BERRIMAN AND OTHERS v. ROSE THOMSON YOUNG (UNDERWRITING) LTD.

QUEEN’s BENCH DIVISION (COMMERCIAL COURT)

[1996] L.R.L.R. 426

Nov. 14, 15, 16, 20, 21, 27, 28, 29; Dec. 4, 5, 7, 11, 12, 13, 14, 18, 19, 20, 1995; Jan. 17, 18; Mar. 19, 1996

Before Mr. Justice Morison

Reinsurance – Lloyd’s litigation – Negligent underwriting – Duty of care – Names incurred losses – Whether underwriters and agents in breach for failing to underwrite business with skill and care – Whether losses sustained by Names due to underwriters’ negligence – Whether Names estopped from asserting that agents liable – Whether limitation defences applied to amended points of claim – Quantum.

The first defendants (RTY) were the managing agents of marine syndicates 255/258 and employed Mr. N. T. W. Bullen as the syndicate’s active underwriter and Mr. Green as his deputy. RTY were also a members’ agency.

There were 1092 plaintiff Names and each of them either participated in the syndicate directly through RTY or indirectly through a members’ agency other than RTY which in turn appointed RTY as its sub-agent. There were 42 such members’ agencies which had been joined as defendants by the indirect Names.

The stamp capacity of a syndicate was the total of the sums allocated to it by the participating Names for a given year of account. From time to time an underwriter might exceed his stamp capacity and an underwriter had to be careful when writing business to allow for the possibility of reinstatements i.e. a reinstatement of the cover on payment of another premium.

In 1988 and 1989 there was a substantial increase in the number of participating Names and stamp capacity which were attributable in large measure to RTY’s decision actively to recruit new Names and thus attract a greater capacity for more business.

In 1988 and 1989 the syndicate suffered large losses, which were attributable to an accumulation of claims arising out of major catastrophes which were retained net for the syndicate’s account as a result of the way that Mr. Bullen wrote his book of business. A large part of Mr. Bullen’s book of business was catastrophe reinsurance written in the LMX market. The principal business written by Mr. Bullen were XL on XL contracts and whole account protections where the account being protected might itself contain a large proportion of XL on XL business.

The plaintiff’s case was that RTY failed to act with reasonable care and skill in the conduct and management of the underwriting for the two relevant years of account and that RTY were thereby in breach of duty and liable for the resulting losses. The plaintiffs alleged that:

(1) Mr. Bullen failed to write his book of business competently in both years of account in that he assumed greatly excessive aggregate liabilities and took out little vertical insurance. (The main issue.)

(2) Mr. Bullen was negligent in accepting INX Re as part of the security of the special priority treaty (SPT) (a form of proportional reinsurance) and in relying on the security of Aneco and Korea Foreign to the extent that he did (the SPT issue).

(3) Mr. Bullen negligently wrote personal stop loss (PSL) reinsurance which for the 1988 year of account involved the acceptance of aggregate liabilities of some 5,450,000 [or 19 per cent. of stamp] which were retained net for the syndicate’s account in 1988 and for the 1989 year 10,450,000 aggregate liabilities [or 30 per cent. of stamp] retained net. It was argued that a reasonably competent underwriter would have recognized the potential for aggregation between the syndicate’s LMX exposure and its PSL exposure and would not have retained net for his account any or all of the PSL aggregate exposure (the PSL issue).

(4) In relation to the 1989 year of account, nine layers of his outwards whole account protections between Sept. 1, 1988 to Jan. 1, 1989 were subject to one reinstatement only, whereas Mr. Bullen was writing reinsurance protection for other syndicates and companies [inward business] on the basis of two reinstatements. The allegation was that no competent underwriter would have written business which did not have matching reinstatements and would either have obtained two reinstatements on his outward protection or limited the inwards business to one reinstatement (the “matching reinstatements” point).

(5) There were issues as to causation and the way damages should be assessed. (The causation issue.)

(6) In relation to certain specific working Names who did not rely on their members’ agents to advise them to participate in the syndicate for the relevant years, whether the Names were estopped from asserting that the members’ agents were liable because each impliedly represented that he knew enough about the syndicate to be able to make an informed judgment that the syndicate was an appropriate part of his portfolio (the estoppel point).

(7) The matching reinstatement point and part of the allegations in relation to the SPT were added to the points of claim by way of amendment. It was argued that a limitation defence was available in relation to the amendments in that such amendments fell within the provisions of O. 20, r. 5 being claims arising out of the same facts or substantially the same facts as a cause of action in respect of which relief had already been claimed in the action by the party applying for leave to make the amendment.

(8) There were issues as to quantum.

-Held, by Q.B. (Com. Ct.) (Morison, J.), that as to A.

The main issue: (1) any underwriter who specialized in writing business in the LMX market, such as Mr. Bullen ought to have recognized the danger of the spiral effect; having elected to specialize in this market Mr. Bullen was obliged to give the most careful thought to the implications of the potentiality of the spiral upon his book of business (see p. 437, col. 2; p. 438, col. 1);

(2) it was the duty of a competent underwriter who wrote a significant amount of business in the LMX market to calculate his probable maximum loss (PML) in the event of the worst catastrophe which the underwriter could foresee as a practical possibility and then make a judgment as to the amount of vertical reinsurance he required and the amount of exposure he proposed to run net for the account of his Names (see p. 438, cols. 1 and 2);

(3) it was common ground that every competent underwriter would need to plan his account so that he could exercise the judgments necessary to protect his Names (see p. 440, col. 2; p. 442, col. 2);

(4) Mr. Bullen’s evidence about the way he wrote his book of business would be rejected; it was not accepted that he planned his book of business or that he used a PML calculation to determine the amount of vertical reinsurance protection he should buy or that he made any informed judgment as to the net exposure to which he should properly expose his Names (see p. 445, col. 1; p. 446, col. 1; p. 447, cols. 1 and 2; p. 448, col. 1; p. 451, col. 1; p. 452, col. 1);

(5) Mr. Bullen did his incompetent best but fell well below the standards to be expected of any underwriter who specialized in this market; if Mr. Bullen had been more responsibly managed there was little doubt that he would have acted more competently; he was as incompetent as he was allowed to be by RTY and RTY were not only liable vicariously for the acts and defaults of Mr. Bullen but they were also in breach of duty as managing agents (see p. 454, col. 2; p. 455, col. 1).

B. The special priority treaty: (1) it was the duty of a broker to satisfy himself that the security put before an underwriter was reasonably appropriate to the type of business underwritten but the broker would not be concerned with the finer details of such a book of business; and Mr. Bullen himself (and RTY) was under a duty to satisfy himself that the security was appropriate for the purposes he had in mind (see p. 459, col. 2; p. 460, cols. 1 and 2; p. 461, col. 1);

(2) on the facts and the evidence no competent underwriter could have concluded that INX Re was a reliable enough security for the SPT; Mr. Bullen had no information on which to make any kind of sensible judgment and he was never shown any figures; and no competent underwriter could have accepted INX Re as a participant in an important treaty which to a significant extent governed Mr. Bullen’s underwriting strategy for 1989 (see p. 462, cols. 1 and 2);

(3) although a broker had greater expertise than an underwriter any reasonably competent underwriter would have made a simple inquiry to check whether the ISI rating which he saw gave him the comfort he was seeking; Aneco was expressly not being warranted by the brokers as suitable security and it was for Mr. Bullen to have made some limited inquiries to establish whether Aneco was the medium sized company he was sensibly looking for having regard to the amount and nature of the aggregate which he intended to cede to that company (see p. 462, col. 2; p. 464, cols. 1 and 2; p. 466, col. 1);

(4) although Mr. Bullen was properly entitled to take the view that Aneco was a reasonably respectable participant in the treaty be ought not to have regarded them as a medium sized company which was suitable for taking a 30 per cent. share in the treaty; the amount of aggregate ceded to it was greatly in excess of what any competent underwriter could properly have ceded to a company of that size and financial standing; and any competent underwriter would not have ceded to Aneco, having regard to its size and financial standing a PML exposure of more that 10 per cent. of the syndicate’s stamp capacity (see p. 464, col. 2; p. 466, col. 1);

(5) Korea Foreign was a state insurance company of North Korea; its assets were controlled by the government of North Korea and it depended for its continued existence on the willingness of that government to keep it in funds; the company had an ISI rating of A and the downside of using the company was the risk that the government would cease to support it; on the basis of that information Mr. Bullen did not fall below the standards expected of him in accepting this company as suitable security (see p. 466, cols. 1 and 2).

C. The PSL issue: it would not be held that Mr. Bullen ought not to have written his PSL book of business; it was risky business; there was a risk of aggregation but that risk was quite unquantifiable; the control which the underwriter exercised was to limit the amount of business he wrote in terms of premium income and to write a spread of business so as to minimize the risks; there was a limited number, only, of syndicates who were significant players in the LMX market at higher levels of catastrophe reinsurances; and there was no sound reason for believing that merely because the sector made losses all the other syndicates would be adversely affected in the same way; it might well be that Mr. Bullen was not as perceptive as he should have been but he could not be said to have fallen below the standards to be expected of every competent underwriter (see p. 468, col. 2).

D. Matching reinstatements: an underwriter was under a duty to match his reinstatements where that could be done; Mr. Bullen could not obtain more than one reinstatement on his outward protections after Piper Alpha and from a practical point of view he could not refuse to accept two reinstatements on his inward business if the market was generally accepting it; he was therefore faced with a choice of either not writing some business at all or running the risk that there would be no more than two major catastrophes which would impact his inward business; it could not be concluded that every competent underwriter would have elected to limit his inward business in the way suggested (see p. 469, cols. 1 and 2).

E. Causation: (1) the writing of insurance was the deliberate acceptance of a risk, the risk that an event giving rise to a claim would materialize; the 1988 and 1989 catastrophes were not unexpected in any sense relevant to this litigation and if the plaintiffs could not contend that this was a no- transaction case they had to accept that they were susceptible to the loss even if Mr. Bullen had purchased adequate reinsurance protection in accordance with his duties (see p. 470, col. 2; p. 472, col. 1);

(2) the plaintiffs were entitled to be put in the same position they would have been in if a competent underwriter had competently chosen a PML event, competently calculated the PML exposure, competently decided how much of that exposure to retain net for the Names’ account and then purchased vertical reinsurance for the balance of the exposure in relation to the book of business which was actually written by Mr. Bullen; the loss which the Names would have suffered on that hypothesis would be set against the losses which they had suffered (see p. 474, col. 1);

(3) for a marine syndicate such as this it would have been sensible to take as the PML event what Mr. Bullen said he had in mind i.e. a collision between a rig (or platform) and a tanker; and there was nothing wrong with taking a figure of between 3 billion-3.5 billion; the PML amount was the amount of loss which a competent underwriter would reasonably have considered his book of business to be exposed to in the event of the occurrence of the PML event (see p. 474, col. 2);

(4) it was clear from the expert evidence that every competent underwriter would have assumed his exposure on his XL on XL account written in the LMX market to be wholly exposed to a PML event (see p. 475, col. 1);

(5) while there might be various different ways of seeking to measure the syndicate’s exposure to the PML event, no competent underwriter could have concluded that his whole account covers were other than exposed 80 per cent.; in principle XL on XL business should be treated as 100 per cent. exposed whether it was reinsured as XL on XL or was part of the cedant’s whole account; the assessment of the PML amount was not an actuarial or purely mathematical exercise; it involved judgment and feel and the decision had to be taken on the basis of all the information and experience available to the underwriter; and the assessment had to be on the whole class of business and not on a risk by risk basis (see p. 475, col. 2; p. 476, col. 1);

(6) for each of the two years an exposure of 80 per cent. in relation to the liability account and 30 per cent. for the hull XL account would be assumed (see p. 476, col. 1);

(7) it was wrong to assess a PML amount by reference to the impact of an actual loss after the event; and it could not be assumed that the impact of any catastrophe on the book would bear any proportionate relationship with the impact of the PML event on the book; how a catastrophe of any particular size would impact the book was a matter of pure speculation (see p. 476, col. 2);

(8) what a Name could reasonably anticipate depended on what he knew or ought to have known about the syndicate’s business; proper advice would reflect the market’s appreciation of the general nature of the syndicate’s business and of the underwriter’s reputation and what was published by the managing agency about the syndicate including its results and underwriter’s report (see p. 478, col. 1);

(9) on the expert evidence, while it was the duty of the members’ agencies to advise the Names about the choice of syndicates, they were not required so to question the underwriter of any particular syndicate so as to endeavour to second guess the way the risks were written; further it would have been unlikely before late 1989 that any members’ agents would have asked specific questions about the underwriters PML or whether there was any deliberate exposure (see p. 482, col. 1);

(10) an underwriter for this syndicate who was acting competently could not have assumed that a Name had accepted that he would be deliberately exposed to a loss of greater than 40 per cent. of stamp (see p. 483, col. 1; p. 484, col. 1);

(11) based on the pleaded case the plaintiffs were entitled to be compensated on the basis that had INX Re been a sound participant the syndicate would have been able to recover from it such sums as were due; in relation to Aneco they were entitled to recover such sums as were due but unpaid in relation to claims in respect of that aggregate exposure ceded beyond the acceptable limit; this was not a no-transaction case; and the losses which flowed from Korea Foreign’s inability or unwillingness to pay were not due to Mr. Bullen’s incompetence but were due to the risks in competently choosing any security; those losses had nothing to do with his incompetent choice of INX Re or of Aneco (see p. 485, col. 2).

F. Estoppel: (1) there was nothing inconsistent, or unfair or unjust in a Name choosing to go on a syndicate and thereafter enforcing his rights against the members’ agent under the contract which he must make with the agent; under the contract the agent warranted that the syndicate’s business would be written competently and carefully; by choosing which syndicate he would join the Name might absolve the agent from his responsibility to advise him about his portfolio selection but there was no warrant for the idea that the Name also represented that he would not hold the agent responsible for the underwriters’ incompetence (see p. 486, col. 1);

(2) the facts did not establish either that the Names consented to the lack of reasonable care or represented that they knew of the matters giving rise to claims against RTY and the agencies; the agents were liable under the contractual arrangements whether or not they could be blamed for not discovering that a syndicate’s business was being badly managed (see p. 486, col. 1).

G. Limitation: there was a real overlap between the amended allegations and the original allegations in relation to SPT and that what was new arose out of the same or substantially the same facts as were already in issue; the same point arose in relation to the matching reinstatement issue; no limitation defence was available in relation to the amendments to the points of claim (see p. 486, col. 2).

H. Quantum: (1) the plaintiffs had to give credit for the cost of any additional insurance Mr. Bullen might have purchased (see p. 487, col. 1);

(2) the plaintiffs did not have to give credit for the benefit of any reinsurance which ought not to have been written in favour of other syndicates on which any of the plaintiffs participated for the years of account in question; there was no link in terms of causation or foreseeability between Mr. Bullen’s negligence and the fortunes of 255’s Names on other syndicates (see p. 487, col. 1);

(3) the plaintiffs did not have to give credit for the benefit of any tax refunds or allowances in respect of losses suffered by individual Names as a result of their participation on Syndicate 255 for the relevant years of account (see p. 487, col. 2);

(4) the plaintiffs did not have to give credit for the benefits of any stop loss policy insurance they might have received for the relevant years of account (see p. 487, col. 2);

(5) the additional cost of reinsurance should fill the gap above the existing cover from the bottom upwards (see p. 487, col. 2).

The following cases were referred to in the judgment:

Arbuthnott and Others v. Feltrim Underwriting Agencies Ltd., Jan. 15, 1996 unreported;

Arbuthnott v. Feltrim Underwriting Agencies Ltd., Mar. 10, 1995;

Banque Bruxelles Lambert S.A. v. Eagle Star Insurance Co. Ltd., (C.A.) [1995] L.R.L.R. 195; [1995] 2 All E.R. 769;

Brown v. KMR Services Ltd. and Sword Daniels v. Pitel and Others, (C.A.) [1995] 2 Lloyd’s Rep. 513;

Deeny v. Gooda Walker Ltd., [1996] L.R.L.R. 183;

Eckersley v. Binnie [1988] 18 Con.L.R. 38;

Henderson v. Merrett Syndicates, (H.L.) [1994] 2 Lloyd’s Rep. 468; [1995] 2 A.C. 145;

Jebsen v. East and West India Dock Co., (1875) L.R. 10 C.P. 300;

Livingstone v. Rawyards Coal Co., (H.L.) (1880) 5 App.Cas. 25;

McElroy Milne v. Commercial Electronics Ltd., [1993] 1 N.Z.L.R. 39;

Robinson v. Harman (1848) 1 Exch. 850;

Yorkshire Dale Steamship Co. Ltd. v. Ministry of War Transport, (H.L.) [1942] A.C. 691.

This was an action by the Names represented by Sir David Berriman to recover damages from their managing agents and members’ agents, Rose Thomson Young (Underwriting) Ltd. in respect of substantial losses which they had incurred as underwriting members of Lloyd’s in the 1988 and 1989 years of account.

Mr. Michael Crane, Q.C., Mr. Stephen Moriarty and Mr. Matthew Reeve (instructed by Messrs. Richards Butler) for the Names; Mr. Jonathan Hirst, Q.C., Mr. Andrew Popplewell and Mr. Harry Matovu (instructed by Messrs. Reynolds Porter Chamberlain) for the managing agents; Mr. Richard Slade (instructed by Messrs. Cameron Markby Hewitt) for some of the members’ agents.

The further facts are stated in the judgment of Mr. Justice Morison.

Judgment was reserved.

Tuesday Mar. 19, 1996

JUDGMENT

Mr. Justice MORISON:

PRELIMINARY

This is yet another Lloyd’s case in which Names seek to recover damages from their managing agents and members’ agents in respect of substantial losses which they have incurred as underwriting members of the Society of Lloyd’s in the 1988 and 1989 years of account.

To be admitted to membership of Lloyd’s at the relevant time and become a Name, a person was required to demonstrate a certain minimum amount of wealth. The amount of premium income which a Name was permitted to write in any year was calculated by multiplying his proven wealth by 2.5. A Name could only become a member of a syndicate through the agency of a members’ agent or a combined agent. In consultation with, and usually upon the advice of such agent, a Name’s permitted premium income would be allocated between a number of syndicates [“a portfolio”].

At the relevant time, the first defendants, Rose Thomson Young (Underwriting) Ltd. [“RTY”] were the managing agents of marine Syndicate 255/258 [“the syndicate”] and employed Mr. N. T. W. Bullen as the syndicate’s active underwriter, and Mr. Green as his deputy. RTY were also a members’ agency. There are 1092 plaintiffs. Each of them is a Name, or the personal representative of a Name, who either participated in the syndicate directly through RTY or, indirectly, through a members’ agency other than RTY which, in turn, appointed RTY as its sub-agent. There are 42 such members’ agencies which have been joined as defendants by the indirect Names.

It is the plaintiffs’ case that RTY failed to act with reasonable care and skill in the conduct and management of the underwriting for the two relevant years of account, and that the defendants are, thereby, in breach of duty and liable for the resulting losses.

The duties owed to direct and indirect Names by a combined agency such as RTY, and by a members’ agency, in relation to the manner in which a syndicate’s underwriting is carried out and managed have been fully analysed by the House of Lords in Henderson v. Merrett, [1994] 2 Lloyd’s Rep. 468; [1995] 2 A.C. 145. In these proceedings it is common ground that, if the plaintiffs have made good their contentions, RTY are liable in both contract and tort to the direct Names; and the members’ agents named as defendants are liable in contract to the indirect Names, to whom RTY are also liable in tort.

On July 4, 1995, this Court ordered that the issues for determination at the trial were to be limited to: (1) issues of liability; (2) such issues of principle relating to quantum as may hereafter be agreed or ordered.

Apart from an order which I made on Nov. 1, 1995, there has been no agreement or other order made in relation to quantum. The ambit of par. (2) has caused some minor controversy, to which it will be necessary to revert in due course.

The trial commenced on Nov. 14, 1995; the evidence began on Nov. 27, 1995, and finished on Dec. 20, 1995; closing submissions were made orally on Jan. 17 and 18, 1996, following the delivery of full and helpful written submissions. At the outset of this judgment I would wish to record my thanks to Counsel and solicitors for their assistance. In many ways the manner in which this case has been conducted by the lawyers is a model of its kind. Counsel have made an effective and efficient use of Court time; witnesses have been treated courteously, even when rigorously, but properly, cross-examined; the solicitors have presented the documents in an orderly and convenient manner. During the writing of this judgment it has become increasingly clear that I cannot do full justice to all of Counsels’ arguments without creating an [some might say “a more”] unacceptably long document. I can say, however, in case it may be thought otherwise, that I have read and re-read all their written submissions with great care.

THE SYNDICATE AND ITS BUSINESS

The stamp capacity of a syndicate is the total of the sums allocated to it by the participating Names for a given year of account. The syndicate’s underwriter will need to know the stamp capacity because that controls the amount of business which he can write. From time to time an underwriter may exceed his stamp capacity as Mr. Bullen did in the 1983 and 1984 years of account, but this was regarded with disfavour by Lloyd's. An underwriter had to be careful when writing business to allow for the possibility of “reinstatements”, that is a reinstatement of the cover on payment of another premium. That was what caused Mr. Bullen to exceed his limit.

The syndicate was formed in 1980. Mr. Bullen succeeded a Mr. Davies as the syndicate’s active underwriter in November, 1981. 1991 was the last year of account. Each year of account from [and including] 1988 has been left open, and the syndicate is in “run-off”. “Run-off” managers [Syndicate Underwriting Management Ltd., “SUM”] were appointed as from Dec. 9, 1994, two days after RTY went into liquidation. The following table sets out the number of Names on the syndicate and the syndicate’s stamp capacity during its 12 years of business. It is to be noted that there was a substantial increase in stamp capacity from 1987 to 1988 [53 per cent.]. This was partly attributable to new rules which came into force in 1987 and which permitted “grossing-up”; [In the table, the 1987 capacity is expressed net; when grossed up, the figure would be approximately 23 m.] but at the same time the number of participating Names increased by 23 per cent. followed by a slightly larger increase for the following year [30 per cent.]. These increases in Names and stamp capacity were attributable in large measure to RTY’s decision actively to recruit new Names and thus attract a greater capacity for more business, and reflected the general market trend towards such an increase.

YEAR STAMP CAPACITY - 000’s NO. OF NAMES 1980 625 30 1981 745 36 1982 1,002 48 1983 2,035 90 1984 4,250 175 1985 6,930 290 1986 9,004 385 1987 18,160 808 1988 27,849 996 1989 34,143 1286 1990 13,058 545 1991 8,060 347

For reasons which lie at the heart of the case on liability, the syndicate suffered huge losses for the two relevant years of account. The extent of the syndicate’s liabilities is not yet finally known, as there are still claims being made and the final amount of the provisions for bad debts in relation to the syndicate’s reinsurance programme has yet to be calculated. These matters will have to be further investigated at a damages hearing, but for present purposes it is sufficient to record the fact that the losses are likely to lie in a range of 315.9 per cent. to 410.76 per cent. of stamp for 1988 and 434.9 per cent. and 704.92 per cent. of stamp for 1989. As a matter of simple arithmetic, the average amounts of premium allocated by each Name to the syndicate for the 1988 and 1989 years of account were, approximately, 28,000 and 26,500, respectively. Several Names had committed to the syndicate a line of 40,000 and will have lost between 126,000 and 164,000 in respect of the 1988 year; and between 174,000 and 282,000 in respect of the following year.

Such large losses are attributable to an accumulation of claims arising out of major catastrophes which were retained net for the syndicate’s account as a result of the way that Mr. Bullen wrote his book of business. The Names allege that Mr. Bullen assumed greatly excessive aggregate liabilities, and took out far too little reinsurance. This is the same allegation as was made in the Gooda Walker and Feltrim cases [see the speech of Lord Goff in Merrett at p. 483, cols. 1 and 2; p. 171G-H]. As in those cases, a large part of Mr. Bullen’s book of business was catastrophe reinsurance written in the LMX market.

One of the reasons why this case has taken relatively little time to try is that, to a significant extent, it is a re-run of what has been tried before. I have the benefit of an analysis by Mr. Justice Phillips of the characteristics of that market, and of the duties of an underwriter who participates in it. Both parties in this case have tended to rely on those parts of the two judgments which they say support their case, urging me to the view that consistency between two Judges of the Commercial Court is very desirable if not essential. When making that submission reliance was placed by the defendants on the following passage in the judgment of Mr. Justice Longmore given on Monday Jan. 15, 1996 in Arbuthnott and Others v. Feltrim Underwriting Agencies Ltd. and Others:

I do not consider that Phillips J. was wrong; even if I did, nothing could be more counter-productive in the Lloyd’s litigation as a whole if judges of co- ordinate jurisdiction were to take diverging views on the same or similar points.

However, neither of the two main parties felt so constrained by this approach as to be unable to suggest, when it suited them, that I should depart from the earlier judgments on the grounds that this Court had received different evidence from that adduced in the other two cases. In fact, it would appear that much of what Mr. Justice Phillips held to be the characteristics of the LMX market and the duties of an underwriter who participated in it is not in dispute in this case. There are, however, issues relating to causation and to the proper approach to damages where the evidence may be different and where I have attempted to tread with particular care.

