Friends Provident Life & Pensions Ltd v Sirius International Insurance

This is an appeal from the determination by Moore-Bick J by judgment dated 22^nd July 2004 of preliminary issues directed to be tried in proceedings brought by the respondents, Friends Provident Life and Pensions Ltd., against the appellant insurers. The respondents are successors to London & Manchester Assurance Co. Ltd. (“LMA”), to whom the appellants provided excess layer professional indemnity insurance cover on a claims made basis for the period 1^st February 1993 to 31^st January 1994. The preliminary issues relate to the incorporation, nature and effect of notification provisions in the insurance, the respondents’ compliance with them and the consequences of any non-compliance.

The background giving rise to the issues, the relevant terms of the policies and the preliminary issues are as set out in the following paragraphs taken from the judge’s judgment:

“2.

LMA's professional indemnity insurance for the 1993-94 year took the form of a primary layer providing cover in respect of losses of up to £1 million any one claim and in the aggregate (subject to various deductibles) and an excess layer providing cover of £4 million in excess of £1 million any one claim and in the aggregate. The primary layer was underwritten by Syndicate No. 657 at Lloyd's through an agent, Resource Underwriting Ltd. The first excess layer was underwritten in part by a group of Lloyd's Syndicates, including Syndicate 657, and in part by the defendants, all of whom were members of the London companies' market. Both layers were placed by the brokers Bowring Marsh & McLennan Ltd ("Bowrings") and were written on a "claims made" basis, that is, the policies were expressed to provide an indemnity against losses arising from claims made against the insured during the period of the policy.

3.

The business of LMA included giving financial advice to individuals in relation to personal pension plans. On 28^th January 1994 in the context of negotiations for the renewal of cover for the year beginning 1^st February 1994 Mr. Harvey, the Legal Services Manager of LMA, wrote to Lloyd's underwriters at the address of Bowrings in Exeter in the following terms:

"I confirm that after due enquiry I know of no circumstances likely to give rise to a claim under the Group's Professional Indemnity Policy save as follows:-

1.

Matters which are currently under investigation but are not likely to exceed the deductible under the policy.

2.

Pensions Transfers and Opt Outs which are a matter of public record and relate to all pensions providers. Detailed investigation will be conducted into pensions related transactions in accordance with any SIB/LAUTRO guidelines and notification of any potential claims given to underwriters in the usual way."

4.

The reference to "pensions transfers and opt-outs" was a general reference to LMA's involvement in giving financial advice to employees who were considering whether to transfer from, (or, in the case of new employees, opt out of), private pension schemes run by their employers in favour of personal pension plans available in the market. By the latter part of 1993 the regulatory bodies had expressed concern that the advice given to many clients by their financial advisers was inadequate and had led to what was later to become known as "pension mis-selling". They had already indicated their intention to conduct an investigation, but its precise nature and scope had yet to be determined. In the event, as a result of those investigations LMA was required to pay sums totalling over £9 million to various clients by way of compensation.

5.

Clause 2 of the General Conditions forming part of the primary layer policy obliged the insured to notify the underwriters as soon as possible of any circumstances that might give rise to a claim. It also provided that any claim arising from circumstances notified to the insurers in accordance with that clause should be deemed to have been made during the period of the policy. Accordingly, the claimant sought to recover its loss from the underwriters for the 1993-94 year on the grounds that, although the claims themselves had not been made during that year, they arose out of the circumstances described in Mr. Harvey's letter of 28^th January 1994 and were therefore to be treated as having occurred during the period of cover. The Lloyd's Syndicates have accepted liability in respect of those claims, both under the primary and excess layer policies, but the defendants have declined to do so on the grounds (among others) that their policies only cover claims actually made within the policy period and that even if they do extend to claims arising out of circumstances notified during the policy period, LMA failed to notify them of any such circumstances within that time.

6.

By orders made on 25^th February 2004 and 30^th April 2004 directions were given for the trial of a number of preliminary issues relating to the construction and effect of the excess layer policies. Before identifying those issues, however, it is necessary to set out the material terms of the various policies.

The policies

7.

The material parts of the primary layer policy provided as follows:

"Now we, the underwriters, to the extent and in the manner hereinafter provided, hereby agree:-

1.

To indemnify THE ASSURED against any claim or claims first made against them during the period of insurance set forth in the First Schedule in respect of any Civil Liability whatsoever or whensoever arising . . . . . . .

. . . . . . . . . . . . . . . . . . . .

EXCLUSIONS

The Policy shall not indemnify THE ASSURED against any claim or loss:-

. . . . . . . . . . . . . . . . . . . .

2.

Arising out of any circumstances or occurrence . . . . . . . . which were known to THE ASSURED prior to the inception of this Policy

. . . . . . . . . . . . . . . . . . . .

GENERAL CONDITIONS

. . . . . . . . . . . . . . . . . . . .

2.

THE ASSURED shall as a CONDITION PRECEDENT to their right to be indemnified under this Policy give to the Underwriters notice as soon as possible during the period of this policy as set forth in the Schedule:-

2.1

Of any circumstance of which THE ASSURED shall become aware which may give rise to a claim or loss against them or any of them.

2.2

Of the receipt of notice from any person whether written or oral of an intention to make a claim against them or any of them.

. . . . . . . . . . . . . . . . . . . .

Such notice having been given to Underwriters THE ASSURED shall give to the Underwriters as soon as possible full details in writing of the circumstances which may give rise to a claim or loss against them or any of them. Any claim or loss to which that circumstance has given rise which is subsequently made after the expiration of the period specified in the First Schedule shall be deemed for the purposes of this Policy to have been made during the subsistence hereof."

8.

