Baker v. Black Sea & Baltic General Insurance Company Ltd [1998] UKHL 18; [1998] 2 All ER 833 (20th May, 1998)

BAILII is celebrating 24 years of free online access to the law! Would you consider making a contribution?
No donation is too small. If every visitor before 31 December gives just £1, it will have a significant impact on BAILII's ability to continue providing free access to the law.
Thank you very much for your support!

	[Home] [Databases] [World Law] [Multidatabase Search] [Help] [Feedback]
	United Kingdom House of Lords Decisions
You are here: BAILII >> Databases >> United Kingdom House of Lords Decisions >> Baker v. Black Sea & Baltic General Insurance Company Ltd [1998] UKHL 18; [1998] 2 All ER 833 (20th May, 1998) URL: http://www.bailii.org/uk/cases/UKHL/1998/18.html Cite as: [1998] UKHL 18, [1998] 2 All ER 833, [1998] WLR 974, [1998] 1 WLR 974

I have had the advantage of reading in draft the speech of my noble and learned friend, Lord Lloyd of Berwick. I agree with it and for the reasons which he gives I would allow the appeal to the limited extent which he proposes and would remit the question to which he refers to further hearing in the Commercial Court.

I have had the advantage of reading in draft the speech prepared by my noble and learned friend, Lord Lloyd of Berwick. For the reasons he gives I would allow the appeal to the extent which he proposes and would remit all questions relating to trade practice and usage for a further hearing in the Commercial Court.

The plaintiff, Mr. Colin Baker, is a member of Lloyd's Syndicate 947 ("the syndicate"). He sues on his own behalf, and on behalf of all other members of the syndicate. The defendants, Black Sea & Baltic General Insurance Co. Ltd. ("Black Sea"), carry on insurance and re-insurance business in London. They re-insured the syndicate under a continuous contract of re-insurance which incepted in 1957 and terminated on notice being given on 31 December 1968.

The run off operated satisfactorily until 1984, when Black Sea refused to reimburse the syndicate in respect of outstanding asbestos claims emanating from the United States, and claims by U.S. soldiers for long term injury to their health due to the use of chemical defoliants during the war in Vietnam. From the third quarter of 1988 Black Sea declined all further liability.

The syndicate commenced proceedings on 3 May 1989. Two other writs were issued at a later date. On 22 March 1991 Saville J. made an order for the trial of preliminary issues in respect of 12 sample claims insured by the syndicate. The preliminary issues came before Potter J. in the Commercial Court in May 1993. Potter J. identified ten questions for decision. He answered all questions save one in favour of the syndicate. This enabled him to allow or disallow all the items in the sample claims. The work involved in preparing the case for hearing was prodigious, since much of the documentation had long since been lost or destroyed. The hearing lasted 11 days. The judgment of the learned judge [1995] L.R.L.R. 287 is a model of how such cases should be handled. In a subsequent judgment he dealt with the question of interest costs and a number of other matters.

On 24 March 1995 Black Sea served notice of appeal. The syndicate cross-appealed in respect of the single issue on which it had lost. The appeal and cross-appeal came before the Court of Appeal on 13 May 1996, and were both dismissed with costs: [1996] L.R.L.R. 353.

The syndicate are the appellants before your Lordships. There is a cross-appeal by Black Sea in respect of one only of the many issues on which they lost in the Court of Appeal. In the course of the hearing before your Lordships the parties reached agreement on that issue, as a consequence of which the cross-appeal has now been withdrawn with no order as to costs. That leaves only the single issue on which the syndicate lost before the judge and the Court of Appeal. That issue may be stated as follows: Where an insurer incurs costs in investigating settling or defending claims by his insured, can the insurer recover a proportion of these costs under a quota share or other form of proportional re-insurance?

Three points need to be emphasised at the outset. The first is that the costs in question are not the costs incurred by the insured in meeting third party claims. These will often be recoverable from the re-insurer as one of the risks insured under the underlying policy. The costs which are in issue are the costs incurred by the insurer in investigating and defending claims by the insured.

It is said on behalf of the syndicate that a proportion of these costs ought to be recoverable, not because they are insured under the underlying policy--they clearly are not--but because it is of the very nature of proportional re-insurance that the re-insurer is liable for a proportion of the reasonable costs of defending and settling claims by the insured. The essential characteristic of a proportional re-insurance, so it is said, is that there is a sharing of the same underwriting fortunes, both as to risk and premium. If the syndicate were to be liable for the whole of the costs of defending and settling claims by the insured, while receiving only a proportion of the premium, the nature of the bargain would be fundamentally altered. The central question is whether the syndicate's argument in this respect is correct in law.