Mr. Bullen specialized in the writing of excess of loss reinsurance of which the major proportion was LMX business. Excess of loss reinsurance [“XL”] indemnifies a reinsured against his ultimate net loss from a claim or series of claims arising out of one event once such claim(s) exceed(s) the deductible (or excess point), and up to the limit of indemnity fixed by the contract. Such covers were customarily placed in layers, one above the other, with the limit of indemnity on the lower layer being the same figure as the excess point of the layer which sits immediately above it. The lower layers may be liable to be impacted frequently by events which might be described as commonplace, such as windstorms. As Mr. Justice Phillips put it in Deeny v. Gooda Walker Ltd., [1996] L.R.L.R. 183 at p. 189, col. 1:

. . .In general, the higher the layer of reinsurance the more serious, and the less common, will be the event that impacts that cover. Low layers of cover are sometimes referred to as “working layers”. Excess of loss insurance of higher layers is commonly referred to as “catastrophe insurance” on the basis that the cover will only be likely to be impacted in consequence of a catastrophe, whether caused by man or by nature . . .

A company which writes direct business runs the risk that its book of business will be impacted by a single catastrophic event and will obtain reinsurance to protect itself against that. The reinsurer who writes that business is known as the primary reinsurer; he, in turn, will wish to lay off the risk he has undertaken by retroceding all or part of it to another retrocessionaire (or reinsurer) who, in turn, may wish to reinsure, and so on. As Lord Justice Hobhouse observed in Brown v. KMR Services Ltd. and Sword Daniels v. Pitel and Others, [1995] 2 Lloyd’s Rep. 513 at p. 545, col. 2:

. . .the market includes a wide range of indirect or reinsurance business. Much of this is of extreme technicality and complexity and may be written in a large number of different ways. These include the straight reinsurance of part of the original risk, the reinsurance of the risk in excess of a certain level, the cession by one underwriter to another of a proportion of his business, the cession of losses on a whole account over a certain level. There are many variants and elaborations of these types of reinsurance business. . .

The principal business written by Mr. Bullen were XL on XL contracts and whole account protections, where the account being protected might itself contain a large proportion of XL on XL business.

The LMX market, in which Mr. Bullen largely did his business, enabled London companies and Lloyd’s syndicates to retrocede all or part of their own reinsurance accounts to other participants, who themselves reinsured their own accounts, whether on an XL on XL basis or a whole account basis, within the market. Thus the participants in that market increasingly transferred catastrophe exposure to each other. This is well illustrated by the facts in this case. RTY, in appendix A to their further and better particulars served in March, 1995, identified 48 Lloyd’s syndicates, including Syndicate 255, as writing more than 50 per cent. of their business as XL, or whose book of business was predominantly retrocessional LMX, and 36 companies whose business had one or other of those characteristics. Upon analysis, it is clear that Syndicate 255 participated in the whole account reinsurance cover of 42 of those syndicates and of 32 of those companies in the 1988 year of account. Furthermore, in relation to the years of account 1988 and 1989 respectively, most of the syndicates [89 per cent. and 85 per cent.] and companies [84 per cent. and 86 per cent.] participated in Syndicate 255’s reinsurance programme. This mutual reinsurance of one syndicate by another and then upwards into higher layers, and round again, produces the spiral effect which is fully analysed in Mr. Justice Phillips’ two judgments.

Mr. Bullen’s book of business 1988

Using the figures from the aggregates book, which was kept at the box, the relevant LMX figures for 1988 and 1989 are set out on p. 433 post.

In total, Mr. Bullen wrote 5332 contracts [giving rise to aggregates of U.S.$325,553,000 at rates on line of less than 5 per cent. i.e. in the 0-4 per cent. bracket] with a total aggregate exposure of U.S.$434,491,000. In relation to whole account and XL on XL contracts in the London market, the total aggregate exposure was U.S.$327,732,000, of which just under $254 m. was attributable to contracts written at rates on line of less than 5 per cent. For the 1988 year of account, the syndicate initially enjoyed the benefit of vertical reinsurance cover [the precise figures are set out in Mr. Thomson’s report at pars. 5.9-5.13] in the sum of U.S.$85 m.; after Piper Alpha, in July of that year, Mr.Bullen obtained further cover in the sum of U.S.$24m., giving him a total overall cover in relation to thelatter part of that year of U.S.$109 m., or about one-third of his total aggregate exposure on the LMXmarket.

It will be seen that the 1989 figures do not include information about the number of contracts written in each category. However, the total aggregates written during the year were U.S.$484,539,000 of which about 50 per cent. were written in the rate on line band of less than 5 per cent. and of those, U.S.$190,543,000 were written in the LMX market. Whole account and XL on XL aggregates in that rate on line accounted for U.S.$164,697,000, that is, about one third of the total aggregates written and 86 per cent. of the LMX aggregates.

The vertical reinsurance cover which Mr. Bullen arranged for the year 1989 was, in round terms [and the precise details are set out in pars. 5.14-5.17 of Mr. Thomson’s report], U.S.$125,000,000, excluding the protection afforded by what has been called the special priority treaty [“SPT”], which forms the subject matter of a separate head of complaint.

In the light of this summary, I turn to the principal allegations which the plaintiffs make in this case:

1. Mr. Bullen failed to write his book of business competently in both years of account, in that he assumed greatly excessive aggregate liabilities and took out far too little vertical reinsurance. This issue raises questions as what Mr. Bullen, did, and what he ought to have done as the underwriter of this book of business. I include within this general issue the allegations made against RTY, as managing agents. I shall call this “the main issue”.

2. Mr. Bullen was negligent in accepting INX Re as part of the security for the SPT, and in relying upon the security of Aneco and Korea Foreign to the extent that he did. This raises questions as to the duties of an underwriter when entering into such a contract, relative to the duties of the broker; and as to the reliability or otherwise of the security as it would have appeared to a reasonably competent underwriter in Mr. Bullen’s position at the relevant time. [“The SPT issue”.]

3. Mr. Bullen negligently wrote personal stop loss [“PSL”] reinsurance which, for the 1988 year of account involved the acceptance of aggregate liabilities of some 5,450,000 [or 19 per cent. of stamp] which were retained net for the syndicate’s account in 1988; and for the 1989 year, 10,450,000 aggregate liabilities [or 30 per cent. of stamp] retained net. It is said that a reasonably competent underwriter would have recognized the potential for aggregation between the syndicate’s

LMX exposure and its PSL exposure and would not have retained net for his account any or all of the PSL aggregate exposure. [“The PSL issue”.] These figures are the losses which were known at the date of the pleading. The position will have to be investigated at a further hearing.

4. In relation to the 1989 year of account, nine layers of his outwards whole account protections between Sept. 1, 1988 to Jan. 1, 1989 were subject to one reinstatement only; whereas he was writing reinsurance protection for other syndicates and companies [inwards business] on the basis of two reinstatements. The allegation is that no competent underwriter would have written business which did not have matching reinstatements, and would either have obtained two reinstatements on his outwards protection or limited the inwards business to one reinstatement. [This allegation was called, inargument, the “matching reinstatements” point.]

5. There are issues as to causation [“causation”], and a dispute as to the ambit of those issues. Iinclude within this point, an issue as to the way damages should be assessed.

6. In relation to certain specific working Names who did not rely upon their members’ agent to advise them to participate in the syndicate for the relevant years, there is an issue raised by the members’ agents as to whether the Names are estopped from asserting that those agents are liable because -

. . .each impliedly represented that he knew enough about the Syndicate to be able to make an informed judgment that the Syndicate is an appropriate part of his portfolio. [“The estoppel point”.]

7. The matching reinstatement point and part of the allegations in relation to the SPT were added to the points of claim by way of amendment. I am asked to decide [the “limitation issue”] whether such an amendment falls within the provisions of O. 20, r. 5: namely whether these new claims arise out of:

The same facts or substantially the same facts as a cause of action in respect of which relief has already been claimed in the action by the party applying for leave to make the amendment.

8. The issues of principle relating to quantum which I ordered to be tried [“quantum”].

I shall consider each of these matters in turn.

1. THE MAIN ISSUE

The plaintiffs’ pleaded case on this issue may be summarized thus:

LMX business has, inherently, the following characteristics:

1. In general terms, the longer the chain of reinsurance arrangements (that is, the further from the placing of the direct insurance) the less transparent the risk will be to the reinsurer, who cannot assess, with any degree of comfort, the accumulation potential of the business he is reinsuring.

2. There is a significant risk that the various portfolios of business reinsured by the LMX underwriter will be exposed to the same loss events, so that a single loss event or catastrophe will give rise to losses on several reinsurances which thereby aggregate together.

3. A loss above a particular size can circulate between the relatively limited number of reinsurers within the LMX market, apparently escalating in size as it spirals into higher and higher levels of the XL reinsurance programme of each reinsurer until one of the participants runs out of cover and the “parcel” stops with him. There may be “leakage” in the sense that part of a risk leaves the market and does not come back in; but there was no requirement for co-insurance and therefore no stipulation that any cedent had to retain some of the risk net. Thus risks were written and then passed on without there being any real increase in the capacity of the market. As the brokers placed the business so they retained a percentage of the premium. At the top of the spiral, at the highest layers, the premium rates were small, but as will be seen, the level of exposure was potentially great.

4. It is inherently impossible for a participant in the LMX market to be able to know what size of loss will trigger the spiral effect. This feature is largely a consequence of the paucity of the placing information which was proffered to a secondary or tertiary reinsurer.

5. The participant who is left with the “parcel” is the one who has less reinsurance cover relative to the cover which the other participants have arranged for themselves. The relative amount of reinsurance cover is not known to any one participant, and, therefore no participant can be confident that he will not run out of cover first.

6. Once a loss of the critical size occurs which exceeds the deductibles of the reassured in the chain, then it will penetrate into the upper layers of the covers until a participant has run out of cover. Thus, it is alleged in the amended points of claim, at par. 27(5)(a), that:

It is not necessarily the case that the probability of a loss event affecting a particular layer of excess of loss will decrease in proportion to the size of the excess to which the layer is subject.

7.

A general principle of insurance is the dispersal of risk. Thus the original risk written on a Lloyd’s slip is divided between a number of syndicates each composed of many Names. Reinsurance of that original risk normally further disperses the risk to other syndicates and other Names (and outside Lloyd’s to other insurers). The vice of the LMX class of business is that it tends to have the opposite effect. There are a limited number of LMX underwriters and writing LMX business tends to concentrate not disperse the risk, particularly at the higher levels. [per Lord Justice Hobhouse in Brown.]

The plaintiffs put it thus [par. 28 amended points of claim]:

. . .the overall effect. . .is to focus that loss on the few LMX reinsurers who have written excess of loss cover without having effected sufficient reinsurance in relation thereto

8. LMX contracts reinsuring marine accounts became more and more exposed to non-marine losses with the result that a non-marine loss could cause the spiral effect to occur in marine XL programmes as well as in non-marine XL programmes, which would thereby aggregate with one another.

The pleaded case for the defendants may be summarized as follows:

1. LMX insurance is a specialist business concerned in part with catastrophe losses and therefore involves a greater potential for profit in good years and a greater risk of losses in bad years than other forms of insurance. In that sense it can be described as high risk. One catastrophe can make the difference between substantial profit and loss.. . .The LMX market operated on principles of continuity and “pay back”. The market had to and did work on the basis that underwriters would be able to recoup in good years the losses resulting from catastrophe claims which had been suffered and which would inevitably be suffered in future years.

2. Because of the “pay-back” concept and the need for continuity inherent in the LMX market -

. . .the underwriting judgment of the active underwriter could properly be governed by a longer term view of profits and losses than a simple year by year approach.

3. On the basis of the information provided to him on placement of a risk, and upon his own knowledge and judgment of the practice and skill of the reinsured -

. . .a retrocessional underwriter could form a reasonable and properly informed assessment or judgement of his likely exposure in respect of the proposed risk.

4. It was generally accepted that risks written at lower layers were less prone to loss than those written at higher layers.

A reasonably prudent LMX underwriter would not have believed that anything short of an enormous catastrophe loss would rebound within the network so as to have a substantial impact on his top layers of cover.

This was confirmed by the syndicate’s own previous experience in relation to other catastrophes namely:

5. Those reinsurers who accepted risk at the highest layers of cover afforded by the market, were providing protection against catastrophic losses either in respect of an enormous single risk loss or in respect of an enormous aggregation of smaller losses caused by one event; the premiums were set across the market which reflected the perceived exposure to loss of such classes of reinsurance.

6. At all material times a Name must have appreciated that the conduct of underwriting at Lloyd's, as with all insurance, involved the balancing of risk and carried with it the risk of losses. Names on such a high risk syndicate should adopt a balanced portfolio of syndicates in which to participate; he should protect himself from losses by making appropriate provision or by taking out personal stop loss insurance [“PSL” insurance].

7. Although the Lloyd’s market was substantially governed by regulations set in order to maintain standards of underwriting practice -

. . .there were no regulations defining a degree ofexposure beyond which it was unacceptable toexpose Names nor any prescribed minimum levels of reinsurance.

It would be unrealistic, uncommercial and impossible in practice to reinsure the entire aggregate exposure of the syndicate. . .in question, so as to avoid carrying any risk at all.

Accordingly, the underwriter in question should make a realistic assessment of how a worst case loss would impact on his book; such a realistic worst case was sometimes referred to, more specifically in the non-marine market, as a “probable maximum loss” or “PML”. In the light of such an assessment the underwriter should exercise his judgment as to the amount of vertical reinsurance cover he should purchase taking into account those parts of his book where he might run what he perceived to be a small risk; the availability and cost of such reinsurance protection and the cost of the other part of any reinsurance programme, including horizontal cover against attritional losses.

8. The spiral phenomenon was recognized to exist to some degree but it was generally accepted “throughout the market” that its effect was mitigated by a number of factors, including the amount of the deductibles or retentions, exclusion clauses which prevented a loss from being passed from one underwriter to another, stipulations for co-insurance and “the possibility of exhaustion of the excess of loss outward reinsurance programme of any participant in the network of reinsurances through which a loss progressed.”

It seems to me important to note what this case is not about. It is not “a portfolio selection case” such as Brown or Sword-Daniels. In other words, it is not a case where the plaintiffs are accusing their members’ agencies of having negligently caused or permitted them to allocate any part of their premium income to the syndicate. I adopt the same approach as Mr. Justice Phillips in Gooda Walker:

Insofar as I have to consider how the nature of the business of the syndicates was, or should have been, perceived by the Names, there is no room for distinguishing between the knowledge of one Name and the next. The question must be judged having regard to the knowledge and information that would have been generally available to members’ agents and their Names and on the premise that the Names were receiving competent advice from their members’ agents.

Second, there is no allegation in this case that Mr. Bullen should not have engaged in the LMX market at all. There is much to be said in favour of such a contention, had it been made. It became quite apparent to me in the course of the evidence that the very nature of the way the market operated made it difficult for any underwriter to make soundly based judgments about the risks he was writing. It is a market which has, I believe, ceased to exist since 1991 because it was recognized to be an aberration. However, such a case would have had to take account of the fact that there were many syndicates and companies who participated in that market, some of whom appeared to do so quite successfully.

Third, this case is primarily concerned with an allegation that the Names were negligently over-exposed to risk, in the sense that Mr. Bullen took on excessive exposure without adequate vertical re-insurance. In other words, save only in relation to par. 36(2)(d) of the amended points of claim, the “matching reinstatements point”, this case is not concerned with the syndicate’s horizontal exposure. The real complaint is, as it was in Gooda Walker, that there was insufficient vertical protection and not that the book of business was excessively exposed to horizontal risks; that is, losses coming from a large number of events than he had catered for.

Within the framework of the pleaded cases, I shall consider, first, the nature of the market in which Mr. Bullen was writing his business, what he knew or ought to have known about it and what his duties were in relation to writing in that market. I shall then consider what Mr. Bullen did and identify the several respects in which he conspicuously failed to carry out his duties in a competent manner.

(1) The LMX market in 1988 and 1989

It will be seen that, to some extent, there was an overlap between the two pleaded cases. Having heard the evidence, it became clear, as one would expect, that there was a further substantial amount of common ground. The following are my findings based upon the evidence which I have heard.

By the beginning of 1988 the marine market was soft. There was too much capacity chasing the business available to it, and rates were slashed. The LMX market provided little effective capacity: there was a proliferation of XL underwriting in the 1980s. Underwriters were aware that premium rates were at rock bottom: as Mr. Bullen put it, the rates were at the bottom of the pit, as low as they could go. During the renewal season for the 1988 year of account, the ratio of premium to exposure was declining. Yet, conversely, the retentions in the direct market were reducing and therefore the XL market was absorbing greater exposure. Cheap reinsurance meant that insurers in the direct market were willing to take on risks at rates of premium which, in Mr. Bullen’s view, were ridiculous. Mr. Bullen was aware, therefore, that the LMX business which he wrote for the 1988 year of account was more exposed for less premium.

In his annual underwriter’s report, published in May, 1987, Mr. Outhwaite, an underwriting expert called on RTY’s behalf, wrote, in relation to the 1985 and 1986 years of account:

Again there are few large marine syndicates that do not underwrite excess of loss reinsurance but the great majority only do so on the back of their own excess of loss protection. They provide no effective capacity to the market and have little interest in the terms on which they accept business, provided they can reinsure their liabilities at similar or lesser cost.

This last point is confirmed by Mr. Alexander, the defendants’ other underwriting expert:

[Q.] In those circumstances, apart from a deductible at the bottom of the excess of loss specific programme, assuming it exists, there would be no additional capacity added to the LMX market other than the susceptibility of each of its participants to retain losses net once those losses have gone through the roof of their programmes? [A.] Yes.

[Q.] That is right? [A.] That is right [Q.] When we see in various documents references to the fact that the increased trade in whole account covers was adding nothing in terms of real capacity, that is what is being talked about, is it not? [A.] Yes

[Q.] All that was happening is that exposure was being transferred within the market? [A.] Yes

[Q.] You knew that at the time? [A.] I am not sure that we knew the extent to which it was happening

In his report published in May, 1988, Mr. Outhwaite wrote in relation to his 1985 year of account:

. . .there are large numbers of reinsurance contracts from overseas companies placed in London, where the company can only underwrite business by competing with the same London underwriters. The logic of this is impossible to follow, but is again a symptom of a market which is too large for the business available to it. . .more risks have been placed as marine business without accounting for the appropriate Non-Marine premium being classified as such. Indeed, there is one example where there were two substantial refinery fires. . .where the entire loss was collected from marine underwriters under contracts which were placed as wholly Marine risks. Against this can only be described as irresponsible and I have no doubt that some of the problems evident in the Non-Marine market. . .will appear in the Marine market, but of course not for several years. Our own attitude to these risks has continued to be one of extreme caution, attempting to limit any impact from this source on the overall account.

Under the heading “General Matters” Mr. Outhwaite wrote:

The present state of the marine market is as depressing as at any time during the thirty years that I have been at Lloyd's.. . .What is so unfortunate about the present position is that the major competition and irresponsible underwriting is here in London. A short conversation with any Marine broker will confirm this. The fact is there are too many Marine underwriters in London; the consequent scramble for business has virtually destroyed the war and drilling rigs markets as viable commercial and profitable areas and is affecting all others. There seems little prospect of any significant improvement until attitudes change radically.

Although I accept from Mr. Outhwaite that he regarded his annual reports as an opportunity to “sound off” about various matters and, therefore, to an extent, what he has written is to be understood as being in the nature of a polemic, he accepted, (as one would have expected) that the substance of his message reflected his views of the market. In my judgment, that was an appreciation of the state of the market which every sensible and reasonable underwriter who specialized in LMX business would have had prior to the Piper Alpha disaster. After Piper Alpha, the position did not improve in terms of rates in the direct market, although there was an immediate increase of rates in the reinsurance market which continued to be buoyant. In his 1989 report [June 9, 1989] Mr. Outhwaite said:

Excess of loss reinsurances form a major proportion of our account and have proved to be the most consistent and stable area of all. This account is written across areas of the business, with the intention of achieving as great a spread of business as possible so that no individual claim in any particular area can have too much influence on the overall result. This is in marked contrast to most excess of loss accounts underwritten in London where the normal practice is to accept huge aggregate liabilities in any given area (often many times the syndicate’s premium income capacity) and therefore rely entirely on theirown reinsurances (or on there being no catastropheloss).

As will be seen, the last sentence in this report describes Mr. Bullen’s practice.

Furthermore, any underwriter who specialized in writing business in the LMX market, such as Mr. Bullen, ought to have recognized the danger of the spiral effect. Quite part from the many articles which were in evidence in this case and which were in evidence before Mr. Justice Phillips and referred to in his judgment in Gooda Walker there was agreement among the witnesses, including Mr. Bullen, as to the characteristics of the spiral as they were known at the time: that is, when Mr. Bullen was writing the business for the 1988 year of account. These characteristics are, essentially, those pleaded in the amended points of claim namely:

1. It was not known what size loss would produce the spiral effect.

2. Once a loss has entered the spiral, it cannot be predicted how far up the spiral the loss will penetrate.

3. The loss would fall on that reinsurer who ran out of cover first. That depended upon the extent of protection which each “player” in the spiral had effected and the relative reinsurance positions were not known.

It is sufficient for present purposes simply to recite the passages of Mr. Bullen’s evidence on this point:

[Q.] Mr Bullen, I am sorry to press you. Before Piper Alpha, is it right that you appreciated that really it was -[A.] I always appreciated that there was a chance that the spiral would happen, yes.

[Q.] Did you appreciate that you could not really identify or determine, in advance, the size of original loss of a given type that would cause the spiral to be triggered? [A.] Yes, I think that is basically true, yes.

[Q.] Can I just ask you to look briefly again at bundle C2/3, before leaving this topic?

Mr Justice Morison: While that is being looked for, Mr Bullen, can I ask you this: you know that Mr Justice Phillips has described the spiral effect as a game of pass the parcel. So I can understand all the answers you have given, do I understand this to be correct, that, in the first place it was not known what size of loss would trigger the spiral effect? [A.] That is true sir, yes.

[Q.] Secondly, if the spiral effect was triggered, it was not known on whom the loss would fall? [A.] That is also true, sir, yes.

[Q.] You could not predict on whom the loss would fall because that, to an extent, would be a function of competing reinsurance protections? [A.] Yes.

[Q.] As I understand it, you did not know, and it was not available in the market to be known, how much reinsurance any one of the other players in this game held? [A.] Well, we had some idea, sir, because the Hymn sheets, they do show what protections people have.

[Q.] As to who was going to be carrying the parcel when the spiral had worked, you would not know? [A.] No, sir.

[Q.] Because, in order to know that, you would need to know what reinsurances other players in the LMX market themselves held? [A.] Yes, sir.

[Q.] Is that right? [A.] Yes.

In my judgment, having elected to specialize in this market, Mr. Bullen was obliged to give the most careful thought to the implications of the potentiality of the spiral upon his book of business. Unlike the underwriters in Gooda Walker, Mr. Bullen shared the same appraisal of the spiral as Mr. Outhwaite; but in common with them, he appears not to have carried his appraisal through to the way he underwrote the book.

(2) The duties of an underwriter in the LMX market

It was also common ground between the experts, and apparently accepted by Mr. Bullen that in the light of the foregoing, it was the duty of a competent underwriter who wrote a significant amount of business in the LMX market to adopt the following basic principles:

1. To calculate his probable maximum loss [PML] in the event of the worst catastrophe which the underwriter can foresee as a practical possibility.

The ordinary catastrophe is not, however, the catastrophe that properly forms the basis of the PML. It is the remote catastrophe, which is nonetheless a practical possibility, which forms the basis of the PML. The effects of such a catastrophe will not be restricted to the lower bands. [Feltrim at p. 66.]

The frequency of such an occurrence is irrelevant for this purpose. There is a two-stage process: making a judgment as to what is the worst catastrophe event which can be foreseen as a practical possibility [”the PML event”], and what loss will his book suffer if such an event were to occur [”the PML amount”].

2. Having calculated the PML amount, the underwriter must make a judgment as to the amount of vertical reinsurance he requires; and, conversely, the amount of exposure he proposes to run net for the account of his Names.

The assessment of a PML is not an end in itself - it is a stepping stone to planning the reinsurance cover that will be required to reduce exposure to an acceptable limit. If no PML exercise is done, no sound basis will exist for calculating exposure. PML calculations will, however, be of no value if they are not allied to a sound reinsurance programme. [Feltrim p. 15.]

Having ascertained his PML, the underwriter will be able to calculate how much reinsurance cover he needs to buy in order to limit his net exposure to the extent that he considers appropriate. [Feltrim p. 102.]

3. If an underwriter has purchased sufficient reinsurance for the PML event, it is likely that such will be sufficient for any lesser catastrophe, whatever its nature.

By basing reinsurance on the PML, the underwriter sets out to cover the worst catastrophes that he can foresee as a practical possibility on the basis that this will protect him against a wide variety of lesser catastrophes, whether or not the features or circumstances of the particular catastrophe are unprecedented. [Per Mr. Justice Phillips in Feltrim at p. 113.]

The relevant passage from Mr. Alexander’s evidence in the present case is as follows:

[Q.] Yes. Could I attempt to establish the extent to which there is agreement in this case as to general principle in this field? For that purpose, could I ask you to get out bundle L? Can you turn to tab 5, which is Mr Justice Phillips’ judgment in the Feltrim case.

It may be a convenient way of establishing whether there is any difference between the plaintiffs’ experts and yourself, in relation to the general principles to which the judge, in this judgment, refers. Can I ask you to focus on the passage beginning just after the paragraph break on page 355. Do you have that?

Look at the page numbers on the top right of the page. Do you have a paragraph beginning with the words, “The experts were agreed. . .” [A.] Yes.

