiality and inducement, will be able to take a realistic and even a robust view about what constitutes “intelligence” which is material for disclosure as distinct from loose or idle rumours which are immaterial, and as to whether a particular underwriter would have been induced to act differently, had he known of an undisclosed circumstance. But, as I have shown, Pan Atlantic identifies a general test of materiality which is on the face of it inconsistent with Mr Millett’s case. Further, I cannot see that the decision in Pan Atlantic that avoidance depends on inducement as well as materiality lends support to a conclusion that avoidance for non-disclosure of otherwise material information should depend upon the correctness of such information, to be ascertained if in issue by trial.

In my judgment, neither principle nor sound policy supports such a conclusion. The assumed starting point is that an insurance has actually been induced by the non-disclosure of matters within an insured’s knowledge which would have influenced the judgment of a prudent insurer. Had the insured made the disclosure which Colman J himself accepts should have been made, the insurance would either not have been written at all, or it would have been written on different terms (including as to premium), and the insured could not done anything about this in either case by later showing the incorrectness of any material intelligence which induced the insurer to refuse the insurance or to insist on more stringent terms. Moreover, had due disclosure been made, there would have been no call for any subsequent investigation or litigation about the correctness or otherwise of the intelligence. The insured has failed to make the disclosure he should have made. That carries with it the risk that underwriters would never learn of the intelligence at all, and never realise that there was anything to investigate. That sound point was, I strongly suspect, in Bayley J’s mind in 1811 when he said:

“As to the assured taking the chance of the event upon himself; he did not tell the underwriters of the fact within his knowledge, and that he was willing to take that chance upon himself; but he took the chance of their finding out his knowledge of the fact, if it afterwards turned out to be true.”

Further, if no disclosure was made, and underwriters did discover the intelligence later, then on Mr Millett’s case the matter does not end there. The position is not restored to that which would have applied, had there been timely, pre-contractual disclosure. On the contrary, underwriters are bound to be prepared to investigate and litigate what might in some cases, and certainly would in the present case, be matters of the most difficult and sensitive nature, both from an evidential and from a substantive viewpoint. In some cases it is possible (as Colman J suggested in The Grecia Express) that underwriters might have conducted some form of pre-placement investigation of intelligence, if it had been disclosed to them. But it is implausible to think that this would compare with the investigation which would occur prior to and at trial of an issue finally to determine the correctness of such intelligence. The present case illustrates this implausibility. A very substantial public enquiry or trial would probably be necessary to form any definite view as to the correctness of allegations like those reported in the Columbian media and the subject of the Columbian investigations.

In summary, it would be an unsound step to introduce into English law a principle of law which would enable an insured either not to disclose intelligence which a prudent insurer would regard as material or subsequently to resist avoidance by insisting on a trial, in circumstances where:

I add that Mr Millett was in some difficulty in answering the question who would in case (ii) have to meet the costs of any such investigation or litigation. Ultimately, he suggested that they would probably just be in the cause, so that insurers would bear them however much they could show that they would never have written the insurance at all in the first place, or never have had to make such investigations, if proper disclosure had been made pre-placement, and however reasonable their investigations and their conduct of any litigation. That underlines the problems which the case advanced on the reinsureds’ behalf by Mr Millett faces.

If, as I consider, both the second and the third strands of the reasoning in The Grecia Express are unsound, that is the end of this appeal. Unless the reinsureds are entitled to a trial to determine whether there was any actual misconduct justifying the allegations and investigations relating to the original insured and its officers, the only relevant issues for trial are those identified in paragraph 9 above. Absent such a trial and determination, there can be no conceivable basis for complaining about reinsurers’ avoidance for non-disclosure of the reports (or the reports and investigations), assuming that such matters are shown at trial to have been material and to have induced reinsurers in the sense identified in Pan Atlantic (cf paragraphs 12 and 18 above).

However, I add some further words on the first strand. Firstly, rescission under English law is not generally subject to any requirement of good faith or conscionability. It is unnecessary to address the academic chestnut, delicately handled by Lord Mustill in Pan Atlantic at p.544, of the juristic origin of the duty of disclosure in insurance law, in particular whether or to what extent the origin is equitable and what role the common law played: cf Carter v. Boehm. The mere fact that a right to rescind has an equitable origin does not mean that its exercise is only possible if that is consistent with good faith or with a court’s view of what is “conscionable”. Secondly, recent authority has in any event tended to limit the scope of any post-contractual duty of good faith to circumstances of repudiatory breach or fraudulent intent: cf Manifest Shipping Co. Ltd. v. Uni-Polaris Shipping Co. Ltd. (The Star Sea) [2001] UKHL 1; [2003] 1 AC 469; Merc-Scandia XXXXII (K/S) v. Lloyd’s Underwriters (The Mercandian Continent) [2001] EWCA Civ 1275; 2 Ll.R. 563; and Agapitos v. Agnew (The Aegeon) [2002] EWCA Civ 247; 2 Ll.R. 42. Even if there were any scope in any circumstances for development of the first strand in English insurance law (and for qualifying, in effect, a clear principle recognised as long ago as 1811 in Lynch v. Dunsford), as to which I need express no view at all, it could not be in a case like the present where reinsurers did not at the time of avoidance accept or know for certain of the incorrectness of the intelligence constituting the basis of their avoidance.