The leading excess layer policy was underwritten by the Lloyd's Syndicates and was known as the 'Co-insurance policy' because it was referred to by that name in each of the other excess layer policies. It incorporated a set of clauses known as the A W G S Excess Wording which provided as follows:

"To indemnify the Assured for claim or claims which may be made against the Assured during the period of insurance . . . . ."

and contained the following clauses:

"1.

Liability to pay under this Policy shall not attach unless and until the Underwriters of the Underlying Policy/ies shall have paid or have admitted liability or have been held liable to pay, the full amount of their indemnity.

The preliminary issues arise from the respondent excess layer insurers’ contention that, although LMA became aware during the policy year 1993-94 of circumstances which might give rise to a claim against LMA relating to the pension transfers and opt-outs problem, no notice was given to excess insurers, or to anyone authorised to receive notice on their behalf, of any such circumstances or claim until 2002. We were given a brief explanation of the background in which this may have occurred, but it is unnecessary for present purposes to go into it. The preliminary issues were directed and were answered by Moore-Bick J in these terms:

The appellants appeal with respect to the judge’s answers on issues 2, 4 and 6. The respondents have served a notice seeking, if necessary, to uphold the answer on issue 2 on a different basis and appealing against the answers on issues 3 and 4, if the appellants’ appeal on issue 2 succeeds, and in any event against the answers on issues 6 and 8.

In these circumstances issue 1 no longer arises. The appellants accept the judge’s answer, which was based on sound reasons with which I entirely concur. The alternative proposition, for which the appellants contended before the judge, would have been uncommercial and improbable. The scheme would have been incoherent. The primary policies would have provided cover in respect of claims made within the policy period and claims made in a subsequent period arising from circumstances notified within the policy period. The excess policies would, in contrast, only have covered claims made within the policy period. This would have meant that, if any circumstances which might give rise to a claim on them had become known during that period, any claim arising therefrom would not have been covered under the excess layer insurance for that period, and the (quite possibly different) excess layer insurers for the next period might well have declined to refuse to renew insurance on a basis including it. Further, clauses 1 and 2 of the excess policy would appear to be unworkable unless the primary and excess layers of any claim all fall within the same insurance year. Finally, the requirement in clause 5 of the excess policy that “any circumstances of which the Assured becomes aware during the subsistence hereof which are likely to give rise to such a claim or loss” be notified immediately “if it appears likely that such claim(s) or loss(es) may exceed the indemnity available” under the primary and any underlying excess insurance indicates that this was intended to be part of a conventional claims made insurance, with the familiar extension to cover claims or losses arising from circumstances notified during its period.

Issue 2 is the first main issue arising before us. The appellants, in challenging the judge’s answer, submit that, once it is accepted that general condition 2 of the primary policy is incorporated into the excess policy by clause 7 of that policy, then, read in that context, general condition 2 obliges LMA to give notice to their excess insurers of “any circumstances of which [they] shall become aware which may give rise to a claim or loss against them or any of them”, as a condition precedent to any right to be indemnified by excess insurers in respect of any such claim or loss subsequently arising. The judge rejected this submission, holding that general condition 2 of the primary policy continues, even in the context of the excess policy, to require notice only to the primary insurers, although notice so given operates as a relevant trigger for cover under the excess (and not merely the primary) insurance.

Mr Hancock QC for the appellants supports their appeal on this issue by submitting that the natural meaning of “the Underwriters”, once general condition 2 is read in the context of the excess insurance, is to refer to the excess insurers. He points out that, where the excess wording wants to refer to the “Primary Insurers”, it does so. But that point is of little weight, when dealing with a condition incorporated only by reference by the general words of clause 7 of the excess layer. Mr Hancock makes more powerful points when he refers to the benefits that could attach to notice being received by the excess layer insurers at the same time as the primary insurers under general condition 2. There could he submits be all sorts of benefits by way of the opportunity to take an interest in the development and handling of the circumstances, or of any claims or losses arising from them, or to make appropriate reserves or reinsurance arrangements. That might be so, but on the other hand, if notice of any circumstances which “may give rise to a claim or loss” on the part of “the Assured” (i.e. LMA) must in every case to be given to excess layer insurers, followed “as soon as possible by full details in writing”, then excess layer insurers could well be receiving a volume of notifications and a great deal of information in which they would, being excess insurers, never realistically be interested. Mr Hancock further points out that, if the wording of general condition 2 as incorporated into the excess insurance is manipulated, in the way the judge did, to make it continue to refer to notice to the primary insurers, it adds little if anything to excess insurers’ protection, since the effect of clauses 1 and 2 of the excess wording will anyway be that they will not be liable unless the conditions precedent to liability under the primary insurance have been met. Again, however, that point seems to me of limited weight, when one is dealing with a condition incorporated only by the general reference in clause 7.

In my view, the judge was right in the answer he gave on this issue. First, clause 5 deals expressly with claims notification to excess insurers. It requires immediate notification in writing of “any claim(s) made against [LMA] or the discovery by [LMA] of any loss(es), or any circumstances of which [LMA] becomes aware during the subsistence hereof which are likely to give rise to such a claim or loss” only “if it appears likely that such claim(s) or loss(es) may exceed the indemnity available” under the primary insurance cover. This on its face suggests that excess insurers are interested only in matters that are likely to affect their layer, and, unless and until such a situation arises, are prepared to leave the handling of lesser matters to primary insurers, even though there is always a risk, as clause 5 acknowledges, that such matters may later escalate and concern the excess layer.

Mr Hancock observes, rightly, that there is nothing to prevent an insurance requiring successive or overlapping notices or notifications. But, in considering whether this is what the present excess insurance contemplated, it is relevant to bear in mind that clause 5 is a specific part of a standard excess wording, while general condition 2 is only incorporated by the general terms of clause 5. That neither provision was drafted with the other expressly in mind is fairly obvious, when one compares their language (“circumstances …. which may give rise to a claim or loss” against LMA in general condition 2 and “circumstances …. which are likely to give rise to such a claim or loss” in clause 7).