[Q.] Let me just read the relevant section to you: “The experts were agreed that (a) if not the principle reason why an excess of loss underwriter buys re- insurance cover is to ensure that the syndicate or company for which he is underwriting will be able to weather the adverse effects of any catastrophe for which it is necessary to make provision as a practical possibility”. Do you see that? [A.] Yes.

[Q.] Do you accept that? [A.] Yes.

[Q.] “It was also agreed that in order to assess the consequences of such a catastrophe, the underwriter has to carry out a PML exercise”. Do you agree with that? [A.] Yes.

[Q.] “Having ascertained his PML, the underwriter will be able to calculate how much re-insurance cover he needs to buy in order to limit his net exposure to the extent that he considers appropriate”. Do you agree with that? [A.] Yes.

[Q.] Does it follow, Mr Alexander, that unless the excess of loss underwriter calculates his PML, he does not know how much re-insurance, on a vertical basis, he needs? [A.] Yes.

[Q.] Can I ask you, now, to go to page 366 of the same bundle? If you would, Mr Alexander, just focus on the last paragraph and, in particular, on these words: “By basing reinsurance on the PML, the underwriter sets out to cover the worst catastrophes that he can foresee as a practical possibility, on the basis that this will protect him against a wide variety of lesser catastrophes, whether or not the features or circumstances of a particular catastrophe are unprecedented”. Do you accept that? [A.] Yes.

[Q.] So, if I can rather brutishly paraphrase it. The philosophy is, that if you assess your re-insurance requirements in relation to your PML, the feeling is that you should be okay for lesser catastrophes? [A.] Yes.

[Q.] Could I ask you to go back to page 319 of the same bundle, Mr Alexander? Can you focus on the second paragraph. You need not worry about the first two-thirds of it, because it all relates to issues specific to the Feltrim case. I would like you to look at the passage, the sentence beginning about five lines up, the penultimate sentence in the paragraph. It is centre page, it begins with the words: “Ordinary catastrophe is not, however, a catastrophe that properly forms the basis of the PML. It is the remote catastrophe which is nonetheless a practical possibility, which forms the basis of the PML. The effects of. . .” Just stop there. Do you accept those propositions? [A.] Yes.

[Q.] So, it is common ground between us, Mr Alexander, is it, that in assessing a syndicate or company’s PML the excess of loss underwriter takes, as his event, the worst catastrophe which he can foresee as “a practical possibility”, impacting his account? [A.] Yes.

[Q.] It follows, does it not, that having identified such an event and having identified it as “a practical possibility”, the underwriter, in assessing his PML, will not be entitled to discount the impact of such an event, because it is likely to happen only infrequently? [A.] Yes, I think that is right. The evidence of Mr. Outhwaite was to the same effect.

Mr. Crane Mr Outhwaite, can we just start, please, by seeing the extent to which there is any difference between us on basic principle? To that end, may I do with you what I did with Mr Alexander, show you certain passages in the Feltrim judgment and ask whether you agree with them? Could you be given Bundle L, please? (Handed)? [A.] L1?

[Q.] There is just one bundle. [A.] I have L2.

[Q.] This gentleman will help you get the bundles, so you need not bother yourself. Could you turn to tab 5 and go to page 355, the pagination at the top right of the document? [A.] Yes.

[Q.] That is Mr. Justice Phillips’ judgment in Feltrim. Just after the paragraph break he says:

The experts were agreed that (a), if not the principal reason why an excess of loss underwriter buys reinsurance cover is to ensure that the syndicate or company, for which he is underwriting will be able to weather the adverse effects of any catastrophe for which it is necessary to make provision as a practical possibility. It was also agreed that, in order to assess the consequences of such a catastrophe, the underwriter has to carry out a PML exercise. Having ascertained his PML, the underwriter will be able to calculate how much reinsurance cover he needs to buy in order to limit his net exposure to the extent that he considers appropriate.

You were one of the experts in the Feltrim case? [A.] Yes.

[Q.] Does Mr Justice Phillips accurately summarise your position in saying that you and other experts agreed to those propositions? [A.] Yes.

[Q.] Could you now go to 366? Look at the last paragraph on that page, where the learned judge says that:

By basing reinsurance on the PML, the underwriter sets out to cover the worst catastrophes that he can foresee as a practical possibility on the basis that this will protect him against a wide variety of lesser catastrophes, whether or not the features or the circumstances of the particular catastrophe are unprecedented.

Do you agree that that should be the approach? [A.] Yes.

[Q.] And that, in fact echoes your evidence in both Gooda Walker and Feltrim? [A.] Yes, I hope so.

[Q.] Could I ask you to go to page 319? There I would like you to look at the first full paragraph and focus on the last five or six lines, you will see a sentence, the pre-penultimate sentence begins:

The ordinary catastrophe is not, however, the catastrophe that properly forms the basis of the PML. It is the remote catastrophe, which is nonetheless a practical possibility, which forms the basis of the PML. The effects of such a catastrophe will not be restricted to the lower bands.

Do you see that? [A.] Yes.

[Q.] Now forget the last sentence because that has a specific reference to the Feltrim case. Do you agree with the two sentences that precede it? [A.] Yes.

[Q.] Could I ask you to come back to page 268? To give the passage at the top of the page some content, could you look at the bottom of 267, where the learned judge says:

In these proceedings the Defence experts tended to muddy the water somewhat by postulating that the casualty envisaged as the basis for the PML calculation should be an event that was “likely”. On analysis it proved that this adjective was used to find the practical, as opposed to the theoretical, possibility, and the expert evidence gave me no cause to revise my findings on this topic in Gooda Walker.

Mr Outhwaite, would you accept that once an underwriter identifies a potential event as a practical possibility in the sense that, albeit unlikely, it might occur and impact his account, that is an event to which you must have regard? [A.] Yes, I would add the word “realistic”.

[Q.] In the context of assessing his PML exposure, once he has identified an event as a realistic or practical possibility, its likelihood thereafter, or degree of frequency, is irrelevant to PML exposure? [A.] Thereafter, but of course the likelihood of it occurring is implicit in his choice.

[Q.] You would accept, would you not, that if he can foresee as a realistic possibility that it might occur, then that is an event to which he must have regard in the context of assessing PML exposure? [A.] Yes.

[Q.] If the event satisfies that test, its frequency, in other words, the regularity with which it occurs, is irrelevant in that context? [A.] Once you made that decision, you do not make another decision as to whether it is likely to occur once every 20 years or 40 years, quite.

[Q.] Yes. Its frequency, however, the frequency of an event, or the frequency of circumstances giving rise to loss, is however, relevant to rating, is it not? [A.] On individual risks, it is a question of whether that risk is likely to be affected by a loss or not.

[Q.] Yes. Would you accept that whereas in rating a risk the frequency of loss is a factor which is taken into account, in the context of a PML assessment, the same criterion is not relevant? [A.] Well, you do not make a further judgment. Once you made your judgment as to the PML event, in which the possibility or the frequency of loss is a factor, you do not do it again.

It is also common ground that every competent underwriter would need to plan his account, so that he could exercise the judgments to protect his Names. Mr. Alexander’s evidence was as follows:

[Q.] If an underwriter wants to know what his likely PML is going to be in order to buy re-insurance incepting from the January of the coming year, he would need to know what his aggregates are going to be? [A.] Yes.

[Q.] Because without knowing the latter he cannot work out his PML, is that right? [A.] Yes.

[Q.] Unless he has a PML, he has no idea how much re-insurance he is going to need? [A.] That is right.

[Q.] Is that your evidence, that a competent excess of loss underwriter would have had to have done that during the renewal season in relation to the incipient or coming year of account? [A.] Yes, I do not say, necessarily, it has to be written down, but he would have had to have formulated his ideas as to what his underwriting strategy was for the following year.

[Q.] That would involve a plan as to his intended aggregate? [A.] I suppose so.

[Q.] Yes. You say you suppose so. Without that idea in your head you have no idea of the PML to which your re-insurance will be asked to respond? [A.] Yes.

[Q.] So you accept the proposition, do you? [A.] Yes.

[Q.] Having done that, you work out your PML on the intended aggregate? [A.] Yes.

[Q.] And then you will know how much re-insurance you need, at least to start with, incepting as from the 1st January in the coming year? [A.] Yes.

[Q.] Is that how you do it? [A.] Yes.

Mr. Outhwaite accepted the need for a plan but in rather more guarded terms than Counsel was putting to him:

[Q.] Now, before we leave Bundle L5, I wonder whether I could ask you to look at page 304 of the bundle? Could you focus on the last paragraph, Mr Outhwaite, if you would be so kind, where the learned judge says:

A competent underwriter when planning his next year’s policy will not merely plan his gross exposure, involving target aggregates, and the extent to which he will cover this by reinsurance, but he will also project the premium that he anticipates he will receive and the premium he intends to spend on reinsurance. Should rates alter from those which he has assumed when making his plan, the competent underwriter will consider the effect of the change and adjust his plans accordingly.

Do you see that? [A.] Yes.

[Q.] Do you agree with it? [A.] In general, yes.

(11.45 am)

[Q.] So you would accept, would you, Mr Outhwaite, that the competent excess of loss underwriter in planning the coming years of account will start by attempting to assess the premium income that he is likely to have and the reinsurance that he is likely to be able to buy in the market with the premium income which he assumes he will have at his disposal? [A.] I would not quite - well, it would depend on the Syndicate very much as to whether that was an appropriate equation.

[Q.] I see. [A.] If you are talking about a specialist excess of loss Syndicate, you are, if you are considering what you are going to do next year, in a position where there is, as it were, a status quo and you know where you are, you know the premium that is being developed, you know how much you are spending on reinsurance. You know where you are. You may make the assumption that things are going to remain exactly the same in the ensuing year. You may think it is going to change because of changing market conditions or you may change your underwriting policy, in which case, you make other decisions.

[Q.] But you will have, will you, from the outset, an idea of the amount, if any, of PML exposure that you intend to run net? [A.] You will have an idea of what PML exposure you have actually at the time, which arrives from the book of business which you are already on. You may then, in deciding what you think you are going to do with your underwriting next year, make some estimate of what you think that means in terms of the PML for next year because you effect reinsurances. If you effect them from the 1st January, they cover, of course, the whole of the 12 months, so it must be appropriate to what you think is going to happen during the next 12 months.

[Q.] Unless you have calculated your probable PML exposure from the aggregates in force from the - [A.] You mean assessed?

[Q.] Assessed or calculated. Can I finish the question before you pull it to bits - [A.] Sorry.

[Q.] - then feel free. Unless you have calculated your PML exposure deriving from the aggregates that you believe will be in force on 1st January of the coming year, you will not know, will you, how much reinsurance you need to buy? [A.] Well, you know what your aggregates are. If you are considering this, say, in November, you know what your aggregates are, and if you underwrite the same book of business for next year, then the aggregates will be no different for next year. The only time when there is any sort of additional assessment to be made is if you are planning to change the aggregates which you are going to write for next change, or you are changing your underwriting policy, which implies a change in the aggregates. Then you would have to take that into account when deciding how much reinsurance you are going to take. You are not doing it on the basis of the fixed aggregates at the 1st January, you are doing it on the basis of what you think the aggregates are likely to be during the space of the 12 months.

[Q.] Yes. Now unless you are merely restructuring an existing book, you will have to do that assessment, to use your word, in relation to the in-force aggregates as opposed to the aggregates you write into a given year of account, will you not? [A.] Yes.

[Q.] Unless you have done that assessment with a view to determining the PML exposure which you are likely to have at the relevant period, you will not know how much vertical cover you need to buy, will you? [A.] You will not know exactly, no.

[Q.] You will not have a clue, will you, because you will not know what your PML exposure is likely to be? [A.] That is going much too far, Mr Crane. As I have said, the practical circumstances are that a Syndicate goes on from one year to the other. You know what your aggregates are. Whatever system you use, whether it was a merely annual, or however you kept them, by November you know what the year’s aggregates are. Therefore, you are starting off with a known base. So to say that you have not got a clue what your next year’s aggregates are is a complete exaggeration of the position. You do not know precisely what they are, and whatever decisions you make for the 1st January must be based on an assessment of what they are going to be during the next 12 months. They cannot be precise because you have not written the business yet.

[Q.] Why should you assume that this underwriter is merely renewing an existing book? [A.] I am saying if that is what he does there would be no change. And of course the vast majority of any underwriter’s book tends to be the renewal of the business which he previously had. The question is, what changes will there be for the ensuing year - whether he is writing additional business or he is going to write less; business of a different type or whatever?

[Q.] That is something he has to address when planning the year of account in question? [A.] He has to take that into consideration, yes.

[Q.] Unless he does, he is unlikely to have the material which enables him to assess the probable PML exposure applying to the coming year of account; that is right, is it not? [A.] You make an assessment, if you are planning the account, as to what it is likely to be. It is just an assessment. There is no material except what the state of play is now and what your - as the underwriter - intention is regarding the next year of account. That is all you are working on.

[Q.] It is a business plan, is it not, that is a grand way of describing it? [A.] Yes.

[Q.] Now, from that plan, you will have an idea in general terms of the aggregate that you intend to commit or accept? [A.] Yes.

[Q.] Having established that, you will have the aggregate, from which you can assess the PML exposure deriving from it? [A.] Yes.

[Q.] My suggestion to you, Mr Outhwaite, is that unless you had done that, you are in no position to determine the amount of vertical reinsurance that you are going to need for the coming year of account? [A.] Well, I suppose that is right, Mr Crane, but I find it very odd. When a Syndicate is actually operating it has its aggregates, it has its PMLs, it has its reinsurance and it is considering what to do for the 1st January. You are saying, “If you do not do any of these things you have got no idea what to do”. I mean, I am bound to agree with that, but it does not seem to me to be a practical proposition.

It seems to me that the qualifications which Mr. Outhwaite is making have no substance in this case: Mr. Bullen was substantially increasing [48 per cent.] the amount of aggregates he was writing for the 1988 year of account over those written in the previous year of account, in the light of an increase in stamp capacity of 53 per cent.; in 1988 he wrote 26.5 per cent. more contracts than in the previous year. By no stretch of the imagination was this simply a renewals book, although it is true that a substantial part of his 1988 book will have been renewals of contracts written in 1987. In my view, Mr. Outhwaite’s qualifications do not impinge, and I do not think he intended them to impinge, upon an acceptance of the proposition put forward by Mr. Justice Phillips in the Feltrim judgment namely that:

. . .A competent underwriter when planning his next year’s policy will not merely plan his gross exposure, involving target aggregates, and the extent to which he will cover this by reinsurance, but he will also project the premium that he anticipates he will receive and the premium he intends to spend on reinsurance. Should rates alter from those which he has assumed when making his plan, the competent underwriter will consider the effect of the change and adjust his plans accordingly.

This was a proposition with which he was prepared to agree, in general terms. But in any event, I have to say that, if he were making any significant qualification I would prefer Mr. Alexander’s opinion on this point. Having accepted the need to calculate a PML for the purpose of deciding what risks to run net, the need for a coherent underwriting plan is, I think, obvious; and was rightly accepted without qualification by Mr. Alexander.

(3) What did Mr. Bullen do

The real issues of substance on liability in this case are:

1. Did Mr. Bullen have any underwriting plan; did he calculate his PML and, if so, how; what judgment did he make about running risks net, for the account of his Names? In short, did he meet the standards, set out above, to be expected of a competent underwriter when writing his business for the two relevant years of account?

2. How would a competent underwriter have written his book of business for those two years? I shall deal with this part of the case under the heading “Causation”.

I bear in mind the following:

1. The definition of a PML event is imprecise and unattractive to an actuary. But, in my judgment, its meaning is clear in the present context. An underwriter is not to be criticized for preferring to take one sensible event as opposed to another, or for excluding some events [which might be described as cataclysmic] as being so massive and so destructive as to destroy the worldwide insurance market [for example, an earthquake in San Francisco Bay].

2. The calculation of the PML amount will involve judgment. And an underwriter is not to be criticised for making a calculation of the PML amount merely because others would or might have made a different calculation. As Lord Justice Bingham said, in his dissenting judgment, in Eckersley v. Binnie, [1988] 18 Con.L.R. 38 at p. 80:

The standard is that of the reasonable average. The law does not require of a professional man that he be a paragon, combining the qualities of polymath and prophet.

I read that passage as saying no more than he said, later, in Banque Bruxelles Lambert S.A. v. Eagle Star Insurance Co. Ltd., [1995] L.R.L.R. 195 at p. 201, col. 1; [1995] 2 All E.R. 769 at p. 840c-e:

. . .In each case the duty is to exercise a reasonable standard of professional care in the circumstances, no more no less.. . .The complaint made and upheld against the valuers in these cases is accordingly not that they were wrong. A professional opinion may be wrong without being negligent. The complaint in each case is that the valuer expressed an opinion that the land was worth more than any careful and competent valuer would have advised.

In each passage, the Master of the Rolls is emphasizing that the standard of care to be expected of a professional is not that he is right, or that he is a paragon who is never wrong, but that he does not fall below the standards to be expected of people doing his job or pursuing his trade.

3. Finally, the underwriter must make a judgment as to the amount of exposure he will carry net for the account of his Names. The fact that this is a matter of judgment, where there is room for more than one view, does not give the underwriter carte blanche deliberately to expose his Names to an unreasonable extent, even if the Names on his syndicate join Lloyd’s on the basis that they are accepting unlimited liability. There is a substantial difference between agreeing to accept unlimited liability and agreeing to being deliberately exposed by an underwriter to unreasonable losses.

The fact that a Name who joins Lloyd’s deliberately agrees to expose himself to unlimited liability does not mean that he anticipates or accepts that when he joins a syndicate the active underwriter will deliberately expose him to the risk of such liability. On the contrary the Name will reasonably expect the underwriter to exercise due skill and care to prevent him from suffering losses. In many categories of insurance the Name will reasonably expect the underwriter to plan to procure profits, year in year out. The fact that syndicates are constituted each year does not, however, make it mandatory for an underwriter to conduct business in this way. Underwriting at Lloyd’s must be conducted as an ongoing business. There is no reason in principle why an underwriter should not write business on the basis that net losses will be made in some years that are balanced by generous profit levels in the other years. If, however, an underwriter is deliberately to expose his Names to suffering losses from time to time, he must make sure that the Names are aware of this and of the scale of loss to which they will from time to time be exposed. [Feltrim pp. 16-17.]

The materials upon which I am asked to resolve the question as to what Mr. Bullen did are: the pleadings in the action, his statement of evidence, his manner and demeanour as a witness, the transcript of his cross examination, the transcript of the evidence he gave to the Loss Review Committee on three occasions between December, 1991 and July, 1992, and the contemporary documents. For reasons which I shall elaborate, I have had no difficulty in concluding that much of what Mr. Bullen says he did, he did not do. He was not a good witness: he was realistically described by RTY’s Counsel as an “unimpressive witness”. In reaching my conclusions I have borne in mind a number of factors urged on me.

In the first place, Mr. Bullen has not been underwriting since 1992, and, to some extent, is out of touch with Lloyd’s and its workings. He was being asked to recall events which occurred seven or eight years ago, and I must make due allowance for the dimming of recollection due to the passage of time. He clearly found the process of giving evidence over four days [Days 6-9, inclusive] stressful, although he remained good-natured and patient throughout. He, too, has suffered loss, as he participated in the syndicate’s misfortunes. Mr. Hirst, Q.C., on behalf of RTY, submitted that “he has been ruined by the disasters which beset his Syndicate”. I am prepared to assume that that is so. I should make it clear, as Mr. Bullen himself did [Day 9 p. 190], that he was sorry for what had happened.

Mr. Bullen’s witness statement, which was signed on Aug. 16, 1995, asserted the following [and I cite extracts]:

All of the Syndicate’s LMX writings were, in theory exposed to loss. However, for reasons I explain below I considered that the majority of this exposure was purely theoretical. I as a marine underwriter and a reinsurer of other marine insurers and reinsurers always based my PML assumptions upon what I considered to be the likely impact of a realistic major disaster in the North Sea involving a rig.. . .I was always willing to run the risk of sustaining a retained loss to the Syndicate in the event of a major marine loss occurring on the basis that, first, the probability of the loss occurring was remote. Second, the loss was only likely to impact a proportion but not all of the Syndicate’s XL book. Third the costs of protecting this remote exposure on an annual basis were unlikely to be viable commercially, having regard to the profit margins of the Syndicate and the low frequency of loss. I always thought, and I believe I was correct in assuming, that even in the event of a major market loss there would not be a wholesale accumulation of claims across the portfolio, in particular the upper layer catastrophe covers, the vast majority of which were rated within the 0-4 rate on Line Band. For PML purposes I had always worked on the premise of a very serious rig disaster, combined with a hull loss. The worst case realistic loss I envisaged was a hull crashing into rig and causing an explosion. For the reasons I have explained. . .I did not consider that other loss scenarios were as likely to impact the Syndicate’s essentially marine writings as seriously. During the relevant period I always considered that a major rig disaster of the type I envisaged could expose my Names to a loss in the order of 100%-120% of stamp capacity. I did not assess this figure for each year of account at any single fixed point in time, but had it in mind when shaping the Syndicate’s Book and reinsurance programme, as both evolved, throughout and following the renewal season for each year of account. I also considered such a level of loss to be acceptable in respect of any major loss event on a worst case basis, given the Syndicate’s successful track record, the very low probability of such large losses occurring and the likelihood that I would be able to recoup substantial elements of any such gross loss under pay-back principles and thereby diffuse it.. . .During the material period in question I never made written PML calculations. There was no formal requirement for me to do so. My deliberations and calculations upon the appropriate level of vertical reinsurance to purchase were entirely mental, since they were invariably carried out during the course of periodic inspections and reviews of the Syndicate’s records.

AGGREGATE EXPOSURE TO PML EVENT

Mr. Bullen then gave detailed evidence about how he made his judgment as to the risks of accumulation in his book of business, and the discount factors which he said he applied, in his mind, when calculating his PML. These factors were conveniently summarized by the plaintiffs’ Counsel in Exhibit D1. Mr. Bullen accepted that the exhibit accurately reflected what he had said in his witness statement: what the exhibit shows is the impact which Mr. Bullen thought a PML event would have on the aggregates recorded in his aggregate book: see diagram on p. 444 ante.

I reject Mr. Bullen’s evidence about the way he wrote his book of business. I do not accept that he planned his book of business; or that he used a PML calculation to determine the amount of vertical reinsurance protection he should buy, or that he made any informed judgment as to the net exposure to which he could properly expose his Names. It is fair to say that whenever asked about his PML event, Mr. Bullen has consistently described it in the same way: namely a collision between a hull and a North Sea oil platform or rig. I am not, however, convinced that he used any PML calculation for the purpose of planning his account or for determining the amount of vertical reinsurance he should buy. What I believe has happened in this case, is that Mr. Bullen has embellished his recollection of events by reference to the decision of Mr. Justice Phillips in Feltrim which was given on Mar. 10, 1995. I am of the view that Mr. Bullen may well now believe that he did what he says he did, so that I can, and do, acquit him of deliberately seeking to mislead the Court. Nonetheless, it is regrettable that he has put forward a case, with much circumstantial detail, which I feel obliged to reject.

My reasons for reaching this conclusion are as follows:

Mr. Bullen’s alleged policy of calculating his net PML exposure and limiting the amount carried net to a figure approximating to stamp capacity must have been regarded by him as fundamental to his underwriting approach. Yet, he never mentioned this until after the Feltrim judgment, when further and better particulars of the points of defence were delivered on Mar. 24, 1995. There is no mention of this aspect of his policy to the Loss Review Committee; nor was any mention made of it in any contemporary correspondence with managing agents. The account given of the way he calculated his syndicate’s PML exposure is wholly inconsistent with a letter sent to one of the managing agents and dated Jan. 4, 1991. RTY wrote [this is an extract, only, of the letter]:

You will recall that question 5 and 7 were as follows: -

5) How was the aggregate exposure of the Syndicate assessed for any one (a) risk and (b) catastrophe?

7) What dictated the Syndicate’s decision to purchase US$85 million of General Excess cover for the 1988 Year of Account?

The answer to these questions are inter-related, for the Underwriter’s assessment of the aggregate exposure of the Syndicate to any one risk or catastrophe was an important factor in his decision as to the amount and type of reinsurance to be purchased. . .

It follows that in answering these questions it is necessary to give a general outline of the policy adopted by Mr Bullen as Underwriter of Syndicate 255. Such an outline cannot be comprehensive, and if what is set out below gives rise to further queries, we shall be most happy to answer any reasonable specific questions which you may have.

The policy adopted by Mr Bullen for the 1988 year of account was broadly that which he had successfully adopted in earlier years. The great majority of the Syndicate’s Premium Income came from reinsurance business, predominantly excess of loss and Whole Account. Modest lines were written on numerous risks, thus producing as big a spread as possible, and risks were spread further by writing all areas from the lower end protections to the top end. Where such business was written at a high level and was given a low premium rate by the Market, this entailed significant aggregate exposure with a very low probability of loss. In those cases where Mr Bullen wrote business at a low rate, his judgment did not differ from that of the Market.

In order to consider the Syndicate’s exposure and the adequacy of the reinsurance programme, an analysis of written business was prepared each month. The analysis showed the aggregate exposure of the Syndicate on different classes of business in a form similar to the annual analysis supplied with our letter of 20 July 1990.

As the aggregate exposure underwritten by the Syndicate increased, additional reinsurance protection would be purchased. At each stage the Syndicate’s exposure in the event of catastrophe would be assessed.

The catastrophe assessment was made on the hypothesis that a rig was involved, this being seen as the highest possible insured value.