The second point to emphasise at the outset is that the syndicate does not seek to include among the costs of defending claims any part of their overheads or fixed costs. The costs in issue are the variable costs which are specific to a particular claim. Thus to take an example mentioned during the hearing the costs of briefing counsel to defend a claim by the insured would be included, but not the costs of employing a claims' manager. The line may be difficult to draw in practice. But the principle underlying the distinction is sound.

The third point is that the syndicate's argument is confined to proportional re-insurances. It is not suggested that the re-insurer is liable for costs under a non-proportional re-insurance, such as an excess of loss or stop loss policy, unless, of course, there is an express provision to that effect.

The only contractual document to have survived is a copy of the cover note dated 10 April 1957 issued by the brokers, Messrs. Swann & Everett Ltd. It provides as follows:

The meaning of the words "basis of cession: one retention equal to a maximum of $50,000 possible loss any one risk surplus to one retention" might not spring to the non-expert eye. Fortunately the effect of the provision has been agreed between the parties. The syndicate was not obliged to cede any of its United States or Canadian business risks. But if it did, then Black Sea was bound to accept a line equal to the line retained by the syndicate up to a maximum of $50,000 any one risk. In other words, it was a 50/50 proportional re-insurance, with an equal sharing of premiums and losses in respect of risks ceded by the syndicate. The premium to be shared was the original premium received by the syndicate net of brokerage, less a deduction of 5 per cent. "overriding commission". This was explained as follows: for every £100 of net premium due to the syndicate, Black Sea would receive £47.50, and the syndicate would receive £52.50. In profitable years the syndicate would also be entitled to receive 20 per cent. profit commission. This was not explained. But I infer (subject to correction) that if, in any period of insurance, premiums amounted to £100 and claims to £60, the syndicate would be entitled to receive 20 per cent. of £40 before distribution, with the result that Black Sea would receive £16 and the syndicate £24. Putting profit commission and overriding commission together, Black Sea would receive £13.50 and the syndicate £26.50.

In the courts below the syndicate's case was put on three grounds. In the first place it was said that Black Sea was liable for its share of the costs under the "follow the settlements and agreements" provision in the contract. But that way of putting the case met with little favour before Potter J., and still less in the Court of Appeal. It has not been pursued before your Lordships.

Secondly, the syndicate relied on an implied term of the contract, such term to be implied in order to give the contract business efficacy, or because it is what the parties to the contract must, as reasonable men, have intended. Thirdly, the syndicate relied on a term to be implied by reason of a trade practice or usage in the insurance market in London.

I shall consider each of the two latter arguments in turn. But first of all I should mention an important development which has occurred since the case was before the Court of Appeal. In September 1996 the offer which had been made to members of Lloyd's as part of the reconstitution scheme became unconditional. That meant that a company called Equitas took over the rights and liabilities of all members in respect of the 1992 year of account and prior years, for a premium in excess of £14 billion. Equitas now has the responsibility, as equitable assignee, of pursuing claims by members against their re-insurers, including the syndicate's claim in the present proceedings. The question whether members can recover a proportionate share of the cost of defending claims from their pro rata re-insurers has substantial consequences for the parties to the present proceedings. The amount at stake is about £200,000. But it has vast consequences for Equitas. The amount to be spent by Equitas in defending asbestosis and pollution claims in a single year is estimated at $100 million. Of this it is thought that about 5-10 per cent. would be recoverable from re-insurers if the appeal on the present point succeeds. Because of its market-wide interest in the outcome of the present proceedings Equitas petitioned the House for leave to intervene in May 1997, and your Lordships granted leave shortly thereafter.

In their written case the interveners indicated that they would be content to adopt the submissions advanced by the syndicate in the syndicate's written case. By agreement it was Mr. Boyd Q.C., for the interveners, who opened the case on behalf of the syndicate, rather than Mr. Clarke Q.C. But more important, the interveners said that they would be applying at the hearing of the appeal for leave to adduce fresh evidence on the question of trade practice or usage. Mr. Boyd argued that the House has a discretion to admit fresh evidence, and suggested various ways in which the fresh evidence might be received. But it is convenient to put that question on one side, and deal first with the other ground on which the syndicate relies, namely, that they are entitled to succeed by virtue of a term implied by law.