Mr Hancock suggested at one point that the language of general condition 2 might be manipulated or understood so that it too only encompassed circumstances “likely” to give rise to such a claim or loss. But there would be no basis for doing that to general condition 2 in the context of the primary insurance, and so further incongruity would result if it were done to general condition 2 in the context of the excess insurance. The suggestion was not however an essential part of Mr Hancock’s argument; but the difference between the language in which the two provisions address a similar subject-matter does underline the point that they cannot be read as if they were both express provisions of a unitary contract prepared by a single draftsman. Moreover, clause 7 provides that “Except as otherwise provided herein” the excess policy is subject to the same terms, conditions and definitions as the primary insurance. Although general condition 2 is incorporated, it should be read in a sense consistent with the general tenor of the express language of the excess wording.

Second, general condition 2 is only concerned with the giving of notice of circumstances of which LMA becomes aware during an insurance period, which may mature into “a claim or loss against them”. A primary purpose of general condition 2 consists in its further role as a trigger to the extension of cover, bringing within one insurance period any claim or loss arising after that period from circumstances thus notified during that period. There is nothing in general condition 2, or elsewhere, to require any notification to excess insurers of any claim or loss, other than under clause 5 a claim or loss which appears likely to impact on the excess layer. A claim may be made against LMA and notified to primary layer insurers in one insurance period which appears to have no prospect of impacting on the excess layer. The prospect of its impacting on the excess layer may only appear years later. A duty to notify it to excess insurers will then for the first time arise under clause 5. But the position will, on the appellants’ case, be quite different if the claim against LMA happens to have been preceded by prior awareness on LMA’s part during the original insurance period of the circumstances later giving rise to the claim. Experience indicates that such prior awareness far from always exists. Claims are made out of the blue. So, on Mr Hancock’s suggested interpretation of general condition 2, there would again be a considerable incongruity in the suggested scheme of the primary and excess insurances.

Third, the importance attaching to the role of the primary insurance and primary insurers is evident from the terms of the excess policy wording. Clauses 1, 2, 3 and 4 all underline it. The second sentence of clause 7 points a contrast, in requiring excess insurers’ consent to any amendment of the primary insurance “in respect of which the Primary Insurers require an additional premium or a deductible”. In the light of this importance and the previous two points, it does not seem in any way incongruous that excess layer insurers should have been prepared to accept that the only notice they would receive was of circumstances, claims or losses likely to impact their layer, and that, unless and until that situation arose, any circumstances, claims or losses should be notified to and handled by primary insurers. There was also before the judge uncontradicted market evidence from a former underwriter, Mr Carey, to the effect that there was nothing surprising or uncommercial about such a situation.

Reference was made to clause 3 of the excess wording. It has no application on the facts of this case. But, in a case where the primary layer was exhausted during the insurance period, so that clause 3(b) applied, Mr Hancock submits that, on the judge’s interpretation of general condition 2, notice of circumstances would still only need to be given to primary insurers, even though they no longer had any interest. I do not think that that follows. The excess insurers would under clause 3(b) have become the primary insurers, and would as such be entitled to the benefit of general condition 2.

For these reasons, and those given by the judge, with which I fully concur, I agree with the judge’s answer to issue 2.

In these circumstances, it is unnecessary to consider the correctness of the judge’s answer to issue 3, relating to adequacy of notice to the brokers to constitute notice to excess insurers, and I express no opinion on it. I turn to the judge’s answer to issue 4, relating to the letter of 28^th January 1994, the terms of which appear in paragraph 2 above. There were two sub-issues below, one whether the letter could be regarded as directed to and as notice to excess layer underwriters, the other whether it amounts, within general condition 2 of the primary layer policy, to notice to the primary layer underwriters of circumstances which might give rise to a claim or loss against LMA at all. The former sub-issue does not arise in view of the answers given above to issues 1 and 2. The latter remains relevant, since, if the letter was not such a notification, the potential trigger to excess layer cover provided by the answer to issue 2 could not exist. Mr Hancock submits that the letter was not a notice of circumstances which might give rise to claims; and that it was not addressed even to the primary layer underwriters on the 1993-94 policy year. Rather, he submits, it was a reassurance to potential underwriters on the next policy year that there were no problems to take into account when fixing the premium and terms for renewal cover. He points out that in prior correspondence the brokers, Bowrings, had communicated to Resource Underwriting Limited, agents for Lloyd’s syndicate 657, which underwrote 100% of the primary layer and were the leading Lloyd’s syndicate on the excess layer, that LMA involvement in pensions transfers was very much at the bottom end of the scale and they had very little exposure, and should not cause any financial problem.

The judge dealt with this aspect of issue 4 as follows:

“38.

I think Mr. Hancock was right in saying that this letter was addressed to the prospective insurers for the 1994-95 policy year, but I do not think that prevented it from providing adequate notice of the matters to which it refers for the purposes of the current policy to the extent that the same underwriters were involved. What the insurers required was prompt notification of any circumstances that might give rise to a claim. Provided they were given the information required by General Condition 2 in a form that made it clear that they were being notified of circumstances that the insured thought might give rise to a claim, it was immaterial that the information was contained in a letter directed to renewal of cover. Since the prospective insurers at Lloyd's for the ensuing year were the same as for the expiring year, this letter was in my view capable of constituting notice to them of the circumstances described in it.

39.