The essential question of judgment for Mr Bullen lay in assessing the need for Reinsurance protection to cover the possibility that a rig loss would penetrate the Excess Layers or Whole Accounts which the Syndicates has reinsured.

Mr Bullen considered that prudence required him to assume that such a loss would work its way through to those Layers or whole Accounts which had a Market rating of 10% or higher. Accordingly, full Reinsurance cover was purchased for such business. In his judgment, however, the danger that a rig loss would penetrate into those Layers and Whole Accounts with Rates on Line of less then 5% was so very remote that it did not require Reinsurance, while in those cases where the Rates on Line were between 5% and 10% partial Reinsurance only was desirable.

This judgment was based on Mr Bullen’s considerable underwriting experience, and took account (amongst other things) of Market experience of risks with these Rates on Line, of the Syndicate’s experience since 1980, and of the Syndicate’s information as to the exposure to rig risks and the loss record of those whom it reinsured. Consistently with Mr Bullen’s judgement, when in April 1988 the “Enchova” drilling rig platform resulted in a $400 million loss, this absorbed only 21% of the Syndicate’s Reinsurance protection.

No judgment of where the Syndicate should retain risk could ever be guaranteed. If all Excess of Loss and Whole Account business written by the Syndicate had been Reinsured, the cost of doing so would have been wholly uneconomic.

We would add that it is recognised in the Excess of Loss and Whole Account Markets that the business is long term: the danger that the Underwriter’s judgment of where to retain risk may prove wrong in a bad year is balanced by high rewards experienced in good years. Those rewards are reflected in the results of the Syndicate prior to Piper Alpha. Following the Piper Alpha loss, the market revised its rates dramatically to ensure that those rewards continued.

The evidence showed that this letter was written on legal advice and after Mr. Bullen had had a chance to approve it. Its importance is that it shows a completely different manner of calculating exposure to loss than that advanced by Mr. Bullen in his evidence.

At the Loss Review Committee hearings [sometimes called the Boatman Inquiry], Mr. Bullen was asked about this letter and confirmed its accuracy, although he told the Court that he had appreciated its inaccuracy during those hearings:

Mr. Justice Morison: It looks as though what you were telling the Boatman Committee accords almost exactly with what we see in the letter of January that we have just been referring to, does it not? [A.] Yes, it would appear so, yes.

Mr. Justice Morison: 4th January letter? [A.] Yes.

Mr. Crane: Mr Bullen, in fact, the position is put beyond doubt, if you turn to page 610 of the transcript in H2, and go to letter F at the foot of that page? [A.] Yes.

[Q.] Because here Mr King actually reads out to you one of the relevant sections of the letter. He says:

Mr Bullen considered (that is reading from the letter) that prudence required him to assume that such a loss would work its way through to those layers or whole accounts which had a market rating of 10 per cent or higher. Accordingly, full reinsurance cover was purchased for such business. In his judgment, however, the danger that a rig loss would penetrate into those layers and whole accounts with rates on line of less than 5 per cent was so very remote that it did not require reinsurance. In those cases where the rates on line were between 5 and 10 per cent partial reinsurance only was desirable.

Then he goes on to talk about Enchova. Then the question:

Do you think that was a fair assumption of your thinking in respect of your buying reinsurance for the 1988 account?

Answer: Yes

Now, you told me earlier that you appreciated the letter was incorrect in the course of your interviews with the Loss Review Committee. Here you are confirming the correctness of the letter as a description of your approach to assessing exposure and purchasing reinsurance. What is the explanation, Mr Bullen? [A.] I have not got one, have I?

[Q.] Can we safely assume that your last explanation, or your last statement given 5 or 10 minutes ago, that it was while being interviewed by the Loss Review Committee that you came to appreciate the incorrectness of this description, that that is not correct? [A.] At the time, I thought it was incorrect. Whether or not I actual said - I thought afterwards when I thought about it after it, I mean, these things were not easy to go through, and maybe at the time I said that and then thought about it afterwards, I am not quite sure.

At the conclusion of the hearings, RTY’s solicitors made a written submission to the committee, which was seen and approved by Mr. Bullen. That submission made no reference to exposing Names to a loss of 100 per cent. to 120 per cent. of stamp; nor did it show that Mr. Bullen adopted the method of calculating his net exposure in the way the case was pleaded in the further and better particulars. In particular, Mr. Bullen was telling the Boatman Committee and, through the letter, the managing agent, that he did not consider it necessary to purchase reinsurance cover for any risks on his XL on XL book or his whole account book at rates on line of less than 5 per cent.; whereas his case before this Court was materially different - as can be seen from the table. Further, the case now put forward was different from the case which his deputy, Mr. Green, advanced before the committee. Mr. Bullen never told his deputy of his policy of deliberately exposing his Names to a risk of 100 per cent., or thereabouts, of stamp capacity in the event of a PML event nor were RTY told and nor were the Names.

The submission, by RTY to the committee, after taking legal advice, was to this effect:

Mr. Bullen and his staff at all material times believed that in the light of the factors referred to above this reinsurance protection was adequate for all foreseeable Market catastrophe losses. Mr Bullen made this quite clear in his underwriting reports.

Such a statement could not honestly have been made if Mr Bullen had calculated his PML and decided to run a net exposure of 100 per cent. of stamp.

Although Mr. Bullen’s case is that his underwriting policy remained more or less constant, it is possible to use the information in the pleaded case and apply them to the previous years of account in order to test whether there is any discernible pattern. There is not. It is clear that Mr. Bullen cannot, in earlier years, have been buying reinsurance protection on the basis of his pleaded case. For the 1984 and 1985 years of account, the vertical reinsurance comfortably exceeded what would have been Mr. Bullen’s PML amount; in 1985, vertical reinsurance was purchased for aggregate which, on his own case, was not exposed to his alleged PML. In 1986, the net exposure was 88.8 per cent. of stamp and for 1987 65.1 per cent. of stamp and for 1988 80.66 per cent. of stamp. These wayward results demonstrate that Mr. Bullen had not purchased vertical reinsurance on the basis that he was deliberately running for his Names a net exposure equivalent to 100 per cent. of stamp. Yet, there is no suggestion that the policy changed over the years; rather, the reverse. And, certainly, the Names were never advised of any change of policy and there is no evidence that he ever so informed RTY of any such change. In these circumstances, it is difficult to understand the change in net exposure between 1985 and 1986. What this wayward pattern demonstrates, I think, is that the vertical reinsurance protection which Mr. Bullen purchased was not calculated in any structured way but was based upon his feel for what he ought to do, the need to show a profit at the end ofthe year, his belief that serious disasters were unlikelyto happen very often and that the aggregates he wrote at rates on line of less than 5 per cent. represented premium income for no risk.

Mr. Bullen’s witness statement departed from the further and better particulars in one significant respect, even though in it [par. 153], he expressly endorsed what had there been pleaded. Given the importance of this part of the case, the inconsistency can only be described as remarkable, and inexplicable. The pleaded case was that Mr. Bullen, when calculating his PML exposure, applied a discount of 20 per cent. to the XL on XL book in relation to aggregates written at rates on line of less than 10 per cent.; in Mr. Bullen’s statement he said he applied a discount of 20 per cent. across the board in relation to this part of the book. There is a significant difference between the two. While errors of this sort very often occur for quite understandable reasons, I am bound to say that the view I have formed is that this particular error is not due so much to carelessness, rather to the fact that Mr. Bullen’s case is founded on sand. If he had adopted a policy which had been in place for some years, I would have expected a greater degree of precision in the presentation of that policy, and an awareness of the departure from it.

Further, his justification for the 20 per cent. discount does not make sense, and shows, I think, that this was not so much evidence of a policy which was applied, but rather a case which was being put forward for the purpose of defending the action. The explanation for the discount was that the XL on XL aggregates would have included back-up covers “which would not all be on risk at the date of any specific loss”. In 1988, the aggregate records did not list, separately, the back-up covers; but in 1989 they did. The use of the discount of 20 per cent. for 1989 could not have been justified for the reason advanced.

Mr. Bullen was questioned about this, but had no explanation:

Mr. Justice Morison: I think the point is this, with respect, Mr Bullen: in order to calculate the general discount of approximately 20 per cent, one of the factors you say you took into account was the fact that at the bottom end aggregate would have comprised back-up covers? [A.] Yes.

Mr. Justice Morison: Therefore, you came to a discount factor of approximately 20 per cent. But for the year 1989, you would have known what the back-up covers were, as a matter of fact. Therefore, it could not have been a factor in calculating your 20 per cent discount; is that the point, Mr Crane?

Mr. Crane: It is the point, my Lord.

Mr. Justice Morison: Therefore, your account of how you did this exercise simply is incredible? [A.] I can only say what we did, sir, I cannot say any more.

There is not a single piece of contemporary written material which refers to or evidences any PML calculation. In his witness statement he said that -

. . .his deliberations and calculations upon the appropriate levels of outwards vertical reinsurance to purchase were entirely mental.

Having regard to the complexities of the calculations, this did not have the ring of truth, although I fully accept that Mr. Bullen could not be accused of falling below proper standards merely because he had no written underwriting plan. However, he said in evidence that he made calculations on pieces of paper which were then thrown away. When asked about the phrase in his statement that he “never made written calculations” he had no satisfactory answer to give:

[Q.] Mr Bullen, why did you say in your statement that the calculations were not written, they were entirely mental, where, in truth, as you now tell us, your evidence is that they were written, in the sense that you made an inscription on a document of the various stages of the relevant calculation? [A.] It was not put into a file anywhere, it was just working it out as one does - an underwriter does, he scribbles and works out and fine, okay, that is good and throws it away.

[Q.] Is this what you are telling the Court, that you described them as “not written” because they were not filed and retained? [A.] Yes.

[Q.] Just keep one hand at paragraph 152 and go back to paragraph 44. Were you saying until 1988 this information, that is the information on the monthly periodic aggregate tables was provided on paper notes which were routinely thrown away? [A.] Yes.

[Q.] Is the effect of your evidence to the Court that that is the same position in relation to the PML calculations: they were recorded in writing on a note, but were routinely thrown away? [A.] Yes, although these were more actually more formal than my scribbling on my pads.

[Q.] Why did you not say that in relation to the PML calculations when compiling your statement of evidence in this action? [A.] I have no answer to that.

[Q.] Did you read the statement before you signed it? [A.] I did. [Q.] Did you read that paragraph? [A.] Yes, but I did not realise what this would mean now.

[Q.] Presumably, Mr Bullen, you understood when you were giving this statement that whether or not you had a method for calculating your PML was likely to be crucial in this case or at least very important; is that right? [A.] Yes.

[Q.] Did you not think it important in those circumstances to satisfy yourself that what you said about your PML calculations was correct and unequivocal? [A.] This is why I have asked the Court to take it into consideration this morning, having read it through again.

It would appear that Mr. Bullen purchased his vertical reinsurance protection on a much more simple basis:

[Q.] You begin by saying:

It would appear from letters I have received that some agencies and certain Names are concerned regarding the loss of the drilling platform, Piper Alpha.

That is the subject of the letter. I will be coming back to this in another context, but could I ask you for current purposes to go to the foot of p. 167? Where you say:

In this report I have tried to explain the situation as it is. Like many syndicates writing excess of protections

- what do you mean by that, excess of loss protections? [A.] Yes.

[Q.] “We have protections. . .”, that is presumably reinsurance? [A.] Yes.

[Q.] “. . .for about three times our income. Cost wise it is not possible to try to cover our aggregate liability and I feel this is a risk we must take.” [A.] Yes.

[Q.] Why did you say that? Firstly, what do you mean, three times our income, what are you talking about? [A.] Our XL income.

[Q.] Your gross excess of loss income? [A.] Yes.

[Q.] And 1988 was no different from earlier years, was it, in this respect, that your cover was about three times your excess of loss gross income? [A.] That is a starting point, yes, or there around, yes.

[Q.] When you say “starting point”, starting point for what? [A.] I have just said, you have to buy your protection before the year starts. So you start somewhere. Then you build it up as you go through the year.

[Q.] I see. So the starting point, before the year starts, in determining how much protection you are going to buy, to incept on 1st January from

[1996] 449 L.R.L.R. Berriman v. RTY (Q.B. (Com. Ct.)) Morison, J.

the following year, will be roughly three times the excess of loss income that you assume you will receive in the course of that year. Is that right? [A.] Yes, around that, yes.

[Q.] Then, having used that to identify at least the initial quantum of reinsurance, how do you adjust it thereafter? [A.] By going through the aggregates, which I told you yesterday about.

[Q.] What do you do when you go through the aggregates, in order to make a decision as to whether you wish to buy some more? [A.] Well, as the aggregates build up, you then compare that with your PML situation, and decide, you know, perhaps it would be advisable to buy some more cover. It is always a hit and miss situation, but I mean -

[Q.] What tells you how much you have to buy extra? [A.] Well, it is something that you have, really. It is - an underwriter has - it is a “feel” situation.

[Q.] How does this “feel” operate? How does the “feel” translate into a figure for reinsurance? [A.] It is very difficult to say that, is it not? I mean, I felt it in 1988, as you know, we had some cover and I decided we perhaps needed another $20 million cover. But it is a feel. It is no more than that.

[Q.] It is no more than that? [A.] Not really, no. But a lot of that is - I mean the whole market works on that basis, I think.

[Q.] Why did you not say here, Mr Bullen, like many syndicates writing excess of loss protections,

. . .we have reinsurance to cover our own PML exposure, save for an amount not exceeding about 120 per cent of syndicate stamp?

[A.] Well, I could have said that, but I did not.

[Q.] Why did you not say it, if it was, in fact, the truth? [A.] Well, it was the truth, but it is easy to look back and say what you should have said at the time, but at the time I thought this was quite adequate, what we had said.

[Q.] In giving your supporting Members’ Agents and direct Names - because this went to everyone, did it not? [A.] Well, it should have done, but I am not sure it did.

[Q.] That was the purpose of the document. It was a circular. [A.] Yes.

[Q.] In giving the Members’ Agents, who supported your syndicate, an insight into the way you determined what reinsurance to buy, you say, like many excess of loss syndicates, “We have protections for about three times our income”, and you felt that was the relevant piece of information to communicate, did you? [A.] Well, I thought it was okay at the time, yes.

[Q.] Presumably you were saying that which you regarded as helpful and important, rather than that which you regarded as irrelevant and unhelpful? [A.] I thought it was relevant at the time, yes.

[Q.] This is an important document, is it not? It is a circular to all your supporting Agents? [A.] Yes.

[Q.] Informing them what the position is in relation to a, potentially, very serious market loss? [A.] Yes, although we did not know too much about it at that time, but, yes.

It must be noted, in Mr. Bullen’s favour, that he kept an aggregates book, and from November, 1988 the aggregates were kept on a monthly basis. It would appear that he bought additional vertical reinsurance in 1988 in March, June and September. As he himself said in evidence there was not much point in having a record of aggregates if it was never looked at. I can infer, I think, that Mr. Bullen did look at the book and he certainly decided in relation to XL on XL contracts that the time had come to stop writing any more aggregates after they had reached a certain level. In a general sense, I am sure that the amount of aggregates he had written at rates on line above 4 per cent. played a part in his thinking as to how much vertical reinsurance he should buy. But not in any rational or orderly way.

It seems to me that without any knowledge as to the “in force” aggregates, he had deprived himself of the chance to make such an orderly or rational judgment, even if he had been minded to adopt any coherent plan. The plaintiffs alleged in their amended points of claim [par. 35(2)(b)] as follows:

Nor did Mr Bullen ever monitor his aggregates on an “in force” basis, so as to ensure that he was properly aware of the 100% aggregate exposure to which Syndicate 255/258 was subject at any given time. Such records of the Syndicate’s aggregates as were kept were maintained on a “written” basis, recording only the aggregate exposure from risks written into (in this case) the 1988 underwriting year but ignoring any continuing aggregate exposure from risks written into the prior year (albeit that they also fell to be protected by the same reinsurance programme as protected Syndicate 255/258’s 1988 year of account.

Mr. Bullen dealt with this in his written statement at pars. 42 to 44. It was his case that:

I was always aware that covers written in one underwriting year with contract dates incepting after 1st January were exposed to claims occurring in the successive [sic] underwriting year. Since the majority of the syndicate’s outwards reinsurance programme had 1st January inception dates this would normally mean that any such claims would be recoverable under the succeeding year’s programme (because the policies are written on a losses occurring basis). . .However, I did not at any stage consider it necessary to maintain a specific system to monitor this “rolling” aggregate. This was for a number of reasons. First I was always aware of the shape of the aggregates and the approximate proportion of the book which would have been written with post 1st January inception dates. I was provided with very regular updates from my staff as to the number of risks written, the aggregates within each class, the cumulative total aggregate and the premium income. Until 1988 this information was provided on paper notes which were routinely thrown away. It was only after Piper Alpha that a more formal record was kept, for the benefit of the Agency. Examples of the more formal monthly cumulative aggregate records are appended. . .From these regular updates I would have built up an impression of the extent of aggregate which was written with post 1st January inception dates. . .I would estimate that this aggregate [1st January inception] consistently comprised two thirds to three quarters of the overall book.

Mr. Bullen supplemented his evidence in the witness box. He remembered pencil totals being made each month, before the monthly records were kept. He gave a great deal of circumstantial evidence about this. That evidence was additional to and not included in his written statement, although the statement had dealt at length with the syndicate’s record keeping. Further, this extra evidence was not entirely consistent with the evidence of his deputy. However, it is clear that whether or not I believe Mr. Bullen about this, the fact is that he never had available to him, even after the introduction of monthly records, details of aggregates on an “in force” basis. In my view, as a matter of common sense, and this is supported by Mr. Alexander’s evidence, an underwriter could only calculate his PML exposure by reference to the written aggregates which were in force, or which he was planning to write:

[Q.] Do you accept that an underwriter ought to calculate his PML by reference to his in force aggregate? [A.] Yes. . .

[Q.] Take from me this hypothesis - his Lordship will no doubt make findings from the evidence - until November 1988 when these monthly sheets were produced and kept in relation to the 1989 year of account, he kept, by which I mean, he retained no such sheets. Any such similar documents in previous years were routinely thrown away after he considered them. Just take that as a hypothesis? [A.] Do we also put into it the time at which they were thrown away?

[Q.] I do not think there was any suggestion that they would be burnt on bonfire day, or anything like that. They were not kept for subsequent reference. They were routinely thrown away? [A.] They were not kept for a year?

[Q.] No. My question is this. On that hypothesis, have you seen any material which would enable Mr Bullen to determine at a given point in one year how much aggregate remained in force from the previous year? [A.] No.

The only basis for Mr. Bullen’s judgment was his feel for the shape of the book. If the book had remained static I could understand it. But Mr. Bullen’s book was dynamic: the number of written contracts was increasing, as was the amount of aggregate. The limits were changing, the rates on line were changing and different proportions of different parts of the book would be written at different times. I have no difficulty in accepting Mr. Green’s evidence that, in retrospect the method adopted by Mr. Bullen seemed unattractive:

[Q.] The reason I am inviting you to look at this document is that, if one spans across, for example, the various classes of business and the rates on line within those classes of business - one can see, as one would expect, that the actual proportion of aggregate written in a particular class in any one month differs between rates on line band? [A.] That is quite normal, I would suggest.

[Q.] It is quite normal and on the whole, the higher the rate on line band, the later in the year it will be written? [A.] That is correct, yes.

[Q.] So we have this book of business that is being written by you and by Mr Bullen which, year on year, is increasing in terms of the number of risks written and the aggregate it generates, yes? [A.] Yes.

[Q.] Where the proportion of the business that is written in particular classes and at different rates on line will differ in any given month; yes? [A.] Yes.

[Q.] I think it is also fair to say, is it not, that, as between one year and another, a particular risk may well shift from one rate on line band to another? [A.] It can do, yes.

[Q.] Which will itself affect Mr Bullen’s calculations as to exposure? [A.] That is correct.

[Q.] When all that is borne in mind, Mr Green, surely it cannot be said by any stretch of the imagination that it is satisfactory to rely upon a feel for what the aggregate in force was in the previous year when determining how much reinsurance to buy for a given year in question? [A.] That was the way we looked at it at the time.

[Q.] Would you accept, in retrospect at least, that it lacks attraction? [A.] I think, looking at it seven years later, yes.

In my view, the absence of records which would have enabled Mr. Bullen to carry out a competent PML exercise is good circumstantial evidence to suggest that he never carried out such an exercise at all. Had he intended to comply with his duties, he would surely have appreciated that he could not properly discharge them without having a clear idea as to the level of his “in force” aggregates. While there must be room for the making of judgments, his book was such that any competent underwriter would have made some attempt to record and monitor aggregates on an “in force” basis. What was needed was a record of when contracts incepted (and expired) which could have been achieved by a coding device.

In any event, it seems to me clear from his evidence that Mr. Bullen did not truly understand the PML process and did not purchase vertical reinsurance on the basis of any calculation implicit in the process of working out a PML figure. He accepted the suggestion put to him by Mr. Crane, Q.C. on behalf of the Names that unless he knew his PML in relation to any one catastrophe he would not know how much reinsurance he needed. In answer to my questions, he revealed that he did not understand a PML event:

Mr. Justice Morison: Can I ask you this, to see if I have understood it all? [A.] Yes.

[Q.] You look at a possible event that might occur, which would produce a very large loss, such as the collapse of a platform, for whatever reason? [A.] Yes.

[Q.] That is your PML? [A.] Yes.

[Q.] You then ask yourself the question, is this how you do it, “is that really likely to happen, based on past experience”, and you think to yourself, “From my wide experience of losses that have occurred, I do not think that it is really very likely to happen, so I will not buy reinsurance to cover that particular loss, because I make a discount from it to take account of its likelihood of occurring”. Is that your thinking or not? [A.] Yes.

[Q.] That was it? [A.] Yes.

[Q.] So that, when you say “foreseeable market catastrophe losses” here, or rather, when your solicitors do in this submission, what you are doing is talking about a potential or probable maximum loss, but then saying, “We have applied some judgment as to whether we think it is likely to occur”? [A.] That is the way we worked, yes.

[Q.] Is that how do you it? [A.] Yes.

Mr. Justice Morison: I do not know if that is of any assistance, but I think that is what the witness is saying.

Mr. Crane: Thank you.

Mr. Justice Morison: You must be careful not to agree with me just because I appear to be the judge. [A.] No, that is exactly what we did. I find it very difficult, sir, because I get negative questions and positive questions and I have to think which is which and it is -

Mr. Justice Morison: I understand that. It is a difficult process. If you do not understand any question, I am sure that you will say so, because it is counsel’s duty to make himself clear.

Of course, it is unsound and incompetent for any underwriter to take an event for PML purposes and then make some kind of allowance for the fact that such an event is unlikely to occur.

Mr. Bullen told the Boatman Inquiry that:

Vertical cover was determined by what had gone on before. We always kept a close watch on all the major losses that we had and how far they had affected our reinsurance programme. Obviously you have to bear in mind the actual worst situation. But our programme had always been catered for the losses that we had and we worked on that basis. . .

He had indicated to the inquiry on an earlier occasion that -

. . .it is like anything else, with life insurance, you are rather inclined to work on things that have happened.

But this reveals the fallacy of his thinking: experience of a past period which is relatively free of catastrophe was not a proper guide to the need for reinsurance in the future. As it was put by Mr. Justice Phillips in Feltrim:

It is possible that Mr. Fagan was, to an extent, proceeding on a basis of following past precedent - concluding that an approach which had worked successfully in past years was likely to continue to do so. That is not a satisfactory basis on which to conduct the business of catastrophe insurance.

The market was changing, and the marine market was writing an increasing amount of non-marine business. I do not need to set out the way this came about, but I am satisfied that it was a fact known to the properly informed underwriter in the LMX market and should have been known to Mr. Bullen.

I need only refer to Mr. Outhwaite’s underwriting report published in 1988, and to his paper in 1988, where he said:

. . .in the marine market high level catastrophe rates are absurdly low, and are based solely on comparison and precedent and bear no relationship to the risk being run. Indeed there are many examples where non-marine XL reinsurance is being included within marine exposure and is being placed at a mere fraction of the cost (say around a quarter) of what the non-marine market itself would require, and without any co-insurance warranty. . .

The point was more pithily made in Mr. Emney’s paper at the same seminar when he said:

. . .with the volume of non-marine business now being written in the Lloyd’s marine market, it would probably be more appropriate to describe its inhabitants as incidental marine syndicates.

In relation to the three events which the defendants contend would have given rise to a belief that only “an enormous catastrophe loss” would trigger the spiral, only one, Hurricane Alicia, was properly to be described as of catastrophic proportions. If Mr. Bullen had applied his mind to the question he would have been able to see that his paid claims in relation to Alicia had dramatically increased [84 per cent.] between December, 1986 and December, 1987 and by over 45 per cent. between December, 1987 and December, 1988. These claims were coming through on Mr. Bullen’s XL and whole account covers. This would have been a good demonstration to him both of the fact that an essentially non-marine incident could have a substantial impact on his book and that such claims might “spiral”. In any event no-one knew what size of loss would trigger the spiral and Mr. Bullen could not, and should not, reasonably have derived comfort from past experience.