Millett L.J. regarded this as the strongest of the syndicate's arguments. But even so he did not accept it. I find his reasons convincing. The starting-point of the argument has already been mentioned. This being a proportional re-insurance it does not make sense for the syndicate to bear the whole of the cost of defending claims. Since Black Sea receive half the premium in respect of risks ceded under the re-insurance, they ought to bear half the losses, including the cost of reducing losses by defending or settling claims, from which the re-insurers benefit. That argument would indeed be a strong one if proportional re-insurance were in the nature of a partnership. But this has never been the law. It might also have been a strong argument if the profits on the business ceded to Black Sea were to be shared equally. But they were not. For the syndicate was entitled to 20 per cent. commission on profits before any distribution. Even in a year when no profits were being made (as must have been the position from quite early on) the syndicate was entitled to 5 per cent. overriding commission. It may well be that the parties intended the cost of defending claims to come out of the 20 per cent. profit commission, or the 5 per cent. overrider, or both. In the absence of any information as to how these provisions worked in practice, your Lordships do not have the material on which to say that a term is to be implied by law. As Mr. Boyd conceded in reply, we simply do not know enough to decide.

Reference was made to a well-known passage in Lord Wilberforce's speech in Liverpool City Council v. Irwin [1977] AC 239 at 253, where he discussed the different "varieties" of implication, all of which, as he explained, shade into each other. The only variety of implication which fits the circumstances of the present case is the first, which he stated as follows:

I will return to this variety of implication when I come to the question of trade practice or usage. I do not regard any of the other varieties of implication as applying in the present case, since they all depend in one form or another on a test of necessity. I would not regard that test as being satisfied in the present case, for the reasons which I have already given.

I turn briefly to see how the matter stands in the leading text books. In MacGillivray on Insurance Law, 9th ed. paragraph 33-65 we find:

The underlying authorities are not easy to reconcile, or even, in some cases, to understand. I have found most assistance from Scottish Metropolitan Assurance Co. Ltd. v. Groom (1924) 19 LI.L.Rep. 131; 20 LI.L.Rep. 44 and Insurance Co. of Africa v. Scor (U.K.) Reinsurance Co. Ltd. [1985] 1 Lloyd's Rep. 312.

In the former case the insurers incurred expenses in successfully defending a claim for a total loss under a marine insurance policy. The insurers sought to recover in respect of these expenses from their re-insurer, Mr. Groom. There was no express term in the re-insurance contract under which they were entitled to recover. But Mr. R. A. Wright K.C. argued that they could recover under an implied term. The implication was put on two grounds: first, that Mr. Groom had expressly agreed that the claim should be contested by the insurers, and, secondly, that there was an implied obligation arising from the relationship between the parties. Bailhache J. and the Court of Appeal rejected both grounds.

It was submitted on behalf of the syndicate that the case was decided on the first ground only, and that the second ground was never argued. But it is clear from the report of the argument in the Court of Appeal at p. 44 that both grounds were argued before Bailhache J. It is clear also that if that very experienced commercial judge could have found a way of deciding the case in favour of the insurers he would have done so. It may be that the second ground was not pressed in the Court of Appeal (the report is not altogether clear) but if not, it could only have been because the future Lord Wright did not think it worth pressing.

It was also argued that the case is of no assistance, since the contract was not one of proportional re-insurance. My impression is otherwise. The sum re-insured was £600 total loss only, of which Mr. Groom's share was £75. It was not suggested that the re-insurers were liable for the whole of the cost of defending the claim, but only their share. This suggests that the re-insurance was surplus business, as in the present case. The fact that the underlying policy covered partial loss as well as total loss does not mean, as Mr. Boyd argued, that the re-insurance was not proportional. It follows that Scottish Metropolitan v. Groom is strong authority in favour of Black Sea.

Insurance Co. of Africa v. Scor (U.K.) Reinsurance Co. Ltd. is another authority to the same effect. The underlying policy in that case covered a warehouse in Liberia against fire. The sums insured were $500,000 for buildings and $3 million for contents. The warehouse became a total loss. The owners of the warehouse brought proceedings in the Liberian courts. The proceedings were defended by the insurers, but they did not succeed. In addition to being held liable for $3,500,000 as the sum insured, they were ordered to pay general damages of $600,000 and $58,000 costs. Leggatt J. [1983] 1 Lloyd's Rep. 541, 557 held that the insurers could recover a proportion of the damages and costs from the re-insurers under an implied term of the re-insurance contract. But the Court of Appeal by a majority disagreed. Robert Goff L.J. [1985] 1 Lloyd's Rep. 312, 331-332 held that there was no basis for implying a term that the re-insurers should bear a proportion of the costs of defending the claim on the ground of business efficacy. Fox L.J. said at p. 334 that the contract worked effectively without any such implication. Moreover the consequences of implying such a term would be that the re-insurers' potential liability would be increased beyond, and possibly far beyond, the sum insured under the contract of re-insurance.