Finally Mr. Hancock submitted that the letter was not a warning that claims might be forthcoming but an assurance that they would not and so did not amount to notification of circumstances that might give rise to a claim. The fact that the letter was addressed to the prospective insurers for the ensuing year inevitably meant that it was directed as much to the risk of claims being made during the period of that policy as to the notification of circumstances that might give rise to claims that would fall under the current policy. The two were directly related. I see no reason, therefore, why the same letter should not be capable of satisfying both objects where the same underwriters are involved. In my view the letter was both notification of circumstances that might give rise to a loss and an assurance that the insured was not aware of any other circumstances that might do so. The underwriters could therefore proceed on the basis that any claims arising out of those circumstances would fall within the expiring policy and could be excluded from the ensuing policy.”

The arguments presented by Mr Hancock before us mirrored those presented to and rejected by the judge in these paragraphs. We did not find it necessary to call on Mr Weitzman to respond to them. In the circumstances, I need only say that I agree with the judge’s reasoning, and would answer issue 4 in the same terms as he did.

There is no appeal on issue 5. Issue 7 was regarded by the judge as covered by other issues, and does not arise before us. That brings me to issues 6 and 8, which I take together and which involve points of general interest concerning the nature and effect of a provision such as clause 5. The points fall to be considered in the particular light of a judgment given by my Lord, Waller LJ, in Alfred McAlpine plc v. BAI (Run-Off) Ltd. [2000] 1 Ll.R. 437.

In that case my Lord identified a possibility that a claims notification provision such as clause 5 might be neither a condition precedent to liability for the claim nor a clause all breaches of which sounded simply in damages. It might be an innominate term, the consequences of which, he said, might be so serious as to entitle an insurer to reject the claim albeit that the breach was not so serious as to amount to a repudiation of the whole insurance contract. This argument had not been pleaded by the insurer (BAI), and Waller LJ had, therefore, some doubt whether it was open. But, apart from that, he said (in paragraph 35) that

“it was, however, bound to fail unless BAI could demonstrate that there was a serious breach of condition 1(a) which had serious consequences and that in reliance on such a breach the claim had been rejected.”

In his view the breach of the claims notification clause, condition 1(a), “was very limited in that BAI had sufficient details to enable them to investigate the claim” and that “Furthermore, by that time BAI had at least some details of the claim they had not suffered any irremediable prejudice”. Waller LJ was also doubtful whether BAI’s conduct could be said to amount to a final rejection of the claim, but, if it did, he said, it was not justified. Any defence which was open to BAI was therefore not available. It follows from these passages that Waller LJ’s statements about the possible nature and effect of a claims notification clause were obiter.

In order fully to appreciate Waller LJ’s reasoning it is appropriate to summarise and set out further extracts from his judgment. First, I note that there was no challenge before him to the conclusion of the judge (Colman J) that condition 1(a) (which provided that “In the event of any occurrence which may give rise to a claim under this policy the insured shall as soon as possible give notice to the Company in writing with full details”) was not a condition precedent. Waller LJ also agreed with Colman J that it was not easy to contemplate that a failure to comply with an ancillary provision relating to one claim under a policy could amount to a repudiatory breach of the whole contract of insurance (paragraph 22). He considered Taylor v. Builders Accident Insurance Ltd. (1997) MCLC 247, a decision in which HHJ Byrt QC contemplated that a breach of a claims notification provision might be “a fundamental breach of an important term in the contract” such as to “enable the insurance company to repudiate notwithstanding the condition breached is not stated to be a condition precedent”. That reasoning, as Waller LJ observed, does not clearly address any possibility of a breach allowing rejection of the claim without repudiation of the whole contract.

But Waller LJ went on:

“26

I do not myself think that the choice should necessarily lie between a construction which would involve condition 1(a) being a condition precedent, and condition 1(a) simply giving rise to a claim for damages. It seems to me that once a condition such as condition 1(a) is construed as something less than a condition precedent, it will still be important to ascertain precisely what its contractual effect is intended to be and what the effect of a breach of that term will be. For example, if no details of the incident in relation to which RCCL was making its claim were ever supplied, despite the insurers' requests for them, would BAI still be bound to pay, and simply be left with a remedy in damages for breach of the condition? Certainly if the consequences for BAI were that they had been seriously prejudiced, it seems to me unreasonable that that should be so. Accordingly it seems to me one should consider the possibility that a breach of condition 1(a) might in some circumstances be so serious as to give a right to reject the claim albeit it was not repudiatory in the sense of enabling BAI to accept a repudiation of the whole contract. The very fact that condition 1(a) is aimed at imposing obligations in relation to individual claims which BAI might be obliged to pay, ought logically to allow for the possibility of a "repudiatory" breach leading simply to a rejection of a claim.

27.

I accept, I should say, that it is possible for the terms of a policy by express language to be clearer than this term as to what its intended effect should be. ….”

He then examined an Australian case, Trans-Pacific insurance Co. (Australia) Ltd. v. Grand Union Insurance Co. Ltd. (1989) 18 NSWR 675; (1990) 6 ANZ Insurance Cases 60-949. Giles J was there concerned with breach of a clause requiring co-operation in relation to claims, including such reporting of claims as was necessary to fulfil that requirement. The clause applied to a particular cession under a facultative/obligatory reinsurance which Giles J regarded as bringing into existence a separate contract of reinsurance. Giles J held that, although there had been tardiness and a breach in supplying information under the clause, the insured had not “evinced an intention no longer to be bound by the reinsurance contract …. or showed that it intended to fulfil the contract only in a manner substantially inconsistent with its obligations”. But, as Waller LJ noted, Giles J referred in the course of his reasoning to a brief passage in Phoenix v. Halvanon [1988] QB 216, 241, where Hobhouse J, in the context of another fac/oblig treaty described an implied duty on the insured to exercise due care and skill in the conduct of its business as involving an innominate term or terms and said that “therefore the consequences of any breach for any particular cession or any individual claim or, indeed, for the contracts as a whole, must depend on the nature and gravity of the relevant breach or breaches”.