In calculating the PML amount, there must be included within it all those aggregates which in the reasonable judgment of the underwriter may be impacted by the PML event so as to cause loss. Yet, on Mr. Bullen’s evidence, he ignored aggregates which he knew would be so exposed on the basis that he was prepared to run them net for his account. Thus, Mr. Bullen was confused as to the purpose of the PML calculations. By leaving out of account those aggregates in the 0-5 per cent. rate on line band in his whole account book, because he had already made an a priori judgment that any risk in that band was worth running, he had denied himself the opportunity of doing a PML calculation and then deciding how much reinsurance to purchase. Because of the way he approached the matter, Mr. Bullen never accurately assessed the PML amount nor the Names’ net exposure. The following extracts of his evidence make the position clear:

Mr Crane: Mr Bullen, you gave evidence yesterday as to your PML methodology in calculating your syndicate’s exposure to a worst case catastrophe, as you saw it, produced by a collision between a tanker and a platform? [A.] Yes.

[Q.] You agreed with the PML factors that I put to you in the form of a list, a schedule? [A.] Yes.

[Q.] In the course of cross-examining you on the subject you will recall, will you not, that your PML factors, in relation to whole account covers, covers classified as XG, was to disregard, in assessing your PML, all such covers written at a rate on line of less than 5 per cent? [A.] Yes.

[Q.] You recall all of that? [A.] Yes.

[Q.] The question I have for you is this: on the hypothesis that all that evidence is accepted, my question is on what material did you come to the conclusion that these covers that we are now looking at, written in favour of the Feltrim Marine syndicate, at a rate on line of less than 5 per cent, were not exposed to your PML event? Do you understand that? [A.] Yes.

[Q.] What is the answer? [A.] All I can say is, as I said before, we knew there could be losses at certain levels of that programme, but that was the area that I was prepared to run. We know now that I was wrong, and certainly from this report, which was dated much later, it shows I was wrong. All I can say in the court is that it was how we assessed things at that time.

[Q.] Let me try to explore that. At the time you realised that these covers could, in certain circumstances, be exposed to loss? [A.] Yes.

[Q.] Those circumstances were foreseeable? [A.] Yes.

[Q.] Did you work on the basis that, for example, if a major platform in the North Sea was totally destroyed, as a result of a collision with a tanker, then these covers, that we are now looking at, could be exposed to claims? [A.] They could be exposed, but still well under my own PML.

[Q.] Try to focus on the question. Before Piper Alpha, would you have regarded these covers as potentially exposed to claims in the event that a tanker collided with and destroyed a high value platform in the North Sea? [A.] It could be, yes. I do not know what his PML was, you see.

[Q.] The question is, Mr Bullen, why did you disregard such covers in calculating your own PML? [A.] I did not disregard them. I was prepared to run where I thought there might be a loss.

[Q.] So the calculations you were telling us yesterday were not designed to identify the exposure of your syndicate to a worst foreseeable loss. They were designed to tell you how much reinsurance you ought to buy. Is that the distinction you are making? [A.] For my aggregates, yes.

[Q.] Do you understand the question, Mr Bullen? [A.] I think I did.

[Q.] The calculations you were telling us about yesterday, which you called your PML calculations, you appreciated that covers excluded from consideration in those calculations could, in fact, be subject to claims in the event of a collision between a tanker and a North Sea platform of high value? [A.] Yes.

[Q.] That is right, is it? [A.] Yes, that is correct.

[Q.] But you excluded those covers from this calculation because you decided that you were not going to reinsure such covers; you were going to run that risk. Is that right? [A.] I think that is correct, yes.

And, on the following day:

[Q.] Forgive me for going back to this, Mr Bullen, and I will try not to do it again. Having made this decision, then, to write a whole account cover at a rate on line of less than 5 per cent, which you foresaw could, in certain circumstances, which were foreseeable, be impacted by catastrophe loss, yes? [A.] Yes.

[Q.] You would then proceed on the basis that it did not require reinsurance. Is that right? [A.] No. As I said to you yesterday, you could expect the claims in some layers, maybe, but that was the risk we were prepared to run, yes.

[Q.] Let me try and approach from a different direction. You told us on Day 1 of a series of calculations you did throughout the year, where you applied certain discount factors to the 100 per cent aggregates in different categories. [A.] Yes.

[Q.] And the discount you applied to your whole accounts was a 100 per cent discount - in other words, a factor of 0 in the range of 0 to 5. [A.] Yes, but I also said that that was where I was running my risk.

[Q.] So, let me try and understand it. You appreciated, as I understand it, that those covers could, to some extent, be impacted by loss in foreseeable circumstances. [A.] Yes.

[Q.] But you decided to run the risk, in the sense that they would be covers that were excluded from the benefit of your reinsurance programme. Is that right? [A.] I have always said, all the way through, that that is where I thought we could make our profits. Some years we did.

[Q.] I am sorry to be so laborious about this, but I want an answer, if I can, to this question. When you say that is where you decided to run the risk, what you mean is, it is those covers that you decided you would not protect by reinsurance. [A.] Yes.

What this evidence shows, I think, is, as the plaintiffs submitted, that:

Mr Bullen elected to run a real exposure to a foreseeable catastrophe without attempting to quantify or limit that exposure.

Although the size of loss resulting from the Piper Alpha disaster [U.S.$1.4 billion] was well short of the size of what Mr. Bullen says was his PML loss [U.S.$3 billion, or thereabouts], his PML event was the loss of a North Sea platform [and the loss of a hull]. In those circumstances, after Piper Alpha, I would have expected him to have been saying to RTY and any managing agent who asked, that it was a loss which fell short of his own PML, or to have referred to his own PML event, at least en passant. He did not. Nor did he ever refer to a “planned” net exposure of 100 per cent. of stamp.

It was part of Mr. Bullen’s evidence [par. 173 of his statement et seq.] that the way the insurance for Piper Alpha was placed in the market was unusual and had some bearing on the losses which fell on his Names. He appeared to be saying two things:

1. Because there were four interests in the platform which were insured separately there was a risk of aggregation which would cause a direct underwriter’s rig specific protection to be exhausted thereby exposing his whole account to the loss. As Mr. Bullen put it in his statement:

It ultimately transpired that a proportion of the direct market had become over committed on Piper Alpha which led to a far greater accumulation of risk within the direct market than had been anticipated, much of which circumvented the specific rig XL coverage and became a loss to whole account reinsurers such as Syndicate 255.

2. The majority of the risk had been placed, not on the line slips but on various facultative or broker’s covers. Thus rig specific covers in respect of these line slips were largely unaffected by the Piper Alpha disaster:

. . .thereby bypassing many valuable underlying retentions to whole account reinsurers, which would have helped to take some of the sting out of the loss. [par. 175 of Mr. Bullen’s statement].

When Mr. Outhwaite gave evidence, he said this:

[Q.] Can I ask you this, Mr Outhwaite: on the hypothesis that the London market received a claim of a billion plus in relation to the loss of a platform, which you assume to be the case here, no-one operating in that market could safely assume that the retentions of the direct market would take the sting of that claim with the result that the LMX market would not be seriously or significantly impacted; that is right, is it? [A.] It is right.

Therefore, if Mr. Bullen was surprised that retentions had not taken the sting out of the Piper Alpha loss he was misguided. Further, it seems to me that Mr. Bullen also demonstrated an unacceptable degree of ignorance about the placing of the Piper Alpha risk. It should be borne in mind that the loss of a platform was essentially what he took to be his PML event, so that he could be expected to have some sound knowledge of the rig market. When he gave evidence, Mr. Bullen accepted that hugely valuable installations such as Piper Alpha could be expected to be in joint ownership; and that separate insurance of those interests was not unexpected. He also accepted that the joint capacity of the lines slips would be insufficient to cover the risks of the largest platforms which he believed to have an insured value of about U.S.$3 billion. In the light of these answers it is difficult to understand why he should have been surprised about the way the Piper Alpha rig had been insured in the direct market. In any event, Mr. Outhwaite dealt with the matter comprehensively in his answers to me:

[Q.] Thank you, Mr Outhwaite. Would you give me a moment to tidy up?

Mr. Justice Morison: While you are doing that, perhaps I could ask the next question, which is would you say that any reasonably competent underwriter for a Marine Syndicate, who was writing XL on XL, where his PML event was the loss of a North Sea platform, ought to have been aware of what you were fully familiar with, namely that these rigs were being insured through oil company package policies? [A.] Yes. Of course, if he is a specialist excess of loss underwriter, he would not have the knowledge that a direct underwriter of drilling rigs would have, or be expected to have. That is why the excess of loss market developed the drilling rig questionnaire. It was to get this sort of information.

[Q.] If I asked you the question whether Mr Bullen should have been aware that rigs were being insured in the way that you were familiar with, what would you say the answer was? [A.] He would and should have known that not all the coverage would come from one particular source, that it would be spread, yes.

Finally, although my decision to reject Mr. Bullen as a reliable witness of fact on the main point has been based, largely, upon the matters referred to above, I also formed an impression of his manner and demeanour in the witness box. I would add that my conclusions are reinforced by that impression. Even making allowances for the matters which Mr. Hirst asked me to take into account, Mr. Bullen’s manner and demeanour gave me no confidence that he knew what he was doing as an underwriter or that he was a reliable rapporteur.

In short, Mr. Bullen did his incompetent best but fell well below the standards to be expected of any underwriter who specialized in this market. That said, Mr. Hirst on behalf of Mr. Bullen’s employers elected to call no other evidence. I have, therefore, been deprived of the help which a representative of the RTY board of directors might have given me about their role in relation to the manner in which Mr. Bullen conducted his business. Mr. Bullen had joined Syndicate 255 from being the marine underwriter at the Turegum. On his arrival he found that the syndicate had “no real record-keeping to speak of”. So he introduced a number of systems and changes at the box which greatly improved the ability of the underwriter to monitor and control his book of business. As he said in his statement, and I accept:

I was not provided with any underwriting criteria or guidelines by RTY as to risk management. I was effectively given sole autonomy to conduct the underwriting as I saw fit.

He developed the business of the syndicate and it appeared to flourish in the early 1980s, not perhaps as spectacularly as some of the other syndicates in the same line of business, but he could say, with some pride, that he had developed the portfolio quite considerably. It does not appear that, during the relevant period, his employers ever asked him any pertinent questions about his underwriting practices, he was never asked for any business plan; he was never provided with any instructions or guidance. New potential Names were usually brought down to the Box shortly before lunch and he would have an anodyne conversation with them. He produced an annual report, but, as I understand the position, he never was asked to produce much else until after the two relevant years of account. Names may legitimately ask themselves whether the managing agents did anything to justify their percentage. Mr. Bullen’s failures have been put under the microscope in these proceedings. If he had been more responsibly managed, I have little doubt that he would have acted more competently. I would not wish the criticisms which I have made of Mr. Bullen’s performance as an underwriter to be isolated from the fact that he was as incompetent as he was allowed to be by RTY. I have no hesitation in concluding that RTY are not only liable, vicariously, for the acts and defaults of Mr. Bullen but that they were also in breach of duty as managing agents in the respects set out in par. 37 of the amended points of claim.

2. THE SPECIAL PRIORITY TREATY

I start with the facts.

As part of his [outwards] reinsurance protection for the 1989 year of account, Mr. Bullen entered into a facultative/obligatory cover, the special priority treaty, which was a form of proportional reinsurance. Under the SPT Mr. Bullen was entitled to cede up to 80 per cent. of marine risks written by him to six different participating foreign reinsurers. The six participants were obliged to accept those risks [hence, “obligatory”] provided they fell within the terms of the contract, and, in return, received their proportionate share of the premium.

The six participants were:

NAME OF PARTICIPANT NATIONALITY

Chamber of Manufactures Insurance Ltd [”CMI”] Australian

Constitution Re Belgian

INX Re Puerto Rican

Korea Foreign North Korean

Aneco Re Bermudan

Cie Europeenne de Reassurance [”CER”] French

There were two parts to the treaty. Each related to risks written by Mr. Bullen into his marine excess of loss account (other than specific liability excess of loss): the first part related to risks written by Mr. Bullen at rates on line of between 6 per cent. and 20 per cent. and was with Aneco, who accepted three units [each unit being 10 per cent. of the whole]. The second part was confined to such business which was written at rates on line of between 2 per cent., and 6 per cent. [I ignore the distinction between CMI and Aneco, on the one hand, which were only prepared to accept business at rates on line of at least 2 per cent., and the four others who did not specify such a restriction]. Between them, the other five participants accepted five units; thus leaving Mr. Bullen with at least 20 per cent. of any risk ceded [and 20 per cent. of the premium]. The business ceded at rates on line of between 2 per cent.and 6 per cent. was coded PD, and at rates on line of 6per cent.-20 per cent., PE. The estimated premiumincome of each unit was $400,000 [or, rather more in the case of Aneco]. The table showing whataggregates were ceded by Mr. Bullen under the treaty is on p. 456 post.

Much of the business ceded under the treaty provided for two reinstatements. Allowing for multiple losses, the total aggregate potentially ceded was about U.S.$170 m. The amount ceded to INX Re amounted to about 20 per cent. of stamp capacity, and the amount ceded to Aneco to about 40 per cent. of stamp capacity [rate of exchange U.S.$2/1]. About 44 per cent. [$26 m., approximately] of the total aggregates ceded under the SPT were ceded to Aneco and 22 per cent. were ceded to INX Re. About 90 per cent. of the total aggregates ceded were at rates on line of less than 6 per cent.

The reason why Mr. Bullen entered into this treaty was because he had become concerned about the impact of the Piper Alpha loss on his book, although he, mistakenly, did not expect any serious problem from that disaster. He started to have doubts about the level of exposure to which his Names had been committed by the way he wrote his book of business. In other words, he began to think that he was over exposed and under reinsured. By the end of 1988, claims, in respect of Piper Alpha, amounting to U.S.$9,398,368 had been paid, of which over one-third was paid during that month. By the end of January, 1989, the total paid figure had increased by a further U.S.$2.7 m. and on the basis of claims advised outstanding, including claims paid, the loss was put at U.S.$63,441,936. Mr. Bullen said this, in cross-examination:

[Q.] By the beginning of 1989, it was a conviction, was it not, that Piper Alpha was going to demonstrate that your thinking, so far, had not stood the test of a catastrophe event? [A.] Yes, on that particular risk, yes.

[Q.] Against that background, you needed to off-load, lay off, have reinsured, a considerable proportion of your upper layer aggregate written into the 1989 year of account, did you not? [A.] Well, I thought that was the best way of doing it. I could have stopped writing business at all.

[Q.] Yes, so those were the options? [A.] I said yesterday, but the problem is if you do that, and you have a clean year, then you are bound to get repercussions from somebody as to: “What are you doing? Just think, if you had said all the business you would have had a nice clean year and we would have had a nice cheque.”

AGGREGATES CEDED UNDER THE SPT

[Q.] This is a half measure to stay with the business? [A.] Yes.

[Q.] But to off-load a proportion of the aggregate? [A.] Yes.

[Q.] Would you not be off-loading an equivalent proportion of the premium? [A.] Yes.

[Q.] So, you are not getting the benefit of a payback increase in rates, your Special Priority Treaty reinsurers are getting that? [A.] My Names are getting the 20 per cent we retained, and then they get 20 per cent profit commission if it is clean.

[Q.] Yes, I see. In order to enable you to feel justified in continuing to accept these aggregate exposures, despite the fact that you no longer had confidence that these exposures in the top end were remote from loss, you needed to off-load the bulk of those exposures, did you not? [A.] Yes, because I said I would. I did not necessarily think that all the layers would have losses and certainly lots of the things had not had losses at Piper. The rates were rising and it was - as I say, you were staying in the market. If we had packed up, the Names would have got nothing at all.

[Q.] Mr Bullen, would you have felt justified in continuing to accept all this aggregate exposure during 1989 in the light of the knowledge, the developing knowledge, of what Piper Alpha was doing to your 1988 year of account? Would you have felt justified without the protection of this Special Priority Treaty? [A.] No, that is why I placed the Treaty.

[Q.] So it was the protection which satisfied you that you were justified in continuing to write this aggregate? [A.] Yes.

[Q.] Now, in order to be satisfied, you had to off-load most of that aggregate, did you not? Otherwise you would be retaining it for the risk of your own Names? [A.] We kept 20 per cent, yes.

[Q.] You would not have been able to keep much more than 20 per cent, would you, and still feel justified in going on? [A.] Well, that follows, I think, yes.

[Q.] It does follow, does it not? [A.] Yes.

[Q.] Really you needed about eight units to make going on justifiable? [A.] Yes.

[Q.] Is that right? [A.] Okay, yes.

[Q.] Do not say “okay”, if it is not okay. [A.] I agree with that.

This accorded with the evidence of Mr. Green. Bearing in mind that rates on line had risen dramatically after Piper Alpha, it follows that Mr. Bullen now appreciated what he should have appreciated before that disaster, namely that business written at low rates on line was susceptible to claims in relation to large scale losses.

In his statement of evidence, Mr. Bullen gave two reasons why he chose this form of protection, as opposed to taking out further top level XL on XL cover within the London market: following Piper Alpha the capacity in the market:

Was either saturated by the end of the renewal season or had dried up as the Market monitored the development of the Piper Alpha loss. This made it practically impossible to purchase the additional levels of cover I had in mind. Second, I wished specifically to cede a large proportion of aggregate out of the London market to reduce the propensity of the book to accumulate or spiral.

In relation to the second reason, I think this reinforces my view that Mr. Bullen simply did not understand the market in which he specialized as an underwriter. He took no steps to provide against the possibility that the aggregates were not reinsured straight back into the London market, assuming such steps were feasible, which I doubt. I accept from Mr. Outhwaite that it would have been customary for such participants’ accounts to be reinsured. Therefore, in the absence of some kind of indication to the contrary, Mr. Bullen would have had no reason to believe that the aggregates would not come back onto the London market. But it goes further. In fact, the very same brokers who introduced Aneco to Mr. Bullen, Messrs. Johnson & Higgins, obtained reinsurance protection for Aneco in the London market, and I accept Mr. Forster’s evidence (of Johnson & Higgins) that Mr. Bullen would have been aware of that, and reject Mr. Bullen’s contrary evidence. This is a good example, I think, of Mr. Bullen’s unreliability as a witness. In my view, he had no reason to believe that he was taking aggregates out of the spiral; he knew that Aneco, at least, were reinsuring their book in London; and I very much doubt that he gave any thought to taking aggregates out of the London market so as to diminish the spiral effect by providing “leakage”, or that any such thought formed any part of his reasoning process when he wrote the SPT. His evidence that it did so is another indication, I think, that he was prepared to try and rationalize what he had done with the benefit of hindsight.

The background against which I must reach conclusions may, thus, be summarized: the SPT was considered to be necessary if Mr. Bullen was going to continue writing business so as to make full use of his stamp capacity. There were good commercial reasons for wishing to remain in the market. Rates had increased since Piper Alpha. LMX underwriting is a chancy business with losses, which can be anticipated in some years, being set against profits which may be expected in succeeding years.

There is no reason in principle why an underwriter should not write business on the basis that net losses will be made in some years that are balanced by generous profit levels in other years. [ - Mr. Justice Phillips in Gooda Walker at p. 197, col. 1].

Although Mr. Bullen was ceding a proportionate part of the premium under the treaty, if the 1989 year had been a good one the arrangement would have given Mr. Bullen at least a 20 per cent. share in such premium income, together with a 20 per cent. profit commission. While the SPT was something of a compromise between obtaining additional layers of vertical reinsurance, on the one hand, and writing less than a “full” book, on the other, Mr. Bullen cannot be criticized for entering into the SPT as a part of his underwriting strategy for the 1989 year of account. Indeed, such criticism is not part of the plaintiffs’ case in this action.

However, he was intending to cede large amounts of aggregate exposure under the treaty, mostly at “catastrophe” layers. As the premium rates increased, so the business at rates on line of less than 6 per cent. became relatively less common than in the previous year. Any reasonably competent underwriter would have appreciated that the events which would give rise to claims in relation to risks ceded under the SPT would be those which became exposed as a result of one or more serious catastrophes. In choosing or accepting the securities offered to him by a broker, Mr. Bullen would have had to take account of the risk that covers ceded under the treaty could be impacted by a series of losses. It was not simply a question of making a judgment about the stability of the participant in relation to the level of aggregate ceded on a first loss basis.

It was Mr. Bullen’s belief at the time, as he told the reinsurers, that a loss at least as great as Piper Alpha would be needed before any claims were made in respect of the risks ceded under the SPT.

[Q.] Now, could we go to Tab 21 of the bundle H2, Mr Bullen, and go to page 800 at E? Here the Loss Review Committee are asking you a few questions about the circumstances in which you came to purchase the fac/oblig treaty, or the Special Priority Treaty as we now know it? [A.] Yes.

[Q.] It is in Tab 21, have you got that? [A.] Yes. Page 803.

[Q.] At the top, Mr King is asking you questions and he is saying:

Whilst I can understand you getting business out of the market. . .

That is a reference to wanting to place - [A.] Sorry, just a moment.

[Q.] Page 800. [A.] Yes.

[Q.] Mr King is saying at B:

Whilst I understand you want to get business out of the market. . .

That is a reference to you wanting to place this Treaty abroad? [A.] Yes.

[Q.] “It was obvious that the security that might write this”, this being a fac/oblig, “would not be of the very best.”

You say “yes”? [A.] Yes.

[Q.] Pausing there, it is right, is it not, that the sort of foreign reinsurers who would be prepared to write a fac/oblig treaty might not be of the front line security? [A.] Well, yes, that is true. Of course, a lot of them were writing business themselves.

[Q.] Yes. [A.] So they would not be interested in a cover of any kind, they would rather do their own. We had to find someone who was interested in the type of business, but not prepared to write it themselves.

[Q.] Such folk, you accepted from Mr King, might not be of the very best in terms of security? [A.] Yes.

[Q.] Then he goes on to say:

And we are looking at the high rate on line and the high rate on line business are the ones which I am sure you would agree would only be hit on a very large major loss where everyone is suffering.

You say “yes” to that? [A.] Yes.

[Q.] That is obviously right, is it not? [A.] Yes, that is what I told my reinsurers.

[Q.] Therefore, you might think, the higher up you go the better the security would be, would you think that is a correct statement.

Pausing there, what it would appear Mr King is putting to you was this: do you not think that the higher up your programme, in terms of last response, you go, the more confident you would need to be about the value of the security. Would you accept that? [A.] Yes. Within reason, yes.

[Q.] Yes, within reason? [A.] Well, the problem is if you use a bad company at the bottom end, you are going to put them out of business if you do that, or you could do, so you have to weigh up the two between.

[Q.] Yes. The quality of security is always important, is it not? [A.] Of course.

[Q.] I think the proposition that was being put to you in that sentence, if I may re-frame it is that those reinsurers that will be expected to respond only in circumstances where there is a major market catastrophe and where, if you like, the fortunes of syndicates may be on the line, those are the reinsurers you would wish to feel particularconfidence in; is that right? [A.] Yes.

Mr. Justice Morison: I suppose the other reason why you want sound reinsurers is because you may be calling on them at a later rather than an earlier date? [A.] That is true, sir, but you are hoping you are not going to call on them at all, because it has to be a really major - I actually told my reinsurers that it would have to be something like at least Piper that they would have a claim there.

Mr Justice Morison: Yes. They might be having to fork out money many years after the incident? [A.] Yes.

Mr. Justice Morison: So you want a reinsurance company which has got a fairly lengthy expectation of life? [A.] Yes.

Mr. Crane: You accept that, Mr Bullen? [A.] Of course, I cannot disagree with that.

[Q.] At the back of your mind was the belief, was it not, that it was unlikely that they would ever be called upon at all? [A.] Well, I have said to you, there was this nagging feeling that I had that maybe we would get another one, which is why I took it out. If I was confident that there would not have been another one, I would not have bothered, and I would have run the risk, but yes, the nagging was there, but I hoped there would not be another one. Certainly I did not hope there was going to be - we had seven in 1989.

Every competent underwriter would also have appreciated that were such events to occur, the reinsurers would be being asked to respond as a matter of “last resort”. As Mr. Outhwaite put it, the survivability of such a reinsurance company would have been a dominant consideration in choosing participants in the SPT:

[Q.] Would you not accept this, that the more questions the underwriter has regarding the likelihood of the reinsurer’s ability to pay claims, the more likely he will be, if he uses them at all, to confine the use to which he puts him? [A.] I do not follow that argument, because if you are talking about there being a significant time-lag, there is no point in putting any reinsurance with a company if you do not feel reasonably confident they are going to pay when a claim arises. It is not a point of quantum at all.

[Q.] Very well?

Mr. Justice Morison: Survivability of a company is not just a consideration; it is a dominant consideration? [A.] If you are talking about the business where claims will appear many years after the declaration.

[Q.] Yes. With regard to Aneco, the brokers specifically provided Mr. Bullen with information about that company with a view to him making up his own mind whether he wanted to use them, and if so, to what extent, as this extract from the evidence of Mr. Forster makes clear:

[Q.] Could you go forward to page 894 of the evidence that we are looking at? You are asked at E - this really echoes a written answer you had given:

Perhaps standing one step back from that, can you tell me what your views would be on the importance of security of a reinsurer in the place that Aneco Re would have in 255’s programme the top end cover.

Do you see that? [A.] Yes.

[Q.] You say:

First of all securities are obviously very important, particularly in the top end, because there will be a time-lag before the losses actually impact the top layers and you have to be pretty confident that that company is going to be around to pay the losses when those losses actually impact the contracts. Although, as you say, all security is important, if anything, the further up the programme you go, the better the security should be.

Answer: I would suggest it should be looked at very closely.

So you are telling the Loss Review Committee that, while security is always important, it is even more important at the top end of a programme such as this? [A.] Yes.

[Q.] And that was your view at the time? [A.] Yes.