For the reasons I have given, I consider that Moore-Bick J’s judgment, with which I agree in its entirety, was correct, and that this appeal fails.

I entirely agree with the judgment delivered by my Lord. With some diffidence, I add only a very few words of my own.

The reassured argued that an allegation or rumour relating to a matter that, if true, would or might affect the decision whether to write the policy was only material, alternatively was only a “circumstance”, if in the event it proved to be true. That was why evidence should be admitted in this case to seek to demonstrate that the allegations relating to Senor Medina were unfounded. The basis of this argument is that if the insurer is not told about an allegation that, after investigation, proves to be false, he has not lost anything or suffered any damage: because the risk that he wrote was in fact the risk as he believed it to be.

I am of course influenced by the fact that three judges of the Commercial Court (Phillips J, as he then was, in The Dora; Colman J in The Grecia Express; and Moore-Bick J in the present case); as well, of course, as my Lord; do not agree with that proposition. However, rather than simply repeat those holdings, authoritative though they are, I would venture to look at the matter as one of principle.

First, it is agreed on all sides that materiality is to be judged at the time of accepting the risk. Materiality depends on the effect of the circumstance on the decision of a prudent underwriter. I do not see how that effect can be sensibly judged in any particular case except in the context of the circumstances that existed at the time of acceptance of the risk. It seems self-evident that a prudent underwriter will or at least may be affected by rumour or allegations about matters that are material to the risk. I venture to repeat my Lord’s citation, in his § 32, from Phillips J in The Dora [1989] 1 Ll LR 69 at p 93:

“When accepting a risk underwriters are properly influenced not merely by facts which, with hindsight, can be shown to have actually affected the risk but with facts that raise doubts as to the risk”

The reaction of the underwriter has therefore to be judged at the time of placement. That is not unfair to the assured, who in relation to a matter likely to be of legitimate interest to the underwriter can disclose to him all the circumstances, and seek to direct the underwriter towards an informed and favourable decision.

Second, it is necessary to examine further the proposition that if the rumour proves to be false the underwriter has lost nothing in writing the policy, because he has written the risk that he thought he was writing. He has however lost the opportunity to take an informed decision at the time of placement. As Lord Mustill said in Pan Atlantic [1995] 1 AC at p 528C:

“the vice of misrepresentation and non-disclosure is not that after the event the underwriter has suffered from having taken on a parcel of risks one of which led to a loss, but that a breach of the duty of good faith has led the underwriter to approach the proposal on a false basis”.

Here again, it is the position of the underwriter at the date of the proposal that is crucial.

Third, great difficulty is caused when the underwriter discovers the undisclosed allegations, and wishes to rescind. On the appellant’s argument, he cannot safely do so unless he satisfies himself that the allegations are accurate or, perhaps, not plainly untrue (the standard to be applied in determining the effect of matters referred to in the allegations upon the underwriter’s continuation on risk, and upon whom lay the burden of establishing that standard at trial, was not plainly set out before us). That would seem to be a complete departure from the important requirement of certainty in insurance dealings. In each case the underwriter, even though he complains of non-disclosure of a matter that on the facts known at the date of placement was material, has to await a trial in order to determine whether he is on risk. That, unsatisfactory, position is the direct outcome of moving the decision on materiality from the date of placement to a later occasion.

Fourth, it seems to me that the question is, if not decided, then strongly influenced, by authority at a higher level than this court. In Pan Atlantic Lord Mustill at pp 534-535 discussed, in the context of Ionides v Pender (1874) LR 9 QB 531, the previous debate between the view that the duty of disclosure was confined to facts which affected the intrinsic nature of the risk, and the view that anything should be disclosed which would probably influence the insurer’s decision. Lord Mustill observed that the latter view, subject to the gloss put on it by Lord Blackburn in Ionides v Pender, was the standard understanding of the duty. I cannot see how Lord Mustill could have dealt with the matter in that way, simply by focussing on the decision of the underwriter at the time of the placement, if he had had any idea that the test was as urged by the appellants in this case, assessing the status of a non-disclosure on the basis of facts that, at the time of placement, were not known either to the insurer or to the insured. I would not regard Lord Mustill’s approach as determinative of the question before us as a matter of authority, but it is strongly persuasive that the appellants’ approach is based on incorrect assumptions about the basic nature of the process of insurance.