Waller LJ continued:

“32.

I see no reason however why condition 1(a) should not be construed as an "innominate" term as per Hongkong Fir Shipping Co Ltd (supra) where the consequences of a breach may be so serious as to entitle BAI to reject the claim albeit the breach is not so serious as to amount to a repudiation of the whole contract. Mr Lynagh submits there is no support in the judgment of Giles J for the proposition that the consequences may be so serious as to give a right to reject the claim. I accept that Giles J took the view that there was a separate contract of reinsurance in relation to the risk the subject matter of that case, and thus did not decide the point. But I do think the inference to be drawn from the passage quoted at 702F - 703A of the judgment of Giles J supports the view that I take. I accept Mr Walker's submission in this regard. It seems to me that the payment of individual claims are severable obligations and that where an insured is bound to carry out one obligation in order to receive the benefit of the insurer's obligation by implication the insured is accepting that if he fails in a serious way to carry out his part of that bargain he will not receive what he has bargained for.

33. Thus the correct analysis of condition 1(a) I would suggest should be as follows. Compliance with condition 1(a) is not by the policy made a condition precedent to liability, thus it is not enough for BAI to establish a failure to supply full details as soon as possible in order to resist the claim. That much is conceded.

34. Condition 1(a) is however an innominate term. Breach of it, however serious, would be unlikely to amount to a repudiation of the whole contract of insurance. Furthermore, it is not a term the breach of which, or any breach of which, would entitle the insured not to pay the claim because that would simply make it a condition precedent. But, in my view, a breach which demonstrated an intention not to continue to make a claim, or which has very serious consequences for BAI, should be such as to entitle BAI to defeat the claim. If a term is a condition precedent to liability, any breach defeats liability but does not lead to a repudiation of the whole contract. I see no reason why although a term is not a condition precedent so that any breach defeats liability, it cannot be construed as a term where a serious breach defeats liability.”

There is some subsequent authority. The statements in BAI were considered in K/S Merc-Skandia XXXXII v. Certain Lloyd’s Underwriters (The Mercandian Continent) [2000] LI.R.I.R. 694; [2001] 2 Ll.R. 563, by Aikens J and the court of appeal. The insured was in breach of a clause requiring the assured to keep the underwriters fully advised. Aikens J held that the approach suggested by Waller LJ involved considering whether underwriters had in fact been seriously prejudiced, which he concluded that they had not been. The Court of Appeal dismissed an appeal relating to this conclusion of fact. Longmore LJ said, again obiter, that in BAI “this Court decided that, at any rate in insurance cases, there was a further category of term in addition to the three categories identified above, namely a term a breach of which was so serious for underwriters that it would give them a right to reject the claim without having to accept the breach of contract as being a repudiation of the contract as a whole”. I have already said why I read the reasoning on this aspect in BAI as obiter. Further, as I understand Waller LJ’s reasoning, one should re-phrase the last part of Longmore LJ’s statement so that, instead of reading “without having to accept ….” it would read “without necessarily either having a right to or having to ….”.

In another case, Glencore International AG v. Ryan (The Beursgracht) [2002] Ll.R.I.R. 335, the Court of Appeal was concerned with a risk which was automatically insured under an open cover, and with a failure subsequently to declare the risk. HHJ Hallgarten QC, whose decision was upheld on this aspect, concluded that the duty to declare a risk did not operate as a condition precedent to liability in respect of a claim on it. Tuckey LJ giving the sole reasoned judgment in this court said:

“If underwriters had felt that the commercial considerations which Mr Saloman has referred to were so important to them, they could easily have made the making of declarations a condition of the cover.”

Applying the reasoning in BAI, HHJ Hallgarten and this court then considered whether the breach was so serious that underwriters were “entitled to avoid liability, not for the whole cover, but for that risk”. Both held that it was not. Whether the reasoning in BAI was or is binding or correct was evidently not argued and was not considered or decided.

The reasoning in BAI has only been applied to enable insurers to avoid liability for a particular claim in one case to which we were referred, the first instance decision of Buckley J in Bankers Insurance Co. Ltd. v. South [2003] EWHC 380, by which we are again not bound. That was a case where the insured riding a jet ski was in July 1997 in collision with another jet ski, whose driver was seriously injured. The insured failed to comply with claims notification provisions requiring him to give insurers notice for some three years. Buckley J held that the breaches were manifestly serious, and that, while it did “not necessarily follow that such serious delay would have serious consequences for Bankers, the position here …. is that by the time the question of cover is resolved Bankers will have lost some three and a half years”. He drew inferences that some significant further fading of memories would have occurred and that some of the witnesses would probably be more difficult to trace and some probably unwilling to assist after so long. He granted insurers a declaration that they were not liable to indemnify the insured.

The first question is whether clause 5 is a condition precedent with respect to the particular claim so that any breach of its provisions means that insurers are not bound to meet the claim. This question cannot be entirely separated from the reasoning in BAI. The reasoning in BAI offers an alternative remedy, which if available might disincline a court from treating a clause such as clause 5 as a condition precedent. If the innominate analysis with the consequences suggested in BAI is not available, then it may be the court would be more favourably inclined to construe clause 5 as a condition precedent.