It was argued that the re-insurance policy was not a proportional contract. This may be the reason why Potter J. attached less importance to the decision in the Scor case than I would do. It is true that the insurers took a very high proportion of the risk. But as Millett L.J. pointed out in the Court of Appeal [1996] L.R.L.R. 353, 363, the extent of the cover is not reduced as the proportion taken by the insurers is increased:

So I would regard the Scor case as another authority in favour of the view that, if the syndicate is to succeed against Black Sea, it cannot be by virtue of a term implied by law. Indeed as Millett L.J. again pointed out, the present case is a fortiori. In the Scor case there was a provision which conferred on re-insurers the right to withhold their approval of a settlement. It was the effect of this provision in particular which made it necessary, in Leggatt J.'s view, and in the minority view of Stephenson L.J. in the Court of Appeal, to apply a term in favour of the insurers. But there is no such provision in the present case.

The cases principally relied on by Mr. Boyd were Uzielli & Co. v. Boston Marine Insurance Co. (1884) 15 Q.B.D. 11 and British Dominions General Insurance Co. Ltd. v. Duder [1915] 2 K.B. 394. I say nothing about Uzielli's case, since, like others before me, I find it difficult to understand. It should not be regarded as authority for any principle relevant to the present case. As for Duder, the question was whether the insurers, having settled the owners' claim for a constructive total loss at 66 per cent. of the sum insured could recover 100 per cent. from their re-insurers. Bailhache J. held that they could, with the result that the insurers would have made a profit out of the re-insurance. The Court of Appeal disagreed. The contract of re-insurance is a contract of indemnity. Accordingly the insurers could not recover more than they had lost. At the end of his judgment Buckley L.J. said at p. 403:

Mr. Boyd naturally relies on the reference to the costs being recoverable. There is a similar reference in the judgment of Pickford L.J. at p. 408. But it may be that the costs which Buckley L.J. had in mind are the costs which the owners had incurred: see the statement of facts at p. 395 where there is a reference to the insurers having settled the action "on the terms of paying 66 per cent. of the claim and the owners' costs." Or it may be, as suggested in MacGillivray at p. 395, that the point was conceded. In any event there was no suggestion that the costs could be recovered under an implied term of the re-insurance contract. So the case is not authority for any such general proposition. If it had been, it would surely have been cited ten years later in Scottish Metropolitan Assurance Co. Ltd. v. Groom, since Mr. R.A. Wright was counsel in both cases, and in both cases Bailhache J. was the judge at first instance.

Although the authorities are not perhaps compelling, the weight of the authorities is certainly consistent with the view that the syndicate cannot succeed in these proceedings by virtue of a term implied by law. On this point therefore I find myself in complete agreement with the judge and the Court of Appeal.

The starting-point here must be the passage which I have already quoted from Lord Wilberforce's speech in Liverpool City Council v. Irwin. As he pointed out, it is very common in mercantile contracts, where there is an established market usage, to add a term to an otherwise complete bilateral contract, on the grounds that it is what the parties would unhesitatingly have agreed. This was indeed the fabric out of which much of the modern commercial law was created in the 18th and early 19th centuries. Unfortunately the evidence of usage in the present case was left in a somewhat unsatisfactory state, owing to the fact that there were many other issues to be covered at the trial, and it was only at a late stage that the points of claim were amended so as to allege an implied term "in accordance with what is customary in proportional re-insurance."

Evidence for the syndicate was given by Mr. D.W. Jordan. He has been in underwriting all his working life. In paragraph 65 of his report dated 6 October 1992 (which stood as his evidence in chief) Mr. Jordan says:

Evidence was also given by Mr. Dodds, the syndicate's claims manager. He was asked what the practice is in relation to surplus re-insurances. He replied:

Neither Mr. Jordan nor Mr. Dodds were cross-examined on these points. It is fair to add that the pleading had not been formally amended at that stage so as to allege a trade practice. But as Mr Glennie very fairly conceded, he could have resisted the amendment, or at least applied for Mr. Jordan and Mr. Dodds to be recalled for further cross-examination.

Mr. Fryer who gave expert evidence for Black Sea, said in chief that it was "common for defence costs to be apportioned proportionate to the interests of the parties involved", but he added that this was not universal. In cross-examination he drew a distinction between proportional insurance on the one hand, and excess of loss and other non-proportional insurance on the other. In the former he would normally expect the re-insurers to pay their share of defence costs, but not in the latter. It was left unclear whether Mr. Fryer was dealing only with cases where the parties had made an express provision one way or the other. What was needed was evidence as to the practice where there is no express provision.