[Q.] And it was for that reason, was it not, Mr Forster, along with the volume of the aggregate that you anticipated Mr Bullen ceding to Aneco, that you left to him the decision as to whether or not to use Aneco, and if so, for how much? [A.] That is correct.

[Q.] And that is what you are saying at F to G:

Did Mr Bullen actually ask your advice on the quality of the security of Aneco, or did you provide him with information and allow him to make his own judgment?

Answer: we provided him with information as to allow him to make his own judgment.

Do you see that? [A.] Yes.

[Q.] Is the picture this: that you left it to Mr Bullen on the information you gave him to reach his own judgment as to whether he regarded Aneco as appropriate for the purpose he had in mind? [A.] Yes.

[Q.] And would you accept this: that while the inclusion of Aneco on your company’s marketing listmeant that you were happy to proffer them as a market, it was for Mr Bullen to reach a judgment as to whether Aneco were appropriate in the context of the amount of aggregate that he intended ceding to? [A.] Yes, I would.

[Q.] Would you accept that? [A.] Yes, I would.”

Against that background, I turn to the first issue:

What, if any, is the duty of a Lloyd’s underwriter to satisfy himself as to the adequacy of the security offered to him by the broker who acts on his behalf in placing his outwards reinsurance arrangements?

On this issue I had the benefit of expert evidence from a Lloyd’s broker, Mr. Alington, for whose assistance I am grateful. He was, if I may say so, a thoughtful, well-informed and capable expert witness. I place great weight on his evidence on this and the other SPT issues. He accepts that it is the duty of a broker to satisfy himself that the security put before an underwriter was reasonably appropriate to the type of business written; for example, he would have regard to whether it contained a high proportion of long tail business. But the broker would not be concerned with the finer details of such a book of business. Brokers such as his own company, Bain Hogg, would have a security committee [of which he had been chairman for a number of years] and would maintain a list of approved securities. That list would be prepared from information which the brokers had acquired over the years from various sources, including Standard and Poors and ISI rating information. He accepted that a broker owed a duty to the client, whether he be a professional underwriter such as Mr. Bullen or someone with no such expertise.

In many ways, Mr. Alington’s evidence convinced me that the defendants were right in this case to accept, also, that Mr. Bullen himself [and RTY] was under a duty to satisfy himself that the security offered was appropriate for the purposes he had in mind. The broker could not be expected to know the details of the underwriter’s book of business. The list of approved companies prepared by the broker was compiled for general use, although some would be more appropriate than others in some instances. But all the companies on the list would have been approved as sound, in principle. It would be quite inappropriate for a broker to maintain within that list different tiers or levels of soundness. The underwriter, on the other hand, knows better than the broker the nature of the risks which he is writing and the length of the tail, if any. He will know how much risk he intends to cede, and he will have been able to evaluate the nature of those risks and the likely impact on them of foreseeable events. The roles of broker and underwriter neatly complement one another: in normal circumstances, an underwriter would be entitled to assume that the broker would only have presented him with a prospective reinsurer which the broker considered to be generally sound, and the broker can assume that a competent underwriter will make a judgment as to the appropriateness of that security having regard to his particular needs. However, there may be circumstances which alter the general position: if, for example, a broker made it clear that he was giving no “comfort” to the underwriter about the reinsurer he was putting forward. When satisfying himself that the security is suitable for the particular purpose he has in mind the underwriter will wish to pay particular attention to those aspects of “soundness” which will be most challenged if foreseeably disastrous catastrophes were to impact his book. Thus, the underwriter will look carefully at the “size” of the reinsurer, having regard to the amount of business he intends to cede, and to the reinsurers’ “expectation of life”, having regard to his knowledge of the likely length of the tail. The names on the brokers’ list are “off the peg”; whereas the underwriter must tailor the material to meet his own special needs.

My conclusion that Mr. Bullen owed a duty of care is re-inforced by a number of factors:

1. Mr. Bullen accepted in evidence that he had a responsibility to satisfy himself that the reinsurer offered to him was reliable for the purpose for which he wished to use it:

[Q.] Perhaps we can come on to Aneco in a moment. Mr Bullen, you recognised, did you not, your duty as active underwriter of Syndicate 255 to satisfy yourself that the Syndicate’s reinsurers were reliable security? [A.] Yes.

[Q.] The mere fact that a broker proffers a reinsurer to you does not absolve you in your own mind of any duty to satisfy yourself that it is reliable, does it? [A.] No, I have never complained - I would hope the broker would only give me security that he thought was okay, but it is my responsibility, yes.

[Q.] That was a responsibility, as I understood it, which was entirely yours personally, in the agency, in the sense that the board did not have a security committee or involve itself in such matters; is that right? [A.] They had not got a clue.

[Q.] “They had not got a clue”, I see. So, the fact that Mr Forster suggests Aneco for this Treaty did not, as far as you were concerned, absolve you from the duty to ensure that Aneco were reliable for the purpose you intended to put them? [A.] From Aneco’s point of view, I had their security rating and I also met the underwriter.

[Q.] Pause there. Is the answer to my proposition “yes”, that you regarded yourself as under a duty to satisfy yourself that Aneco and the other reinsurers were sufficiently reliable for the purposes you intended to put them to? [A.] Of course, otherwise I would not have accepted them.

[Q.] Had you thought the security was not sufficiently reliable for the purpose to which you intended putting it, you would have regarded yourself as duty bound to take it off the cover note? [A.] Yes.

2. More importantly, Mr. Wilshaw, an experienced underwriter, and Mr. Outhwaite, both of whose evidence was also of great assistance, accepted that the underwriter was under a duty to satisfy himself that the security would be adequate for the purpose for which he intended to put it. Mr. Wilshaw also said, and I accept, that a competent underwriter would think in terms of placing large lines with a large strong security and smaller lines with a somewhat weaker and smaller security. In other words, the underwriter would rank reinsurers; whereas the broker would not. His evidence was consistent with Mr. Outhwaite’s view [see the passage from his evidence cited below] that the underwriter would fulfil his duty by “feel” and “judgment” rather than by any formula or rule of thumb. All this seemed to me to be good sense.

3. The Names were expressly led to believe that Mr. Bullen had exercised judgment on these matters. In his underwriters’ report dated May 19, 1988 Mr. Bullen wrote:

Reinsurance arrangements

All our covers are placed by experienced brokers. The security of our reinsurers is mainly other Lloyd’s syndicates or London Market companies. Any other security has to be approved by us before being used.

The case advanced on the pleadings on behalf of RTY and the members’ agents to the effect that -

Mr Bullen reasonably relied upon [the brokers] to ensure that the reinsurers with whom each line of the Treaty was placed were of a financial status and/or creditworthiness and/or reliability sufficient to meet any claims. . .

is without foundation. However the defence came to be put forward, it is to be noted that in their closing submissions RTY, with whose submissions the members’ agents defendants agreed, accepted that the underwriter/managing agent owed -

. . .a duty of care to Names on his syndicate in the selection of reinsurance security.

Once again, Mr. Bullen was left alone to make his own judgments in his own way, entirely unsupervised or assisted by his employers. RTY have thrown him to the wolves on this part of the case, by running a defence with which, so far as one can tell, Mr. Bullen has never agreed.

Accordingly, I find in favour of the plaintiffs on the first issue under this head of claim: Mr. Bullen was under a duty to satisfy himself that the security offered to him was suitable for the purpose for which he intended to use it.

The second issue is whether there was a breach of duty. Here, it seems to me that I should bear in mind the following factors:

1. Managing agents/underwriters do not have the resources, human or otherwise, for making as detailed a review of the soundness of reinsurers offered to them as do the brokers. Only since 1994 has it been a Lloyd’s requirement that managing agents have security committees, although this fact of itself would not have convinced me that they were not in breach of duty by failing to have one before that date. However, I consider that on the evidence such a failure was not of itself a breach of duty in 1988/1989.

2. The broking houses will be likely to have personnel within them, such as Mr. Alington, who are quite at ease in reading financial statements and in understanding the niceties of classifications whether by Standard and Poors or ISI. Underwriters and managing agents were less well equipped at the relevant time.

3. The brokers who placed this business were all respectable firms; and although Mr. Bullen was introduced to INX Re through his brother who worked for one of the brokers [Hanson & Roberts], that fact does not itself render the transaction suspect.

I deal with three participants only, as the plaintiffs’ complaints were, finally, limited to them: INX Re, Aneco & Korea Foreign. There was a claim by the Names relating to CER, but, for reasons which do not need to be spelt out, the parties agreed that all issues relating to this participant should be stood over.

INX Re:

Mr. Bullen’s evidence was as follows:

[Q.] INX Re were a company about which you knew nothing; that is right, is it not? [A.] Very little, yes.

[Q.] What did you know about them? [A.] That they were operating in France, and that is why he asked the underwriter to come and see me.

[Q.] When you say they were operating in France, they had an underwriting agency in France? [A.] Yes.

[Q.] In other words, they had given someone in France authority to bind risks on their behalf? [A.] Yes.

[Q.] Did you know where they were incorporated? [A.] I think I did, yes.

[Q.] That was Puerto Rico, was it not? [A.] Yes.

[Q.] Did that give you much cause for confidence? [A.] That is why I asked the underwriter to come and see me.

[Q.] Their underwriting agent in France was a Mr Dominic Testue, is that right? [A.] Yes.

[Q.] When he came to see you, what did you talk about? [A.] The place of the Treaty and his companies.

[Q.] His companies? [A.] He was writing for two companies.

[Q.] What was the other company he was writing for? [A.] Not CER, the other French company.

[Q.] Anyway, he had an agency for another principal? [A.] He had two companies he was underwriting for.

[Q.] What did he tell you about INX Re? [A.] He said they were a sound company and he had authority to underwrite on their behalf and I particularly asked and said, “Well, what about funds?” He said, “I have funds in Paris for settlement claims.”

[Q.] Was that a premium reserve, those funds? [A.] I do not know, sir.

[Q.] You are familiar with the practice of underwriting agencies, of retaining a reserve out of premium, otherwise due to their principals, to meet claims as they arise on business? [A.] Yes.

[Q.] Is that a practice you are familiar with? [A.] Yes.

[Q.] These funds in Paris, could they have been a premium reserve of that sort? [A.] It could have been, I suppose, yes.

[Q.] What were they, Mr Bullen? [A.] I do not honestly know.

[Q.] How big were they? [A.] Again, I do not know. He was quite convincing that he was okay to underwrite this business. I took him on face value. He seemed to be a very nice person to me. As you know, he was a Name at Lloyd’s and, as far as I was concerned, he was writing in France. France, in my view, had always had fairly good restrictions on their agencies and I thought it was okay.

[Q.] What was the scheme of regulation enforced in France in late 1988, early 1989? [A.] I do not know offhand.

[Q.] It did not exist, did it? [A.] I thought it did.

[Q.] But you did not know what it was? [A.] No, it is like a company underwriting in London. If they are DTI companies, they should be all right.

[Q.] You did not know whether or not there was a scheme of regulation in France in force at the time which required overseas companies to establish reserves as a condition of authority to conduct business in France? [A.] No, I did not know that.

[Q.] You assumed there was? [A.] I assumed there was, yes.

[Q.] Now, he told you that INX Re was sound; is that right? [A.] Yes.

[Q.] Did you say, “Fine, I will see their balance sheets and audited accounts for the last. . .” [A.] I think I asked for that at one time but he did not have any at the time.

[Q.] You did not get it later? [A.] No.

[Q.] You saw absolutely no financial information of any sort relating to this company? [A.] No, I took him on face value.

Although the defendants accept that INX Re were not, in fact, acceptable security for the SPT, they nonetheless maintain that in the light of the evidence to which I have just referred, and in the absence of any warning from Hanson & Roberts, or their US parent:

Mr Bullen was not negligent to take at face value Mr Testue’s characterisation of INX Re as a sound company with funds in France for the settlement of claims.

Were such an analysis correct, it would reduce, so as to extinguish it, the underwriters’ duty to satisfy himself as to the soundness of a security offered to him. Mr. Bullen had no information on which to make any kind of sensible judgment. He knew he ought to see some figures but was never shown any. Ultimately, he based his decision on the fact that the gentleman from France seemed very nice. That, I think, is a good example of Mr. Bullen’s failure to recognize the importance to his Names of the decisions he was taking on their behalf. In my judgment, no competent underwriter could have concluded that INX Re was a reliable enough security for the SPT. Mr. Alington did not support the defendants’ position:

[Q.] Do I correctly infer from your report that you have never heard of Mr Nadjar until these proceedings? [A.] That is correct, yes.

[Q.] Have you been able to find out anything about him since then? [A.] I have not made any enquiries about him.

[Q.] In short, this security was hopeless, was it not? [A.] It was not particularly good, no.

[Q.] If an underwriter, who had had it left to him to make his own assessment of the security, formed the view that the security was acceptable upon the basis of a meeting with the underwriter where he took the view that he was a nice man, and that he was told that there were funds in France, albeit he never established what their nature was or how much there were, in so far as you can assist the court at all on matters within your experience, he will be falling well below what one would expect of a competent underwriter, would he not? [A.] I would not regard that as a defensible position, no.

[Q.] Would you also agree that, to the extent, as it were, that reliance were placed upon the company being subject to French regulation, then that also would have been misconceived in 1988 and 1989, because purely reinsurance companies were not at that stage, were they, subject to French regulation? [A.] That is correct.

It was, I think, unfortunate that Mr. Alexander, an underwriting expert called on the defendants’ behalf, should have sought to support this part of the defendants’ case. He was, otherwise, a reliable and straightforward witness and I am grateful for his help. It seems to me quite clear that no competent underwriter could have accepted INX Re as a participant in an important treaty which to a significant extent governed Mr. Bullen’s underwriting strategy for 1989.

Aneco

I can deal with Aneco quite shortly, despite the amount of evidence which I received. Aneco were given an ISI rating of A5. Mr. Bullen regarded this company with some suspicion initially but was of the view that in the light of that rating it was safe for him to accept it for the purposes he had in mind. He met Aneco’s underwriter. Otherwise he knew nothing of the company. In my view his judgment was flawed because he simply did not understand the rating system and never bothered to ask anyone to help him, as this passage in his evidence shows:

[Q.] Let us go to the security rating. Now, they had a security rating from ISI of A5. What does that mean, Mr Bullen? [A.] It is one of the better ratings.

[Q.] What is the scale? [A.] A plus, and up and down, or B it can go down to, C.

[Q.] If you have to translate A5 into words, what are the words? [A.] It was a medium-sized, well-respected company.

[Q.] Did you know how to use the ISI security ratings? [A.] I had seen them before, yes.

[Q.] Did you have your own guide as to interpretation of them? [A.] No.

[Q.] Did you receive a guide to interpretation of them from Mr Forster when he proffered the rating itself? [A.] I do not remember having one, but I might have done, or I might have had it later, yes.

[Q.] Did you know what the scale 1,2,3,4,5,6, meant? [A.] Yes I think I did. 5 was quite a reasonable rating, it was not the top, but it was -

[Q.] What is the top, 6 or 1? [A.] 1.

[Q.] What does 1 mean, does it mean big or small? [A.] Big, I think, but I am not sure about that.

[Q.] Did you know then, Mr Bullen? [A.] I knew about the ratings, yes.

[Q.] Did you know what the scale 1 to 6 meant in terms of relative size at the time when you looked at this rating? [A.] No, I am not sure that I knew about actual size, no.

[Q.] So, A5 could have meant big as far as you were concerned, bigger than 1, or it could have meant small, smaller than 1, and you did not know which. [A.] The fact it was an A5, I was quite convinced it was okay.

Mr. Justice Morison: Just pause there. I think you may be at cross purposes and I want to clear this up. Did you know that the numbering after the letter was an indication of size, or did you think it was, on a scale of 1 to 10, how good it was under the letter A? [A.] I think I said it was a medium-sized respectable company. 5 was.

Mr. Justice Morison: It may be I that am being silly, but I would like an answer to my question. Did you think that the numbering against the letter was an indication of the size of the operation or did you think that it was an indication of the goodness of the operation, in other words, on a scale of one to 10, A9 - [A.] I think it was a little of both actually, sir.

(12.15 pm)

Mr Crane A little of both? [A.] Yes.

[Q.] Where does that leave us, Mr Bullen, a little of both of what? [A.] Well, as I say, of it a medium-sized, respectable company.

[Q.] What was the index of size in its rating? [A.] I am not sure I know that, to be honest.

[Q.] You were not aware that 5 was a size indicator?[A.] I knew it was a medium size, but I would not know what size -

[Q.] If 5 was a medium size, what was small size, as far as you understood? [A.] It would probably be B if it is very small.

[Q.] It would be B. You were familiar with ISI as a company of a security rating service? [A.] I had seen it before. We did not usually have to use it because most of our reinsurances were, as you know, placed in London.

[Q.] Well, is the fact of the matter that you have never had to use it before? [A.] No, I had seen it before, we had seen it at the Turegum.

[Q.] At the Turegum? [A.] Yes.

[Q.] Had you ever looked at these security ratings in your time as active underwriter of Syndicate 255? [A.] I had not had to at that time.

[Q.] So the answer is “no”, you had not, had you? [A.] All right, sir, yes, okay.

[Q.] Is that right, Mr Bullen? [A.] Yes, because I had not had the -

[Q.] Because you had been reinsured with London - with Lloyd’s or London- approved companies? [A.] Yes.

Mr. Alington, who understands these matters better than Mr. Bullen could or should have done, said this in cross-examination:

Mr. Moriarty: Let me try a slightly different tack to see if we can at least agree on this, a hypothetical question: if a person were looking for what one might call a well respected and medium size company for security, on any view, Aneco would not really fit that ticket, would it? [A.] I do not think you can say they are not well respected, but they are not medium sized.

[Q.] On size alone they would fall on that hurdle? [A.] They are not medium sized.

[Q.] They would fall on that hurdle, full stop. Can you really call them well respected, when not even widely known? [A.] There is a sort of fine line between well-known and notorious. I am not quite sure whether being well-known is necessarily a good thing. I have no adverse information on Aneco’s reputation. I was doing the business at the time.

[Q.] Having an absence of information, that would reflect badly upon someone, is different from saying they are well respected, is it not? [A.] Yes. I cannot say positively that I have information, because, as I say, we did not deal with them. Equally I was not aware of any rumours in the market that Aneco were not well respected.

[Q.] Let me try to rephrase the question slightly. If someone had decided what they were looking for was a medium sized and well respected company and they looked at the ISI document and said, “A5 is what it says, that fits the ticket”, they would be misleading themselves somewhat, would they not? [A.] Somewhat, yes.

[Q.] And actually less than competent? [A.] Certainly as a matter of opinion, I would say that Aneco was not a medium sized company in terms of either that category or the market generally. Again, size is not absolutely everything. As I was trying to explain early on, once you have passed that threshold of minimum size, then, by far the most important thing is ratio. All the other criteria are ratios, not absolute numbers.

[Q.] I am not going to interrupt when you make a point, I do not want to cut you off, Mr Alington, but that is going over ground I am not dealing with. What I am saying is the hypothesis is the underwriter has set himself the objective of having a medium sized, well respected company. If that is the reinsurer he has in mind to place his business with, and he satisfies himself upon the basis of an A5 rating that that is what he is getting, he is misleading himself? [A.] Yes, if he is specifically looking at medium sized. I think an A5 rating does give you an indication that it is probably a well respected company. In terms of the 5 rating, it is not of medium size.

[Q.] To that extent at least we could agree he would be acting less than competently in drawing that conclusion from that particular rating? [A.] Yes.

It seems to me that making all due allowances for the fact that a broker has greater expertise than an underwriter, any reasonably competent underwriter would have made a simple inquiry to check whether the rating which he saw gave him the comfort which he was seeking. Aneco was expressly not being warranted by the brokers as suitable security: their information was given to Mr. Bullen so that he could exercise his own judgment. In these circumstances, I would have expected him to have made some limited inquiries to establish whether Aneco was the medium sized company he was sensibly looking for, having regard to the amount, and nature, of the aggregate which he intended to cede to that company. Aneco took the largest share under the SPT: their ability to manage such aggregates and their stability were matters of importance to any underwriter in Mr. Bullen’s position. He ought to have realized the risk to which he was exposing his Names in the event that Aneco was unable to meet its obligation. As I have indicated, the amount of exposure ceded to Aneco was substantial, particularly having regard to the possibility of reinstatements. The passages from the evidence cited above shows, I think, that Mr. Bullen was quite out of his depth and was attempting a task which, frankly, he was not competent to perform. Although apparently a respectable company, Aneco was not of the size which Mr. Bullen was looking for. The papers accompanying the ISI rating showed that Aneco was making underwriting losses in three of the past five years; as a proportion of its net premium income, its technical reserves were only 7 percentage points above the minimum which Mr. Alington would have been prepared to accept; its liquidity had been deteriorating since 1983; it was a small company by any standards, as the size of its shareholders’ funds showed and its capital base would have been eroded if Aneco’s parent reneged on its obligation to repay the debt it owed Aneco. This was a long way short of being the sort of size of security that Mr. Bullen was looking for, and mistakenly believed he was getting. In any event, it seems to me that Mr. Bullen should have been on guard as a result of the brokers expressly making it clear that they took no responsibility for Aneco’s suitability as security. Although he believed that his employers were useless, it might have been better had he consulted with the board of RTY before he made the contract. In fact, it was quite some time before Mr. Bullen appears to have made any mention of the SPT.

Thus, although Mr. Bullen was properly entitled to take the view that Aneco was a reasonably respectable participant in the treaty, he ought not to have regarded them as a medium sized company which was suitable for taking a 30 per cent. share in the treaty. The amount of aggregate ceded to it was, in my judgment, greatly in excess of what any competent underwriter could properly have ceded to a company of that size and financial standing. This conclusion is consistent with Mr. Green’s evidence.

The plaintiffs’ pleaded case is that:

. . .the quantity of exposure ceded to the special priority treaty by way of declarations to it far exceeded the amounts which a competent underwriter would have relied upon the treaty to protect given the identity and financial strength of the participating reinsurers making up its security.

In their further and better particulars the plaintiffs say this [File A1 p. 327]:

Nevertheless, leaving aside INX Re (where the Plaintiffs’ contention is that no exposure would have been ceded by a competent underwriter), the Plaintiffs do contend that a competent underwriter would not have ceded a PML exposure of more than between 5%-10% of the Syndicate’s stamp capacity to each of the reinsurers on the special priority treaty.

In their closing submissions, the defendants rightly comment upon the contradictions between Mr. Thompson’s oral evidence on the plaintiffs’ behalf and what was said in his written report and in the plaintiffs’ pleadings, to which I have just referred. The conclusion they ask me to draw from all this, is that the figure of 10 per cent. [or 5 per cent.] is just a figure “plucked out of the air”. They ask me to say that this is so much a matter of feel for the underwriter that I cannot properly conclude that Mr. Bullen fell below the standards of competence in the judgment he made, and they rely upon a particular passage in Mr. Outhwaite’s evidence:

[Q.] For example, say you were considering whether or not, and if so to what extent, to accept as a reinsurer one of the companies that was neither Lloyd's, ILU, or a foreign household name. How would you approach the question of the extent to which you would use this reinsurer, the amount of exposure that you were willing your syndicate should run to that particular reinsurer? [A.] Well, I do not think it is possible to make a sensible assessment of that kind. It seems to me the judgment is to whether you find that they are acceptable security, or not. I am not saying there are not limits to what you would give them, because obviously you can get to a point of absurdity, but generally speaking, what you are looking for is security, which you are persuaded is satisfactory.

[Q.] You would not want your syndicate, and its fortunes, completely dependent upon, for example, two foreign companies that were not substantial, well-known, household names of the sort to which you have referred? [A.] That is why I say you can take it to absurdity. Then of course the answer is obviously, no. What I am saying is you have no touchstone, there is no way of calculating what exposure it is reasonable to give a particular reinsurer. You cannot possibly have the facts.

[Q.] Say -

Mr. Justice Morison: What facts do you not have? I have not quite understood. [A.] If you are convinced that a particular company is a sound company-

[Q.] It is a small sound company? [A.] A small sound company, you may know what the capitalisation of it is. You may know what the policy holder’s surplus is, but that tells you nothing about what it is actually accepting from everyone else. You do not know what the exposures are. Implicit in accepting that security is the belief that they manage the affairs of the company properly, as any syndicate or London company do, and, whatever business they write gross, there is no point in relating that to the capital. In the first place you do not know what it is, because you do not know what they are writing. You know what they are writing to you, but you have no idea what they are writing to anyone else. You do not know what they are doing in order to manage that exposure. You just have to accept the fact, if you are going to take them, that they are reputable, small or large, and that they manage their affairs properly.

[Q.] Presumably, the amount of business that you would cede to them would be, to some extent, dependent on your perception of their size and reliability. “They are reliable enough, they are small, they are in a very small way of business, so I will not cede very much business to them, because I do not want to place overmuch reliance on them.” That might be the sort of thinking. [A.] I suppose it is possible that an underwriter may think along those lines, but he has no method of calculating what exposures it would be safe to give the company. That is the point I am making.

[Q.] I follow that. [A.] I think it may well be, as I have said, that he may say, “This is a small company,” or, “I think it is fine, I will use them for security, but I will not use them for a great deal. I do not mind if they write 10 per cent on this layer of my excess of loss”, or something, “but I do not want them to write 100 per cent.” They may make that sort of judgment, but there is no mathematical relationship between this. You cannot possibly say, “Yes, it is all right for them to write $5 million of aggregate exposure, but it is not for 15”. This would be a matter of judgment of how the underwriter feels about it, it seems to me.