Accepting that, even if the reasoning in BAI is rejected I cannot accept that compliance with clause 5 can or should be construed as an implied condition precedent to excess insurers’ liability for any claim, loss or circumstances to which such non-compliance relates. The excess wording is, we are told and as its title (AWGS Excess Wording) suggests, a standard market wording. It is used by and between professionals in the market, who are or should be taken to be well aware of the various possibilities open to insurers. They are able, if they need to, to take advice to make such wordings either more or less stringent. Very probably, such a wording, either in its drafting or in its use, reflects a general appreciation, on the part of the market professionals using it, of what will be acceptable and/or attractive to each other. Where the wording intends to introduce a condition precedent, it shows itself well capable of doing so expressly: see the language of clause 1 and in my view clause 2 as well as clause 4 and the second sentence of clause 7. For reasons I have already given, I do not think that it is realistic to try to construe clause 5 by way of a close comparison with general condition 2, but, if one were to do so, the result would be to underline the difference between on the one hand the clearly worded condition precedent in the first part of general condition 2 and on the other hand clause 5. Even in general condition 2 there is a contrast between the condition precedent in the first part and the absence of any provision making its subsequent provision requiring full details in writing as soon as possible a condition precedent. Finally, I observe that there is a difference between a provision requiring notice as soon as possible to primary insurers of circumstances which may give rise to a claim, which means that the primary insurers will have the opportunity to investigate and handle the matter, and a provision like clause 5 which looks to a potentially different stage when excess insurers have to be involved. It follows that I agree with the judge in so far as he held that clause 5 was not a condition precedent.

This brings me to issue 8. If clause 5 is not a condition precedent, is it an innominate term and may it then have the consequences suggested by Waller LJ’s reasoning in BAI? For my part, I am prepared to proceed on the basis that clause 5 may be regarded as an innominate term, although I agree with Waller LJ in BAI that it is not easy to conceive of a breach of such an ancillary term in an insurance like the present as going to the root of the whole contract, and, even if this is conceivable, there would be further questions as to whether such a breach could, if accepted, enable insurers to reject liability for any claim accrued by the time of such acceptance, including any to which the breach related. But no such breach is suggested in this case. What is suggested, relying on the reasoning in BAI, is that what is described as a repudiatory breach arising from the failure to give notice for some years may mean that insurers could deny liability for the particular claim or claims relating to the pensions transfers and opt-outs problem to which such failure related.

Moore-Bick J accepted this. He did so on the basis that Waller LJ was enunciating a conclusion on construction. He was “simply recognising that a serious failure to comply with a notice provision such as clause 1(a) in that case was capable of being treated as a failure to comply with a condition precedent” and was seeking “to identify and explain the nature of the obligation contained in clause 1(a), which is essentially a matter of construction”. Moore-Bick J went on:

“In my view Mr. Hancock was right in saying that the principles relating to discharge by breach are not directly applicable to a situation of the kind that arises in this case. As I understand the passages from his judgment quoted earlier, Waller L.J. was drawing on those principles by analogy to explain how the breach of a term such as that contained in clause 1(a) could have different consequences depending on its gravity. I do not understand him to be importing from the law on repudiation the necessity for the insurer to "accept" the breach as terminating his liability to pay, much less the principles governing accrued rights. If, as paragraphs 33 and 34 suggest, a serious breach of clause 1(a) has the same consequences as a failure to comply with a condition precedent, no "acceptance" is necessary; the insurer is simply entitled to reject the claim on that ground. Moreover, there is another reason why I am unable to accept Mr. Weitzman's argument. The submission that in the present case the claimant already had an accrued right to be indemnified by the defendants at the time when they decided to reject the claim for failure to comply with the notification requirements proceeds on the assumption that that right was unconditional. In fact, however, it was conditional on its complying, at least substantially, with the requirements of clause 5. If the claimant failed to comply with clause 5 in a manner that was sufficiently serious to entitle the defendants to reject the claim, its right to be indemnified remained at best conditional until the underwriters elected not to rely on their rights or in some other way lost their right to reject the claim on those grounds. However, neither of those events occurred.”

Mr Hancock supports the judge’s reasoning. He submits that the correct reading of BAI is that it reflects the parties’ implied understanding that, in the event of a serious breach (which is what he submits that the parties would, like the judge in his declaration, probably have concentrated on, rather than serious consequences), the insurers would be free of liability. He disclaims any submission that Waller LJ was basing himself on a rule of law. However, as I read BAI, the reasoning was based on a concept of repudiatory breach of the particular claim, requiring both proof of “a serious breach …. which had serious consequences” and proof “that in reliance on such a breach the claim had been rejected” (see paragraph 19 above). As to this, a particular contract may be severable into separate parts. For example, a failure under a contract for sale by instalments to make due delivery of one instalment within the contractually stipulated time may be accepted as repudiatory of that instalment, but does not necessarily mean that the whole contract comes to an end (cf Benjamin on Sale of Goods (2002) paragraphs 8-073 to 8-076). But that is a very different position from the present. The present insurance is a composite contract. The primary quid pro quo for insurers’ obligation to pay claims under the insurance is the premium, which is incapable of being severally allocated to any particular risk or claim. The common rule whereby fraud in pursuit of a claim may forfeit the whole of the particular claim (cf Axa General Insurance Ltd v. Gottlieb [2005] EWCA Civ 112) derives from policy considerations present in the case of fraud, which have no real parallel in the present context. No authority was cited to us, apart from BAI and its successor cases, in which any court has suggested that a party to a contract may be relieved from a particular obligation under a composite contract such as the present, by reason of a serious breach with serious consequences relating to an ancillary obligation, absent some express or implied condition precedent or other provision to that effect. Either some conditional link can, as a matter of construction, be found between performance of the two obligations or it cannot. Where such a link cannot be found as a matter of contractual construction, I see no basis for a new doctrine of partial repudiatory breach to, in effect, introduce one.