I would, therefore, reject the appellants’ contentions as to the nature of material circumstances.

The judge was right on this point, for the reasons that he gave in §§ 31-32 of his judgment in Drake Insurance v Provident Insurance. Like him, I am unable to accept the contrary view of Colman J in The Grecia Express.

The first and obvious basis on which this argument fails is that rescission is an act of the party, effective as soon as made, and regarded by the courts as so effective provided that the appropriate circumstances for rescission existed at that time. That is established by the authorities cited by Moore-Bick J, namely Lord Atkinson in Abram Steamship and Megarry J in Horsler v Zorro.

There is no authority for the appellant’s argument that the court retains some power of equitable intervention to control the exercise of the right to rescind. I deal briefly with the cases that were said to support that view. When this court said in Banque Keyser [1990] 1 QB at p 779F that the right to rescind

“depends not on any implied term of the contract but arises by reason of the jurisdiction originally exercised by the courts of equity to prevent imposition”

it was saying, in the context of an attempt to promote a claim for contractual damages in a case of non-disclosure, that the right to rescind was based not on an undertaking by the parties, but on the recognition by the court of the effectiveness of an act of rescission. No claim was made for the further retention by the court of a power of intervention. When Lord Mustill said in Pan Atlantic [1995] 1 AC at p 549C that it would be unjust

“to enable an underwriter to escape liability when he has suffered no harm”

a passage much relied on by the appellants, he was plainly addressing the altogether different question of whether a representation could be material for the purposes of section 20 of the Marine Insurance Act when it had not induced the making of the contract. He was not addressing at all the question of whether, if a material representation had induced the making of the contract, the power of the representee to rescind could be controlled, the act of rescission could be reversed, by the court on the ground that, although the representor could not show the representation to be true at the time of rescission, he can do so by the time of the trial.

The appellant however sought to avoid this general jurisprudence of the law of rescission by urging that insurance was a special case, being a contract uberrimae fidei in both directions. It showed mala fides on the part of the insurer to stand on a rescission when he now knew that the rumour on the basis of which he rescinded had been untrue. The court should find some means of depriving the insurer of the fruits of that act on his part.

This argument does not circumvent the difficulties arising from the nature of rescission that have already been set out. Nor does it have merit in general terms. The duty of good faith, or of utmost good faith, applies in the formation of the contract. It is simply inept to extend it to the enforcement of the contract in litigation. Nor would the insurer be acting wrongly in the circumstances posited. Once it is accepted that he is entitled to complain of the failure to disclose relevant allegation or rumour; and entitled to rescind the contract once he learns of the undisclosed allegation or rumour; then in resisting claims on the basis that the contract no longer exists he is doing no more than standing on his rights both in law and in equity. The unconscionability argument is in truth no more than a way of seeking to avoid by a side-wind the effects in law of the insured’s non-disclosure.

I am not sorry to reach these conclusions. Although I agree with Mr Millett’s submission, in a spirit of fiat justitia ruat coelum, that if the insured is truly entitled to re-open the issues of allegation and rumour then the burden on courts and parties of his doing so must not stand in his way; nonetheless that burden is striking. In any allegation case, it would on the appellants’ argument be open to the insured to demand a trial of the truth of what might be complex issues, affecting many persons other than the insured. The instant case is a striking but, I suspect, not wholly untypical example, where it is proposed that the English civil courts should investigate allegations of acts alleged to have been criminal by the law of Colombia, in the process forming a view of the implications for the truth of those allegations of fallings-out within the Colombian political class. Mr Millett said that courts often had to conduct such investigations, for instance where fraud is alleged against the insured. An example close to hand was the Royal Hotel case, that occupied 22 days of valuable court time before my Lord in 1996. I would only comment, first, that such exercises are in some cases unavoidable is no reason for welcoming them in other cases; and, second, a great deal of the investigation in allegation cases is likely to be conducted at second hand, with and in relation to persons who are not party to the contract, and thus with much greater difficulty of organisation and much less discipline imposed by the test of relevance.

As I have said, if our law offers the insured that option as a means of avoiding the consequences of his non-disclosure, then so be it. My view, like that of my Lord, is that the law does not offer him that option.

I agree, and there is nothing that I could usefully add.