Mr. Crane: To look at it in broad terms, it would be appropriate, were an underwriter to identify a foreign company as small, apparently reputable, but by no means having the profile or status, or size, of a household name. It would be appropriate in those circumstances for the underwriter to say, “I will limit the line or the number of layers to which this particular reinsurer will subscribe”. [A.] He may do that, but it is just a matter of feel. There is no possible way in which he could calculate and say, “I think, because I perceive that the capital and surplus of the company is X, that I therefore will only give it so much exposure”. There can be no precise relationship of that kind.

[Q.] But if we confine ourselves for the moment to conventional excess of loss reinsurance, there will be a category of reinsurers, will there not, where the underwriter says, “Fine, I will have him on my security list, but on condition that his subscription is limited to the following lines on the following layers, I do not want too much exposure to this layer”? [A.] Yes, for a whole variety of reasons that may be right, yes.

I understand Mr. Outhwaite to be saying that there may be circumstances in which an underwriter may limit the amount of protection sought from a particular reinsurer, but there are no hard and fast rules which constrain the way such a judgment is made. I do not understand him to be saying that there may not be circumstances where a judgment must be made: he is emphasizing that such a judgment must be respected for what it is. I fully accept that a Court should be slow to criticize and hold liable a professional man who has made judgments which cannot readily be tested against absolute standards. However, an underwriter is not above the law; and he cannot escape legal responsibility for the consequences of his actions by saying “I did this by feel and you cannot therefore judge me”. I do not regard this passage of Mr. Outhwaite’s evidence as preventing me from forming an objective judgment, based upon the whole of the evidence, as to the extent to which a competent underwriter could properly have ceded risks to Aneco, even though such a judgment itself may involve an element of “feel”. Making due allowance for the fact that the amount to be ceded is a matter of “feel” and judgment for the underwriter, I have no difficulty in concluding that any competent underwriter would not have ceded to Aneco, having regard to its size and financial standing, a PML exposure of more than 10 per cent. of the syndicate’s stamp capacity, taking into account the total exposure ceded, and allowing for the possibility of multiple losses.

Korea Foreign

This entity is the state insurance company of North Korea. Its assets are controlled by the Government of North Korea. It, therefore, depended for its continued existence upon the willingness of that government to keep it in funds in the event of bad results in any given year. The company has been a client of Bain Hogg for many years and on its approved security list for many years. The company had a conservative ratio between their capital funds and net premiums written. It had an ISI rating of A. The downside to using this company was the risk that the government would cease to support it; and the potential problem of foreign exchange which might make it slow to meet its liabilities. Some brokers were more cautious about using Korea Foreign than Bain Hogg, who overcame the exchange control problem through the medium of Korea Foreign’s Paris Office and, in any case, were able to set-off amounts owing from the company against amounts due to them.

On the basis of this information alone, I feel quite unable to conclude that Mr. Bullen fell below the standards to be expected of him in accepting this company as suitable security. If he had asked Mr. Alington for his views, at the time, he would have been given a favourable response. That seems to me to be an end to the matter, despite all the various points which have been urged on me in the plaintiffs’ submissions. Accordingly, I do not consider that the case for the plaintiffs in relation to this company can succeed. I reject it.

4. THE PSL SSUE

Personal stop loss insurance is a form of proportional insurance which gives the insured protection against the specified aggregate of losses in excess of a deductible. They were increasingly used by Names in the Lloyd’s market during the 1980s. There were mixed views about the need for such policies and about their effectiveness. But this case is not concerned with those issues. Mr. Bullen participated on a number of PSL line slips, but he did not record the aggregate exposure from this participation nor did he consider the question of aggregation.

The allegation in the amended points of claim is that:

. . .in underwriting the amount of PSL business he did, particularly in conjunction with the Syndicate’s participation in LMX business, Mr Bullen acted negligently, and aggravated still further the losses to which the Names have been exposed. A reasonably competent underwriter would have recognised the potential for aggregation between the Syndicate’s LMX exposure and its PSL exposure, and accordingly would not have retained net for his account the amount of PSL business that Mr Bullen did, even if (which is not admitted) it would otherwise have been prudent to do so.

By 1988, I accept from Mr. Wilshaw that PSL had generally become recognized as high risk business and in relation to comparable business [an Estate protection plan] he contracted specific reinsurance protection in relation to it. Mr. Wilshaw’s statement continued:

Mr Bullen should have recognised that in writing personal stop loss business he was writing high risk business. Firstly it is stop loss business: the personal stop loss underwriter is taking on liability for an aggregation of losses. Secondly, he is accepting liability for a high proportion of policy holder’s potential underwriting loss for a very small percentage of that Name’s total premium: it is therefore highly geared business. Thirdly, it is very difficult to assess a PML for the aggregate exposure deriving from a book of this business because the stop loss reinsurer will have extreme difficulty in identifying the events which will give rise to his PML. He ought nevertheless to assume that a loss of sufficient magnitude to cause one policyholder to claim under his policy will often be general to at least part of the Lloyd’s market and thus might cause other policy holders to make claims. There is an inherent risk of accumulation of loss. It was necessary therefore for Mr Bullen to give very serious consideration to the size of the aggregate exposure which he decided to run net on this part of his book. By this time it was no longer possible to regard this class of business as “historically profitable”.

Mr. Outhwaite has helpfully traced the origins of PSL policies in his report at pars. 105-112 inclusive. He says that in 1988/1989 the average Lloyd’s Name had an underwriting capacity of between 450,000-500,000 spread over some 15 to 18 syndicates, giving an average rate on line of about 30,000. He gives an example of how a PSL policy worked in practice:

Assume a Name has an overall capacity of 500,000 and has a PSL policy for 150,000 limits, excess of 50,000. The Syndicate in which he participates with the line of 30,000 makes a 200% loss, ie 60,000 to him. The remainder of the Syndicates produce a 5% profit (ie 5% of 470,000 or 23,500). His overall loss is 36,500, which is some way below the excess point.

He continues:

Provided an Underwriter accepts a large number of PSL policies (in practice, this means subscribing to the main market line slips) there will be a natural balance in the account. This is because the exposure will be spread over every syndicate in Lloyd's: virtually no two member’s participation will be identical. Losses on one syndicate will be balanced by the profits of [sic] others. . .

Syndicate 255 wrote a small volume of PSL business, of the order of 1% of premium income. As result of the disastrous losses in 1988/89, no doubt they will have produced an overall deficit. However, I do not believe that Mr Bullen can be criticised either for having accepted the limited portfolio of such business, or on the terms on which it was accepted.

The real criticism that is made of Mr. Bullen is that he never stopped to think and reflect on what he was doing and to contemplate the risk of aggregation. He ought to have been asking himself whether he wanted this extra exposure on top of what he was already writing in the LMX market. However, in my judgment the evidence did not go far enough to enable me to decide the case in the plaintiffs’ favour on this point. Mr. Wilshaw would, I think, have accepted Mr. Bullen’s judgment to write this business if Mr. Bullen could have shown that he had applied his mind to the issue, and stopped to think about it:

Mr. Hirst Are you saying that no-one should have written Personal Stop Loss business? [A.] No, I am not saying that, not at all.

[Q.] Is that not the consequence of what you were saying? [A.] I think it is like everything else. You actually have to look at the thing. You have to make an underwriting judgment on it, but you have to have done some thinking as to, what can I lose and under what circumstances can I lose? If you make that sort of judgment, fine. If you get it wrong, as I see it, that is underwriting. But you have to have sat and done some thinking about it and you have to look at how it fits in with the rest of the book of business you have.

The leaders on this class of business, certainly Mr Mackinnon looked at how - what sort of loss can I get on this sort of business. He did some “what if” exercises, and based on that, he came to a figure. He said maybe he could lose - I forget what figure he said now, but he ended up something like a 1,000 per cent loss ratio and he tried to buy reinsurance up to that level. He did his exercise.

All right, whether he had it right or wrong - he had it wrong in the end, but he actually sat down and said, “What is the worse scene that I can think of? Yes, the worse scene is that, you know, 20 per cent of the people are going to lose me 10,000 on each one and 10 per cent of people are going to lose 30,000, and 5 per cent of the people are going to lose 80,000, and maybe 1 per cent is going to lose 100,000.”

I have just made those figures up, but he did a “what if” exercise. What if this sort of thing happens? This is the worst I can foresee happening. And he bought himself protection to where he felt - you know, where he came to having arrived at his PML. That means he actually sat and did an exercise on it which was a sensible way of approaching it.

If Mr Bullen had thought about it - I mean, I do not know what figure he would have arrived at, but if you use the same sort of exercise, I mean, if he came to - he might lose - supposing he could lose ten times his income which is going to a 1,000 per cent loss ratio because he was not going to protect himself, he decided to run it net. All right, perfectly reasonable, but is he prepared to lose 10 per cent of his stamp on one class of business? If he says yes, then fine.

Further, Mr. Outhwaite’s evidence on this point was convincing, and I accept it:

[Q.] Did you have any method of recording the aggregate exposures on those - [A.] The aggregate exposures.

[Q.] On the stop loss book? [A.] No.

[Q.] How did you assess the potential impact of that book on your account? [A.] Well, there is no effective way to aggregate the potential or the liabilities or the potential liabilities on a PSL book. So the answer is that, as there is no way of doing it, we did not do it. What we did do was to underwrite what we saw was a very small percentage of our premium income in that field.

[Q.] Did you buy reinsurance cover for that part - [A.] No.

[Q.] Now, it is right, is it not, that this is a field in which the exposure in relation to the premium income is very large? [A.] Well, I do understand that, as a philosophy. It is not - it does not seem to bear any comparison with any other area of insurance, frankly.

[Q.] Is it right that the rate on line, in other words, the premium of the exposure under a given policy tends to be very small on personal stop loss policies? [A.] It is generally of the order of 1 to 1.5 per cent. If you were looking at, for example, super-tankers or semi-submersible drilling rigs or things of that nature, the answer is that would be an extremely handsome rate.

[Q.] Did the claims impacting your stop loss policies which you underwrote into your 1988, 1989 and 1990 years substantially emanate from policy holders who, themselves, were affected by catastrophe losses as Lloyd’s underwriters? [A.] I cannot possibly say. It is a mixture, obviously, they will have been. There will have been losses which a - which derived from Names who were on various syndicates, including Mr Bullen and the Fagan, the Feltrim ones and Gooda Walkers, of course; there will have been claims arising from Names on that Syndicate.

[Q.] The circumstances are likely to give rise to claims under stop loss policies, are likely to be losses common to a large share of the Lloyd’s market; that is right, is it not? [A.] You are likely to get significant claims on a stop loss book of business unless there are substantial claims in the - losses in the market.

[Q.] So, what one postulates is a circumstance likely to cause policy holders in large numbers to claim on their personal stop loss is some event or series of events which causes, if not market wide loss, losses common to a substantial proportion of the market? [A.] Yes.

[Q.] A major catastrophe or one or more majorcatastrophes will fulfil such a description, would it not? [A.] They may, they may.

On the basis of this evidence, I decline to hold that Mr. Bullen ought not to have written his PSL book of business, which was the bold submission made to me in the plaintiffs’ closing submissions. It was risky business; there was a risk of aggregation, but that risk was quite unquantifiable. The “control” which the underwriter exercised was to limit the amount of business he wrote in terms of premium income and to write a spread of business so as to minimize the risks. There were a limited number, only, of syndicates who were significant players in the LMX market at the higher levels of catastrophe reinsurances: there was no sound reason for believing that merely because that sector made losses, all the other syndicates would be adversely affected in the same way. In the light of the way Mr. Bullen wrote his book of business I can understand the concerns felt by Names as to his writing PSL business. It may well be that Mr. Bullen was not as perceptive as he should have been but in my view he cannot be said to have fallen below the standards to be expected of every competent underwriter.

4. MATCHING REINSTATEMENTS

This allegation is confined to the 1989 year of account and relates to certain specific layers in Mr. Bullen’s outwards whole account reinsurance programme where he obtained only one reinstatement, instead of two, despite the fact that his inwards business had two reinstatements. The layers are identified in par. 36(2)(d) of the amended points of claim as:

01.09.88-31.08.89 2.5m xs 18m (100%) 01.09.88-31.08.89 2.5m xs 20.5m (100%) 01.09.88-31.08.89 2.5m xs 23m (100%) 01.09.88-31.08.89 2.5m xs 25.5m (100%) 01.10.88-31.12.89 1.0m xs 28m (83.13%) 01.11.88-30.11.89 1.0m xs 28m (12.5%) 01.12.88-31.12.89 1.0m xs 29m (100%) 01.01.89-31.12.89 1.0m xs 30m (44%) 01.01.89-31.12.89 1.0m xs 30m (2%)

On the basis of the evidence before him, including that of Mr. Outhwaite, Mr. Justice Phillips in Gooda Walker concluded as follows:

My conclusions are as follows. The competent excess of loss underwriter had to give careful consideration not merely to his vertical exposure but also to his horizontal exposure. This was true whether he was writing high level catastrophe business in the spiral or low level reinsurance of direct business and it is axiomatic that the underwriter had to plan his pattern of reinsurance protection. In relation to true catastrophe reinsurance, to grant more reinstatements than those purchased would produce an area of obvious exposure. I accept Mr Outhwaite’s evidence that there might be circumstances in which an underwriter would deliberately take such a risk. Nothing suggests that such circumstances applied in the present case, or that any of the Gooda Walker underwriters were deliberately taking such a risk. They should, in my judgment, have been following a policy of matching reinstatements to the catastrophe excess of loss business that they were writing in the spiral.

It was common ground between the parties at the trial, although not until then, that Mr. Bullen was not able to buy in the market two reinstatements on his outwards reinsurance programme at commercially acceptable rates. The primary case advanced on the pleadings was that Mr. Bullen should have purchased two such reinstatements: thus implicitly contending that such were available. However, in the alternative, the plaintiffs contended that if such were not available, Mr. Bullen should have limited the number of reinstatements he was offering to cedents:

The decision to limit the reinstatements on this inwards business should have been taken when it became apparent that he was unable to acquire two reinstatements on his outwards programme. [File A1 page 322].

The evidence shows Mr. Bullen accepting, by way of renewal, inwards business in June, 1989 with two reinstatements, when there had been only one reinstatement in place the year before. The defendants say that the market was such that Mr. Bullen, who was not a leading underwriter, was effectively faced with the choice of not writing the business at all or of accepting non-matching reinstatements. The case for not writing the business should be rejected. As they put it in their closing submissions:

. . .such an extreme case should be rejected as being unrealistic to the point of absurdity. Mr Bullen was engaged by the Names to write LMX business, not to retire to his tent hurt.

In my judgment, an underwriter is under a duty to match his reinstatements where that can be done. In this case, Mr. Bullen could not obtain more than one reinstatement on his outwards protections after Piper Alpha. From a practical point of view he could not refuse to accept two reinstatements on his inwards business if the market, generally, was accepting it. He was, therefore, faced with a choice of either not writing some business at all or running the risk that there would be no more than two major catastrophes which would impact his inwards business. I cannot conclude that every competent underwriter would have elected to limit his inwards business in the way suggested. Accordingly I reject this part of the plaintiffs’ claim. That said, when a reasonably competent underwriter came to make judgments about his PML event, and the amount of the syndicate’s net exposure, he would, I think, take into account the lack of match referred to above. The fact that this part of the case is concerned with horizontal as opposed to vertical exposure, does not alter the position. If a competent underwriter knows that his syndicate is, unintentionally, exposed horizontally, in a way which would normally be avoided, in my view he is entitled to take that into account when assessing the amount of vertical exposure which he is entitled to run net for his Names.

5. CAUSATION

Although named “causation”, contained within this issue is the question of the measure of damages to which the plaintiffs are entitled against RTY: a) first, in relation to Mr. Bullen’s failure to comply with the duty of an underwriter to plan his account, to make a proper assessment of the book’s exposure to catastrophe loss and to purchase an appropriate level of reinsurance cover [the main issue] b) second, in relation to Mr. Bullen’s decision to accept INX Re as one of the participants on the SPT, and to cede to Aneco more aggregate than he should have done [the SPT issue].

As usual, there are no difficulties in stating the legal principles governing causation and the measure of damages; and I take them from the judgment of the Court of Appeal given by the Master of the Rolls in Banque Bruxelles v. Eagle Star, [1995] L.R.L.R. 195; [1995] 2 All E.R. 769:

Causation

As the Master of the Rolls said, at p. 202, col. 1; p. 842b:

The approach of the Courts to the issue of causation is in principle simple, pragmatic and commonsensical.

He endorses a passage from Lord Wright’s speech in the Yorkshire Dale Steamship Co. Ltd. v. Ministry of War Transport, [1942] A.C. 691 which draws a distinction between the Court’s approach and that of a philosopher or scientist. What is the dominant, real, efficient or effective causative event is thought to be recognisable when it can be observed, in contradistinction to an event which does not possess those characteristics. I must confess that, instinctively, I would prefer not to appeal to “commonsense”, as though it sometimes underlies judicial reasoning and determination and sometimes does not. However, it seems to me clear, from (*470) the judgment of the Court of Appeal, that causation issues must not be allowed to be bogged down by Latin tags or by the sort of analysis given to the concept of philosophers such as Aristotle or Hume. The appeal to commonsense is, I think, simply to reinforce that point and to highlight the fact that a judgment on causation may not be capable of being sustained by logical reasoning. Hence, the citation by Sir Thomas Bingham of a passage from the judgment of Cooke, P. in a New Zealand case:

The ultimate question as to compensatory damages is whether the particular damage claimed is sufficiently linked to the breach of the particular duty to merit recovery in all the circumstances [ - McElroy Milne v. Commercial Electronics Ltd [1993] 1 N.Z.L.R. 39 at p. 41].

Measure

The rule of the common law is that where a party sustains a loss by reason of a breach of contract, he is, so far as money can do it, to be placed in the same situation, with respect to damages, as if the contract had been performed. [ - per Baron Parke in Robinson v. Harman (1848) 1 Exch. 850].

If the claim is in tort then the measure is:

. . .in settling the sum of money to be given for reparation of damages you should as nearly as possible get at that sum of money which will put the party who has been injured, or who has suffered, in the same position as he would have been in if he had not sustained the wrong for which he is now getting his compensation or reparation. [ - per Lord Blackburn in Livingstone v. Rawyards Coal Co., (1880) 5 App. Cas. 25 at p. 39 [This was an appeal from the Court of Session: hence the reference to reparation]].

As in the Eagle Star case, none of the parties suggested for present purposes that there was any material distinction between the two tests.

In the light of the general principles, I consider each of the two issues in turn.

A. The main issue

What the Names are entitled to, in general terms, is to be put into the position they would have been in had Mr. Bullen written his book of business reasonably competently. That begs the question: “what book”? Having negligently taken the view that aggregates written at rates on line of less than 5 per cent. both on his XL on XL account and his whole account covers, were not exposed to loss, he wrote much more aggregate than he otherwise would have done. Had he planned his account and calculated the amount of vertical cover by reference to the PML process he almost certainly would not have ended up with the book which he wrote. The sheer size of the aggregate exposure was beyond the capability of the syndicate to purchase the necessary outwards protection: either the reinsurance would not have been available in the market or would have cost more than the syndicate could afford. But, this is not a “no-transaction” case. The plaintiffs are not saying that they would not have participated in a book of LMX business written in a competent manner. This is not a case where a false valuation or representation had induced a party to enter into a transaction which he would not otherwise have done. In other words, the plaintiffs do not say that they are entitled to be compensated on the basis that they had never participated in the syndicate for each of the relevant two years of account. They say that they are entitled to be put into the position they would have been in, had Mr. Bullen written his book in accordance with the standards to be expected of a competent underwriter. In the further and better particulars of the points of claim, served on Aug. 5 1994 [File A1 pp. 275/276], it is pleaded on the plaintiffs’ behalf:

Preferably, Mr Bullen should have had his outwards reinsurance substantially in place prior to acquiring any significant aggregate exposure on the accounts being protected. In any event, Mr Bullen should not have written any significant amount of aggregate liabilities without first having in place sufficient reinsurance protection in respect of that particular exposure. The outwards reinsurance should have been acquired from reputable and sound reinsurers, and preferably not from those who were themselves contributing to the spiral by reinsuring back into the same market as were being reinsured by Mr Bullen. If it was not possible to acquire sufficient reinsurance from sound and reputable reinsurers to protect the aggregate that it was proposed to write, a competent underwriter would have reduced the amount of aggregate liabilities he was proposing to assume to a size which could be accommodated by the reinsurance which he was able to acquire.

The writing of insurance is the deliberate acceptance of a risk: the risk that an event giving rise to a claim will materialize. Such a business necessarily involves uncertainty as to whether such an event will occur: insurance cannot be written after the event. It is, therefore, simply not possible, in any assessment of damages, to “re-write” Mr. Bullen’s book of business so as to reduce its size. Such a task would involve discarding some but not all of the aggregates he wrote. One could discard those risks which gave rise to loss, or vice versa; or have a mixture of the two. In the end, however it was done, an a priori judgment would have to be made as to the loss [or profit] which the plaintiffs would have suffered had their book been written (*471) competently. Such a judgment could only be made on the basis of some typical or paradigm syndicate whose business largely lay in the LMX market. But there is no such syndicate. As the Chatset statistics make clear, some syndicates in that market fared very much better than others, for the relevant years. It seems to me likely that the LMX market could only have existed if some of the underwriters were less competent than others. An assumption that there was some kind of “standard” syndicate would be to assume that the LMX market did not exist. In any event, different underwriting policies by different competent underwriters will lead to different results. Finally, an element of luck must also be involved, even in competent risk taking.

The plaintiffs do not say, and nor could they, that all the losses the Names sustained in the two years of account are due to Mr. Bullen’s negligence. That is because of the nature of the allegations which have been substantiated. The complaint relates principally to the writing of his XL account. But that was not the only business he wrote: he had some direct accounts as well, and if any losses resulted therefrom they are not attributable to any alleged negligence. And in the light of my findings in relation to PSL policies, the losses associated with them are not recoverable. Further, as the complaints under this head are confined to a lack of adequate and proper vertical insurance, the plaintiffs are not entitled to be protected from losses which might have arisen from the inadequate horizontal cover. As it was accurately put by Mr. Hirst, Q.C.:

. . .1988 and 1989 were dreadful years for the LMX market in London and Lloyd’s generally after 20 years of uninterrupted profits.

In fact, at par. 50 of their amended points of defence, RTY rely on what they describe as the “extraordinary and unprecedented number and nature of major catastrophes” impacting on both years of account as being the true reason why the Names suffered their huge losses, as opposed to any negligent underwriting. It seems to me that this plea confuses liability and causation. If the case being advanced is that these catastrophes constitute some kind of supervening event which breaks the chain of causation leading from Mr. Bullen’s negligent underwriting to loss, I cannot accept it. The Names are entitled to be compensated on the basis that their syndicate’s business was competently written. If, despite being competently written, they would still have suffered some loss, because the years of account were always going to lead to loss, then I can understand the point. Whether it is supported by fact will only become apparent after the damages have been assessed. The defendants supported this part of their case with the evidence of a consulting actuary, Mr. Clarke,who gave evidence to Mr. Justice Phillips in Gooda Walker and Feltrim. It seems to me on the basis of his evidence, which I found more interesting than helpful, that none of the catastrophes identified were outside normal experience. All would have been events against which persons affected would, and did, seek insurance. These are precisely the sort of events which a catastrophe reinsurer may foresee as impacting upon a book of business. Catastrophes do not occur at regular intervals: there is no a priori reason why they should. As Mr. Alexander accepted, the destruction of one rig might be followed by the destruction of three more in the next 12 months. The conclusion which Mr. Clarke arrived at, that for the 1989 and 1990 years the total amount of catastrophe loss was larger than could have been expected, given the experience up to that time, was based upon a model.

[Q.] As I understand it, what your model tries to do is impose some sort of order upon this chaos of random events, some sort of mathematical order? [A.] It attempts to produce a model - it is a model which, by looking at the past data, attempts to make predictions to which you can assign the probabilities his Lordship has just referred to, in looking at future catastrophes.

But his model seemed to me to be unreliable due to the relatively sparse amount of data. As I understand it, the model takes into account all the catastrophe events giving rise to losses over a certain level during the years from 1969. Therefore, it may well not include within it catastrophes whose probability of occurrence is more than 29 years: whether such catastrophes have been included will not be known. The model calculates mathematically the probability of occurrence of losses from catastrophes in certain brackets. The reliability of a probability model can be tested by running within it sensitivity tests, designed to change assumptions to see what effect such changes have on the output. Its reliability can also be judged by looking at its output, and seeing whether it is intuitively correct or suspect. On the evidence I am satisfied that Mr. Clarke’s model was “over sensitive” to changes in parameters within the model and that it gave results which were intuitively suspect, of which I give one example:

[Q.] You are including in your data the data relevant to US and European catastrophes, and you are adding further catastrophes on top in so far as they occur outside the United States and Europe? [A.] Yes. I suspect there are not that many that occur. (*472)

[Q.] It may be the same, but it is likely to be larger? [A.] Yes.

[Q.] It would be very odd if it was smaller, would it not? [A.] It should be higher.

[Q.] What explanation could here be for it being smaller? [A.] There should be one. It should not be smaller.

[Q.] Could you help me on probability estimates and actual amount estimates in the table on page 4? If you, for example, take world-wide catastrophes with a cut-off point of $500 million, your expected amount predicted by your model is 3.968 billion, but, if you then go across and look at European and North American, where you are dealing with a smaller exposure - [A.] I was not paying attention. Could you repeat the question? I was looking at the figures. I was not paying attention.