A claims notification clause like clause 5 is an ancillary provision. Breach of such a provision is capable of sounding in damages. But I am unable as a matter of construction or implication to find in clause 5 any provision that insurers will be free of liability in the event of a serious breach and/or a breach with serious consequences. Even if one assumes that it might or would have been reasonable for the parties to agree such a provision, reasonableness is not the test for implying a term. Nor do I see any basis for creating a new rule of construction or law which would impose such a provision on these parties. Representing opposite sides of the market with opposing interests on the point, they must, as I have said, be taken to be fully aware of the possibilities of agreeing whatever they could, acceptably to each other and in accordance with any other constraints (practical or legal, as to which I say nothing), negotiate. For my part, I do not think that it is obvious that the broking and underwriting sides of the market would regard even as reasonable a provision such as suggested in BAI, and I do not see any reason why the court should impose it on them. A test of “serious breach” and/or “serious consequences” might have some drastic and unfair consequences. Suppose a year’s delay, in consequence of which insurers lost the opportunity to make or (e.g. because of insolvency) to recover a reinsurance claim or a subrogation claim worth £50,000. That would be a breach serious in nature and consequences. But suppose the insurance claim was itself for £1 million or £100,000 or even £75,000. Why should insurers have the right to reject the whole claim? Further, if the test in BAI is applied, insurers must prove serious consequences, as well as a serious breach. If they can prove serious consequences, then these will often be capable of quantification, in one way or another, even if only as losses of a chance or opportunity, and can be set off against the claim.

Of course, there are cases, like the Bankers Insurance case decided by Buckley J, where it may be said that the consequences are too intangible to measure in precise financial terms, although, where an insured’s breach has caused this difficulty, courts should incline to a quantification favourable to insurers. But the difficulty is anyway no justification for the introduction, whether as an implied term or as a rule of law or by a previously unknown extension of the doctrine of repudiatory breach or on any other basis, of a novel form of protection for insurers. If insurers consider that they want or need such protection, they can and should try to express it in their insurance contracts and see if insureds and the broking market will accept it. In a passage quoted in Lord Diplock’s speech in Scandinavian Trading Tanker Co. AB v. Flota Petrolea Ecuatoriana (The Scaptrade) [1983] QB 529, 540-1; [1983] AC 694, Robert Goff LJ said, in relation to a submission that the court could relieve against the supposed unfairness that might result from withdrawal of a vessel under a time-charter for a single late payment of a hire instalment: “Parties to such contracts should be capable of looking after themselves”. That thought is equally applicable here. English insurance law is strict enough as it is in insurers’ favour. I see no reason to make it stricter. I would therefore disagree with the reasoning suggested in BAI, and answer issues 6 and 8 by declaring that the giving of notice in accordance with clause 5 was not a condition precedent to liability, and that, even in the event of a serious breach of clause 5 having serious consequences, the insurers have no such defence or defences as mentioned in issue 8(1), (2) and (3). Since all that Mr Hancock has sought to establish in these proceedings is a right to reject the particular claims, it is unnecessary to say more about the classification of clause 5 in response to issue 6.

I agree with the judgement of Mance LJ.

After reading in draft the judgement of Waller LJ, I prepared a short judgement on issue 8. Having read the judgement of Mance LJ, I have concluded that he has expressed in better terms than I had the reasons why Mr Hancock’s submissions should not succeed. My view was and is that the principle suggested in the Alfred McAlpine case has no basis in the law of contract unless an appropriate term can be implied into the contract. That appeared to be the view of Mr Hancock who sought to support the conclusion of the judge upon the basis that the contract should be construed so as to imply an appropriate term. That submission fails as there is no need to imply such a term and in any case it did not appear possible to set out with sufficient precision the term.

There is only one aspect of the appeal on which I would like to express some views. It is the aspect which calls for consideration of what I said in my judgment in MacAlpine(Alfred) Plc v BAI(Run off)Ltd .

I am not concerned to establish whether what I said in that case in a judgment with which Peter Gibson LJ and Buxton LJ agreed was ratio decidendi or simply obiter. I am concerned to seek to demonstrate that the view I expressed is not as heretical as my lords, according to their judgments would imply, and should be welcomed rather than rejected.

What I was suggesting in relation to the contract of insurance in BAI was that a clause which provides that notice of a claim must be given “as soon as possible” may not be a condition precedent, but may, if it is breached in a way which seriously prejudices the insurer, give a right to reject a claim rather than leave the insurer simply with a claim for damages.

That the law should recognise that possibility seems to me to accord with what the parties would have contemplated where the insured has agreed to provide details “as soon as possible” and where an inexcusable delay in so doing has caused serious prejudice. It also incidentally in my view accords with justice.

The serious prejudice I had in mind in that case was that the insurer had simply had no chance to examine the circumstances of the claim and was as a result simply unable to say whether some action could have been taken to defeat the claim. In such a case a set-off for the damages for “loss of a chance” is illusory as a remedy. By the nature of things it may be difficult if not impossible for the insurer to say whether there were circumstances which would have enabled him to defeat the claim, or what his chances of so doing were.

There are various aspects to considering this question. First is a term requiring notice of a claim “as soon as possible” which is not a condition precedent, the type of term the breach of which could ever give a right to the insurer to assert that the obligation to pay claim has not arisen or that the contract as a whole has been repudiated? Second if it could be construed as such a term, is the obligation to pay one claim a severable obligation, which could be discharged by some failure to comply with the notice clause in relation to that claim? Third what kind of failure will discharge this latter obligation? Fourth is the answer to any of these questions an answer that depends on construction of the contract or is it provided by some rule of law?

I accept that a breach of a notice clause would be unlikely to amount to a repudiation of the contract as a whole, but, I suggest, it is not impossible that it could in extreme circumstances of consistent breach over a number of claims. More likely is a situation in which the insured has details of a claim to the knowledge of the insurer and the insured refuses even on request to supply the details. It is likely that in such a situation the term will at the very least be construed as suspensive disallowing the claim until details are provided. The question is whether if there has been a failure to notify the insurer in any way of a claim when obliged to do so “as soon as possible”, there should ever come a stage where the delay in informing insurers or supplying details is such that the insurer should be entitled to say, he is so prejudiced by the lack of information that he should be entitled to reject the claim. Can, in other words, such a term be construed as an innominate term the consequences of a breach of which depends on the nature and gravity of the breach?