Potter J. in the course of his judgment correctly summarised the effect of the evidence: [1995] L.R.L.R. 261, 281. But when he came to state his conclusion at p. 282 he unfortunately fell into error. In the first place he said that, despite various amendments made by the parties in the course of the proceedings, no such custom had ever been pleaded by the syndicate. This must have been an oversight, for he had himself given leave for the amendment. Secondly, he said that the evidence did not go beyond an assertion that re-insurers often or usually pay a proportion of the insurer's expenses. But with respect Mr. Jordan's evidence goes further than this.

These errors serve to undermine Potter J.'s conclusion that the evidence was insufficient to establish a practice "so universally recognised and followed" as to lead to the implication of a term in the contract.

The test applied by the Court of Appeal was, if I may respectfully say so, entirely correct. Where I differ from Millett L.J. is in his view of the evidence. On its face, Mr. Jordan's evidence does go beyond establishing what is usual. Mr. Jordan says, in so many words, that it is "customary" for the expense of defending claims to be shared on a pro rata basis. He may have been using the word "customary" in a loose sense. But as there was no cross-examination we cannot tell.

It is at this point that Mr. Boyd's application to admit fresh evidence becomes relevant. The House did not find it necessary to look at the fresh evidence. Your Lordships were told that it is directly relevant to the point at issue. What should be done? One possible course would be to allow the concurrent findings of Potter J. and the Court of Appeal to stand in the present proceedings, and to leave it to Mr. Boyd to establish a trade practice by fresh evidence in other proceedings. For the failure to establish a trade practice in one set of proceedings does not prejudice other parties in subsequent proceedings: see Beckhuson and Gibbs v. Hamblet [1901] 2 KB 73. I would not for my part regard that as a satisfactory solution for two reasons. In the first place I would not be happy to leave the adverse findings of Potter J. and the Court of Appeal to stand when both courts seem to have mistaken the effect of Mr. Jordan's evidence. But I would be equally unhappy to substitute a contrary finding in favour of the syndicate on such scanty material. Mr. Glennie himself shared this concern.

Secondly, there is the overriding importance of establishing on adequate evidence for the market as a whole whether the alleged trade practice or usage exists or not. The sooner this is done the better. For these reasons the sensible course would seem to be to admit the fresh evidence in the current proceedings. The exceptional nature of Mr. Boyd's intervention on behalf of Equitas justifies, in my view, any breach of the rules in Ladd v. Marshall [1954] 1 WLR 1489. Although Mr. Glennie's formal position was that the rules in Ladd v. Marshall were not satisfied, he did not seek to press the point.

Then comes the question how the evidence should be received. Mr. Boyd said there were two choices. Either the House could direct the evidence to be taken before an examiner. The House could then itself determine on the basis of the evidence whether or not the syndicate had succeeded in proving the existence of the trade practice or usage. This would have the advantage of finality. The alternative would be to allow the syndicate's appeal, set aside the judgments below, so far as they relate to trade practice or usage, and remit the case to the Commercial Court for further argument on that issue, after hearing the fresh evidence, and any fresh evidence which Black Sea may wish to introduce.

Of these alternatives I much prefer the second. Difficult questions may arise in the course of hearing the fresh evidence. For example, it may be necessary for the parties to amend their pleadings. Or there may be questions as to the admissibility of some of the evidence. It is important that these questions should be the subject of an authoritative ruling. This could be given by the commercial judge, but not by an examiner.

But quite apart from the handling of the evidence, it is desirable, in a case of this importance to the market as a whole, to have the views of the commercial judge himself. In that connection counsel may find it convenient to cite a recent article by Mr. William Hoffman On the use and abuse of custom and usage in re-insurance contracts 1998 L.M.C.L.Q. page 43. It is true that a judgment on the central question at first instance could be appealed to the Court of Appeal and thence back to the House. But it is to be hoped that after a full hearing the parties to these proceedings, and the market as a whole, would accept the judgment of the commercial judge. If your Lordships agree, therefore, I would allow the appeal to the extent indicated, and remit all questions relating to trade practice and usage for a further hearing in the Commercial Court. The costs before your Lordships and in the Court of Appeal on the remitted issue should await the outcome of that hearing.

United Kingdom House of Lords Decisions

Baker v. Black Sea & Baltic General Insurance Company Ltd [1998] UKHL 18; [1998] 2 All ER 833 (20th May, 1998)