It is not unknown to the law of contract that in relation to severable obligations a failure in some respect may give rise to a right to “reject” simply and only the particular counterpart of the severable obligation. Thus in an instalment contract, a breach of condition in relation to a particular instalment, may give rise to a right to reject the particular instalment, and only the particular instalment [see Benjamin para 8-083, and the cases cited in the note]. It is of some interest that Benjamin later draws attention to the fact that there can be a difference between the right to “reject”, and the right to “treat the contract as at an end”[see 12-025].

In the insurance context in Scottish law it seems that a fraudulent claim simply gives the insurer the right to reject “that” claim and not any right to have the contract discharged as a whole [see Macgillivray para 19-61 supported by Reid & Co v Employers’ Accident & Live stock Ins Co (1899) 1 F.1031,1036]. In English law the fraudulent claim gives a right to avoid the contract as a whole, but it has been questioned whether that means “ab initio” or whether it is as from the date of avoidance [see the same paragraph of Macgillivray, and the reference to the “tentative view” of Mance LJ in Agapitos v Agnew [2002] 1 All E R (Comm) 714 at 730].

As I pointed out in BAI furthermore Hobhouse J in Phoenix General Insurance Co of Greece SA v Halvanon Insurance Co Ltd [1985] 2 Lloyd’s Reports 599 at 614 when holding that a term should be implied into facultative/obligatory reinsurance contract that the reassured should conduct his business in accordance with market practice and with due diligence in certain respects including the maintenance of records and accounts, the acceptance of risks and the investigation of claims, stated :- “The term or terms are all innominate and therefore the consequences of any breach for any particular cession or any individual claim [my emphasis] or indeed for the contracts as a whole, must depend on the nature and gravity of the relevant breach or breaches.” He therefore expressly contemplated that the consequence of a breach might be confined to an individual claim and, within that context, depends upon the gravity of the breach.

Conceptually there is, as I would see it, no difficulty in construing an obligation to pay a claim as a severable obligation. Strictly the obligation on the insurer is to pay damages, but it is precisely to that severable obligation that a “condition precedent” provision applies. It is precisely to that obligation that the suspensive effect of the clause applies. Conceptually accordingly, I can see no difficulty in construing a notice provision as being an innominate term which provides the insurer with a right to reject the claim if there is a breach of sufficient gravity i.e. a breach which has seriously prejudiced the insurer.

Mr Hancock was pressed in argument to suggest that some form of implied term is called for. I do not accept that. The question is whether the term is one which if breached seriously i.e. in a way that seriously prejudices the insurer, gives a right to reject the claim. In addition I do not accept that Mr Weitzman is right in suggesting that in some way the whole concept of repudiation, and acceptance has to be built into the operation of such a term. Indeed in relation to the absence of notice prejudicing the insurer, it would be strange if someone who does not know that he has not had notice of a claim, should be disentitled from rejecting the claim because he has failed to accept some repudiation about which he knew nothing.

If the clause is of the requisite type, it works I suggest on the basis that for a serious breach i.e. a breach which has caused the insurer serious prejudice, the insurer has an entitlement to reject the claim.

I should add that the position that was under consideration in BAI was quite different from the position being considered in the instant case. For one thing there is an express condition precedent which is intended to protect against the prejudice that we were concerned with in BAI. I doubt whether that prevents condition 5 being an innominate term because, albeit very unlikely, there could be a refusal to provide details despite requests which might have a consequence for which damages would not be the appropriate remedy. What is clear to me is that the excess underwriters having agreed that notice of circumstances should be given to the primary layer underwriters, would have little chance of demonstrating serious prejudice. But because we are dealing with preliminary issues, that would have to be for another day.

If the above is heretical then so be it. But it is of some comfort that Macgillivray paragraph 10-13 expresses the view that “the increased flexibility in remedy which the introduction of such a term gives to the courts is welcome…” adding I accept “though the legal basis for the term is perhaps open to question”, and I have endeavoured to deal with that latter point. Other leading text books Professor Clarke’s The Law of Insurance Contracts, Section 26-2G, Colinvaux’s Insurance Law 7^th Edition, 2^nd Supplement, paragraph 9.02, and Butler and Merkin on Reinsurance Law, sections B-0006 to B-0011 refer to the principle without adverse comment.

As Macgillivray suggest in para10-13, the decision in BAI, has been followed in the Court of Appeal in K/S Merc-Skandia v Certain Lloyd’s Underwriters [2001] 2 Lloyd’s Reports 563. Longmore LJ with whose judgment Robert Walker LJ (as he then was) and Carnworth J (as he then was) agreed, referred to and applied BAI without any adverse comment. They upheld the judge Aikens J it is right to say in ruling that no serious consequence had been shown.

Tuckey LJ with whose judgment Jonathan Parker LJ agreed also applied BAI again without adverse comment in The Beursgracht [2001] EWCA Civ 2051, in the slightly different context of “declarations” under a floating policy, and in doing so were upholding the application thereof by Judge Hallgarten QC. They upheld that judge also in ruling that serious consequences had not been established.

The principle has been applied by Buckley J in Bankers Insurance Company v Patrick South [2003] EWHC 380(QB). He did hold that the insurers in that case were so seriously prejudiced that they should be entitled to reject the claim.

I would differ from my lords on their answer to issues 6 and 8. Otherwise I agree with the judgment of Mance LJ.