[Q.] I am looking at the last box on the left-hand side on page 4 of bundle Q. For 1990, we can see that your model predicts, using trigger points of 500 million dollars, an expected amount of catastrophe loss worldwide of 3.968 billion dollars? [A.] Yes.

[Q.] If you then go across to the next box on the right-hand side, where you are dealing with the same exercise, only a more limited scope, Europe and North America, you see that the expected amount of catastrophe loss is considerably higher, 4.76 billion dollars. Your model is producing a rather counter-intuitive conclusion, is it not? [A.] I would accept that. I think it must deduce the fact we have already mentioned, that the model is based on a fairly limited amount of data in both cases.

If the model is not reliable, as in this case, then it becomes dangerous to use it for any purpose. For what it is worth, I am confident that it may, at some stage in the future, become sufficiently robust to become a reliable tool, but at the moment I am simply unpersuaded that it has any value. I am not persuaded that the 1989 and 1990 catastrophes were unexpected in any sense relevant to this litigation. It is probable that insured losses of ever increasing proportions and, frequency, will occur due, in part, to the greater urbanisation of the world’s surface, the cost of materials and the sophistication of engineering projects in whatever field. In my view, any underwriter who wrote his book on the assumption that he was being accurately informed of the probabilities by this model would be mistaken.

If the plaintiffs cannot contend that this a no-transaction case, they must accept that they are susceptible to the risk of loss even if Mr. Bullen had purchased adequate vertical reinsurance protection in accordance with his duties. The plaintiffs’ pleaded case [par. 39 of the amended points of claim] is as follows:

In and after 1987, a series of catastrophe losses occurred in relation to which [the Syndicate] was substantially exposed by reason of its participation in XL on XL and whole account reinsurances, and the operation of the spiral effect in relation thereto. Thus:

(1) The 1988 underwriting year is exposed to substantial losses arising from:

(a) the UK windstorms occurring in October 1987;

(b) the blowout and fire on the Enchova Oil Platform occurring in April 1988;

(c) the explosion on the Piper Alpha oil rig occurring in July 1988;

(d) the Exxon Valdez oil spill occurring in March 1989;

(e) the explosion and fire on the Arco Baker platform occurring in March 1989;

(f) the Hurricane Hugo loss occurring in September 1989;

(g) the Phillips Petroleum plant explosion occurring in October 1989.

(2) Furthermore, the 1989 underwriting year is also exposed to substantial losses arising from the Exxon, Arco, Hugo and Phillips losses, as well as from the UK and Continental windstorms (known as 90A) occurring in January and February 1990.

The way it was put in plaintiffs’ closing submissions was as follows:

It is submitted that it would not be right to assess damages on the basis that, had the breaches of duty not occurred, the Plaintiffs would have been in the position of participating in a syndicate which had still written the book of business which Mr Bullen actually wrote but which had also purchased the full amount of reinsurance called for by the aggregate exposure of that book. Such reinsurance would have to be sufficient reinsurance to reduce the competently assessed net PML exposure arising from that aggregate to a figure which the court holds could properly have been retained by a competent underwriter in Mr Bullen’s position. On the evidence, this complicated exercise would involve postulating a state of affairs which was never possible. This contention applies a fortiori to damages sustained as a result of breach of duty in relation to the 1989 year of account. - [par. 19.4]

A number of reasons are put forward for this contention, including a lack of available funds to purchase the hypothetically proper level of vertical reinsurance. (*473)

On this basis, the plaintiffs’ submissions were put in the alternative:

1. I should accept Mr. Wilshaw’s estimate that members’ agents would have expected a loss of between 30 per cent. and 50 per cent. of stamp for each of the years, in the light of what actually happened in these two years; in other words, in the light of all the disasters which occurred.

Mr. Wilshaw’s evidence to this effect was given in answer to a question from me. I considered him to be a reliable and trustworthy expert witness, whose evidence more generally was of great assistance. But it would, I think, be wrong to use this answer as a foundation for the assessment of damages. Mr. Wilshaw must have been forming his own assessment of a paradigm syndicate, when he gave this answer. But, that said, it was of value when coupled with the evidence of Mr. Outhwaite, to which I shall refer later.

2. I should follow the method of measuring damage adopted by Mr. Justice Phillips in the Gooda Walker case, as opposed to that adopted by him in Feltrim. In the former case, the plaintiffs contended that the underwriters were negligent in leaving the names exposed to five main catastrophes [Piper Alpha, Exxon Valdez, Hurricane Hugo, Phillips Petroleum & Windstorm Daria 90A]. Having held that the underwriters were incompetent in much the same way as I have found Mr. Bullen to have been incompetent, Mr. Justice Phillips said this:

It is for the plaintiffs to prove that the losses that they have sustained have been caused by the defendants’ breaches of duty. In this action the plaintiffs have concentrated, almost exclusively, on the attitudes and actions of the Gooda Walker underwriters in relation to vertical exposure to catastrophes and the consequences that these have had in relation to losses flowing from the five central catastrophes. In the case of each catastrophe, losses have been suffered because the underwriters failed to put in place - whether on a first loss or reinstatement basis - cover that extended sufficiently high to cover the claims arising out of that catastrophe. The plaintiffs have satisfied me that the resultant exposure was neither an unforeseeable consequence of excess of loss underwriting, nor a consequence that was foreseen by the underwriters as a possible result of a deliberately calculated risk taken on behalf of the Names. It was a consequence of an incompetent disregard of important principles of excess of loss underwriting. In my judgment the plaintiffs should recover by way of damages such sums as will put them in the same position as if this exposure had been protected by reinsurance. To an extent this is an artificial approach, for in some cases such reinsurance would not have beenobtainable. To have regard to such a situation would not, however, absolve the defendants from liability in damages, but set in train an alternative enquiry of greater complexity which would be unlikely to result in any lesser liability on the part of the defendants. Accordingly, I consider that the test which I have adopted is one which combines relative simplicity with a fair and correct approach in principle. - [pp. 229-230]

On the basis of those facts, the plaintiffs were to be covered in respect of the losses they sustained in consequence of the five named catastrophes, but not:

Losses resulting from the erosion by other loss events of reinsurance cover that would otherwise have been available as protection against the five central catastrophes. [p. 230, cols. 1 and 2.]

In Feltrim, the Judge had second thoughts about this very difficult topic. He noted that, as here:

The only acts or omissions alleged by the Names to constitute breach of duty are failures to put in place adequate vertical reinsurance protection. . .In my judgment the correct approach to damages is to attempt, insofar as this is possible, to place the Names in the same position as if those breaches of duty had not occurred.. . .This is essentially the same approach to damages that I adopted in Gooda Walker. In that Action, however, in dealing with liability I did not have to determine either the PMLs for the various syndicate years or the maximum exposure that the underwriters could competently have run. In these actions I have determined both the factors that should have been applied to aggregates in order to determine PMLs and the maximum net exposure that could properly have been run on the basis of those PMLs.. . .In the result the assessment of damages will involve a more sophisticated, and more complex, exercise than that on which I am about to embark in Gooda Walker. In preparing for the assessment of damages in Gooda Walker a number of points have been raised which I had not anticipated when ruling on the principles to be applied in the assessment of damages. These have persuaded me that in these Actions I should do no more than formulate the basic principle to be applied, leaving open for later argument the manner of application of that principle. That principle is simply that the damages awarded should place the Names on each syndicate year in the same position as if the underwriters had purchased reinsurance protection sufficient to restrict the Names’ net exposure to the PML of 100% of stamp. In calculating net exposure regard should (*474) be had not merely to the gap above the top layer of reinsurance cover but to retentions and co-insurance.

It seems to me that it is neither sensible nor practicable to approach the question of loss in relation to the main issue in any way other than by starting with the book of business which Mr. Bullen actually wrote in each of the two years of account. The complaint against him is that he failed to acquire sufficient vertical reinsurance protection, and thus negligently exposed his Names to the risk of excessive losses. The only sensible and practicable way of putting the plaintiffs in the position they would have been in had Mr. Bullen not been negligent, is to ask what vertical reinsurance protection would a competent underwriter have purchased on the book of business which Mr. Bullen wrote; despite the obvious drawbacks to which the plaintiffs have drawn attention. In my view, a balance must be struck between what is theoretically desirable and what is capable of being worked out in practice. My conclusion is that the plaintiffs are entitled to be put into the same position they would have been in if a competent underwriter had competently chosen a PML event, competently calculated the PML exposure, competently decided how much of that exposure to retain net for the Names’ account, and then purchased vertical reinsurance for the balance of the exposure, in relation to the book of business which was actually written by Mr. Bullen. The loss [if any] which the Names would have suffered on that hypothesis [taking into account the “notional” cost to the Names of buying such reinsurance] will be set against the losses which they have suffered and, having taken into account questions such as interest, which does not have to be decided now, they will be entitled to the difference.

Although I am sympathetic to the submission that such an exercise is largely hypothetical, I am greatly encouraged to follow it, for two reasons. First, it is the approach adopted by Mr. Justice Phillips in the Feltrim case, having refined the approach he took in the Gooda Walker case. In my view, it would be wrong to follow a method which, based upon experience, Mr. Justice Phillips had discarded for a different and more sophisticated approach. As in the Feltrim action, and unlike the Gooda Walker action, I have listened to a great deal of evidence about the calculation of a PML amount and net exposure. Further, I have gratefully adopted much of what Mr. Justice Phillips has had to say on issues common to this case. The parties have spent considerable energy in debating the PML issues and I am, therefore, in a position to make findings which will enable the Names’ damages to be calculated on the basis which I have indicated.

The PML event

As I have already indicated, for a marine syndicate such as this, it would have been sensible to take as the PML event what Mr. Bullen said he had in mind, namely a collision between a rig (or platform) and a tanker. I can see nothing wrong with taking a figure of between 3 billion-3.5 billion. At the end of the day, although there had been discussions about whether a hurricane catastrophe might not have been more appropriate, I regarded that discussion more as an examination of what Mr. Bullen thought he was doing, rather than as a debate about what a competent marine underwriter would have chosen as his PML event.

The PML amount

This is the amount of loss which a competent underwriter would reasonably have considered his book of business to be exposed to in the event of the occurrence of the PML event. Strictly, PML stands for probable maximum loss, and that is what has been described by Mr. Hirst, Q.C. helpfully I found, as the PML amount.

This has been the subject of much dispute and evidence. What the underwriter is doing is having to make a difficult judgment as to the risk of aggregation in his book of business. This judgment has to be made against the background of scant placing information. The LMX underwriter does not have a clear picture of the nature of the risks he is writing; he will not know anything about the cedent’s own PML calculations.

In principle, an underwriter can be sure that there will not be aggregation to the extent of 100 per cent. That is for a number of reasons, such as:

1. Cedents’ PML events against which the cedent was insuring might well be, and probably are, different from one another and from Mr. Bullen's. No underwriter on the LMX market would have known what any other syndicate’s PML event was, but he could safely assume that not all such events would be the same for every syndicate.

2. Some of the risks within the market were “watered down” through leakage, in the sense that in some cases the whole risk was not being assumed by the LMX market; the degree to which there was leakage would be unknown to an underwriter.

3. Some of the risk were partially retained net in, for example, the direct marine market.

4. Some cedents were ceding business as a “sleep easy” cover protecting more than their PML amount.

On the other hand, rate on line was not a satisfactory or reliable indicator as to exposure to risk in the event of the PML event, nor as to the risk of aggregation. As is already clear, Mr. Bullen (*475) knew about the spiral. He may have misunderstood the extent to which it would impact on his book when a loss entered the spiral, but every competent underwriter would have known at least as much as was evident to those who had written about the market, such as Mr. Outhwaite in his notorious paper published in April, 1988. It seems to me very clear from the expert evidence in this case that every competent underwriter would have assumed his exposure on his XL on XL account written in the LMX market to be wholly exposed to a PML event. Therefore, taking the 1988 aggregates from the table at the beginning of this judgment, a competent assumption would have been an aggregate exposure of U.S.$96,228,000; and, for 1989, U.S.$104,071,000.

There was a substantial issue as to the proper approach which a competent underwriter would take to assessing his exposure to a PML event in relation to his whole account covers. It is to be noted that Mr. Bullen did not, during these two years of account, so code the whole account risks that he could distinguish between those whole accounts which had within them XL on XL business and those which did not. But the placing information would have enabled Mr. Bullen to distinguish. He frankly admitted that he had no system in force to enable him to do so:

Mr. Justice Morison: Before you move on, can I ask you this: did you have any system in operation in the box where you were writing an XG cover which attempted to analyse what proportion of the whole account was devoted to XL business? [A.] No, I think is the straight answer to that, yes.

Mr. Thompson, whose help in analysing the data has been of most value to the Court, [in contrast, if I may say so, to his rather Calvinist approach to risk taking which was of less assistance by comparison with the other experts], was prepared to admit that a competent underwriter could make some limited discount, but only of the order of 10 per cent. The plaintiffs ask me to adopt the approach that in the absence of any reasonable grounds for a different conclusion an underwriter should assume his whole account book to be 100 per cent. exposed. Mr. Hirst points out, correctly in my view, that this is no different from saying that every underwriter would assume that all his aggregates were 100 per cent. exposed to any one catastrophe because, due to the inherent uncertainties and opacity of risk in the market, one could never show, or have, reasonable grounds for making a discount. The plaintiffs’ approach, whilst sounding sensible to a lawyer, does not allow, I think, for the vagaries of themarket or the nature of the somewhat imprecise judgment which a competent underwriter is being asked to make.

I discount Mr. Bullen’s evidence on this topic, as I do not believe he ever approached the question of his vertical reinsurance on the basis now being considered, and it is, therefore, to the experts’ evidence I must turn.

Mr. Outhwaite said this:

[Q.] You treated for PML purposes, as an excess of loss on excess of loss cover, any whole account which included excess of loss? [A.] I did.

Mr. Alexander had a different system. He looked at each whole account cover. If there were XL on XL within the whole account, he would apply a multiplier to the estimated premium income given by the reinsured in relation to the account in question and treat the whole account as XL on XL for PML purposes if the figure resulting reached the excess for the layer in question. This was a very imprecise method, bearing in mind the paucity of information available, but it does show that he regarded it as necessary to apply some different judgment to the assessment, taking into account the amount of XL on XL cover contained within the whole account. Mr. Wilshaw said that a small discount of between 5 per cent. and 10 per cent. might be permissible, bearing in mind that not all parts of a book would be 100 per cent. loss.

From an examination of records carried out by and on behalf of Mr. Thompson, it would appear that a substantial proportion of Mr. Bullen’s whole account covers contained XL on XL business. The schedules relate to about 58 per cent. of his whole account aggregates for 1988 and 50 per cent. of them for 1989.

While there may well be various different ways of seeking to measure the syndicate’s exposure to the PML, event, having heard the experts I am prepared to hold that no competent underwriter could have concluded that his whole account covers were other than exposed 80 per cent. In reaching my decision I have taken into account the substantial portion of the whole account book which contained XL on XL business. In principle, XL on XL business should, I think, be treated as 100 per cent. exposed whether it is reinsured as XL on XL or is part of the cedent’s whole account. Mr. Outhwaite treated them in the same way; Mr. Alexander was more complicated about it and I do not accept that his methodology was sustainable as a logical and simple exercise, which this should be. Further, taking the aggregate in the 0 per cent.-4 per cent. rate on line segment of Mr. Bullen’s 1988 whole account book, approximately U.S.$110 of the U.S.$197 m. was written in favour of companies or Lloyd’s syndicates who themselves either wrote more than 50 per cent. XL loss or were known as major retrocession LMX specialists. (*476)

Finally, bearing in mind that what is called for is a broad judgment, any competent underwriter would, I think, have appreciated that not all his risks, which had a potential for some aggregation in the event of a serious catastrophe, were in his aggregate book. Thus, he had an unknown exposure to PSL claims, which had some potential for accumulation in the event of the sort of disaster which Mr. Bullen’s PML event implied. Further, in relation to the 1989 account he knew that he had layers where there was no matching re-instatement. Whilst such factor would have no direct bearing on his writing of LMX business, in terms of assessing exposure and deciding on the amount of vertical reinsurance protection to acquire, it was a factor the underwriter should have taken into account. As Mr. Hirst put it: the assessment of the PML amount is not an actuarial or purely mathematical exercise; it involves judgment and feel and the decision must be taken on the basis of all the information and experience available to the underwriter, and the assessment must be on the whole class of business and not on a risk by risk basis.

I believe that my conclusion is entirely in line with that reached by Mr. Justice Phillips in Feltrim.

In relation to the other three components of his LMX book, I see no reason to depart from what Mr. Bullen said he would have estimated his cumulative exposure to have been in relation to his rig XOL account, namely 100 per cent. If his PML event involved the loss of a platform, then 100 per cent. exposure was the only sensible assumption. With regard to his liability XOL, and using the figures which Mr. Bullen said he used, in 1988 he regarded this part of his book as 77 per cent. exposed and in 1989 as 84 per cent. exposed. The variation in these figures demonstrates that using rate on line was not a reliable yardstick for assessing the risk of accumulation, because there is no reason to believe that that risk varied from year to year, depending on the strength or otherwise of the market. In relation to the fifth category, hull XL, on Mr. Bullen’s case he would have said for 1988 that this part of the book was exposed to the extent of 24 per cent. in 1988 and 30 per cent. in 1989. Bearing in mind that one of the components in his PML event was the loss of a hull, I find the latter figure somewhat surprising. However, having regard to the arguments, which were principally directed to the XL on XL account and the whole account, I am content to conclude that an exposure of 80 per cent. in relation to the liability account and 30 per cent. for the hull XL account should be assumed for each of the two years.

I shall leave the parties to calculate the PML amount in the light of these findings.

Mr. Hirst, in a part of his written closingsubmission under the heading “The proof of the Pudding” made the following submission:

1. The Piper Alpha loss was close to the loss to be expected from a PML event. Hurricane Hugo was a PML event, or close to it.

2. The losses to the syndicate from these two events, on the basis of the current incurred loss, are U.S.$167.7 m. and U.S.$206 m. respectively.

3. Having regard to various adjustments which should be made, the minimum competent PML amounts for 1988 and 1989 should be taken as 85 per cent. of the actual losses suffered by the syndicate as a result of the two disasters giving the following figures: 1988 - U.S.$142.4 and for 1989 - U.S.$174.1 m.

It seems to me that this approach, which can be properly called a “lawyer’s construct” [Day 21, p. 23, line 3], is unsound. I agree with Mr. Crane’s submission that it is wrong in principle to assess a PML amount by reference to the impact of an actual loss after the event. What is in issue is the extent of Mr. Bullen’s obligation as a reasonably competent underwriter to protect his Names by vertical insurance against the happening, in the future, of a hypothetical event. The “reconstruction” of events ex post facto is doing what Mr. Hirst constantly cautioned me from doing: namely judging the case with the benefit of hindsight. Further, the PML exercise is an assessment of exposure to a hypothetical worst case, with an assumption that all lesser catastrophes would be catered for. Piper Alpha was a disaster which fell short of the PML event which Mr. Bullen referred to, and on Mr. Wilshaw’s evidence, which I accept, Hurricane Hugo was not of PML proportions. It seems to me that one cannot assume that the impact of any catastrophe, whether or not of PML proportions, up n the book will bear any proportionate relationship with the impact of the PML event on the book. The assumption implicit in the lawyer’s construct is simply not justified, because it is founded on a proposition which has no evidential justification. How a catastrophe of any particular size will impact the book is a matter of pure speculation. The unpredictable nature of the spiral was the very reason why every competent underwriter had to undertake a proper PML exercise. To assume that any one loss will travel through the spiral in the same way as any other is simply contradicted by the facts. The pudding is only proved if the impact of either Piper Alpha or Hurricane Hugo on Mr. Bullen’s book provides some reliable guide as to how a different hypothetical catastrophic event would have impacted upon it. The point was not put to the expert witnesses for their consideration; it was raised only in RTY’s closing submissions, (*477) which is why it can accurately be described as a lawyer’s construct. Further, the construct involves assuming two PML events, whereas the concept of a PML exercise is that there is one PML event and not two. In other words, the premise upon which this argument is founded is not made good on the evidence. In the end, this was simply an ingenious submission which had no sound basis on the evidence. The pudding is not proved.

Equally defective, in my view, was an alternative approach suggested by Mr. Hirst, which involved taking the combined efforts of Messrs. Bullen, Green and Marsh to construct their best estimate of the actual exposure of different sections of the book to the loss of Piper Alpha, after the disaster. Again, it assumes that the disaster was akin in size to the PML event, when it was not; second it assumes that an underwriter had the benefit of hindsight; third it involves an assumption that Mr. Bullen understood what he was doing by discounting the aggregates written at rates on line of less than 5 per cent. Mr. Hirst says that “Mr Bullen was really conflating the second and third stages of the PML calculations”. I am afraid I cannot agree, for reasons which I have already endeavoured to give. In truth, Mr. Bullen did not plan his vertical protection in anylogical or socnd manner and I am not assisted by a reconstruction of what he did after the event.

The net exposure

This was, also, a very contentious issue.

The background to it may be stated in this way. In the Feltrim case, Mr. Justice Phillips heard evidence to the effect that a reasonably competent underwriter could have deliberately exposed his Names to a risk of loss equal to the amount of the syndicate’s stamp capacity. Thus, having calculated the PML amount, an underwriter would, he held, have been entitled to deduct the stamp capacity and purchase vertical reinsurance to cover for the balance. It was said that joining an LMX syndicate was joining a syndicate which was high risk. On the principle that damages are to be assessed on the basis that the contract breaker would have performed the contract in the way most beneficial to himself, Names cannot be compensated on the basis that the underwriter would not have deliberately exposed them to loss to this extent.

It seems to me that there is a clear distinction to be drawn between deliberately exposing Names to loss, on the one hand, and foreseeing that LMX syndicates are high risk, on the other. Although this case is about the lack of vertical reinsurance, there are all sorts of events which might cause a Name to suffer a loss on such a syndicate as this, even if the underwriter had not deliberately exposed him to any risk, in the sense of retaining net for the account any vertical exposure on the occurrence of the PML event. For example, a competent underwritermay have misassessed his PML event or his syndicate’s true aggregate exposure; the outwards reinsurance protection which he obtained might turn out not to sound, leaving the syndicate with bad debts; losses may have occurred due to non-negligent mistakes in relation to horizontal cover; in respect of some of his book there might not be matching reinstatements because of the way the market moved in a particular year. Further, the LMX book was only a part of the syndicate’s business. Losses might accrue from risks competently accepted, such as risk from the book of personal stop loss business which Mr. Bullen wrote.

In their voluntary particulars of par. 35 of the points of claim, the plaintiffs advanced their claim in this way:

The Plaintiffs’ actual [sic] contention with regard to the net exposure which a competent underwriter can prudently retain is as follows:

(1) A competent underwriter will limit his reinsured exposure to an amount which, on any view, can safely be regarded as representing an acceptable level of potential loss. having regard to the Syndicate’s stamp capacity for the year of account in question.

(2) In setting the overall maximum limit to the Syndicate’s exposure from any one event or occurrence, the competent underwriter will have to take account of the fact that several catastrophes could well occur in any one year, and that attritional losses and expenses could also account for a significant part of the Syndicate’s premium income.

(3) It will be contended:

(a) that because of the Active Underwriter’s ignorance of the position of individual Names and of their underwriting philosophy, he cannot safely assume a risk of sustaining a loss which exceeds the size of the Syndicate’s stamp plus that part of each Name’s proven wealth which notionally supports their underwriting on the Syndicate, though that figure might be reduced still further by factors specific to the Syndicate itself;

(b) that the foregoing considerations relevant to the capacity of Names to bear losses indicate that a competent LMX underwriter would not ordinarily be justified in assuming an unreinsured aggregate exposure to (*478) any one event or occurrence in excess of 40% of stamp. However, as indicated in paragraph 3(a) above, that figure could be reduced still further by factors specific to the Syndicate itself.

I should say, at once, that I do not think that the level of exposure to which a Name can be deliberately exposed has anything directly to do with his proven wealth. The fact that at the time a Name was entitled to write business up to 2.5 times his proven wealth cannot be taken to mean that he ought not on that account to be exposed to a liability greater than his proven wealth [i.e. 40 per cent. of his premium limit] because it is obvious that some Names deliberately only “prove” wealth well short of their real wealth, whereas that may not be so with many other Names. An underwriter may not, and Mr. Bullen did not, know anything about his Name’s financial position. That was no concern of his. He knew what premium income was allocated for the particular year and the decisions he had to take were not, I think, determined in any way that could be related to a Name’s underlying capacity to withstand financial loss. Conversely, as I have already said, the fact that a Name is warned when he joins Lloyd’s that insurance is a high risk business, that losses may be made as well as profits, and that his liability is unlimited [down to his last cuff-link or earring], does not entitle the underwriter to write business on the basis that the last cufflink or earring can be staked.

It seems to me that an underwriter can only deliberately expose his Names to risk on the basis of their implied consent. Or, to put it another way, a Lloyd’s underwriter may not deliberately expose his Names to the risk of a loss of a size which exceeds that which they could reasonably have anticipated and, therefore, that which they could be said to have accepted by becoming or remaining a Name on that syndicate. What a Name can reasonably anticipate depends upon what a competently advised Name knew or ought to have known about the syndicate’s business. Proper advice will reflect the market’s appreciation of the general nature of the syndicate’s business and of the underwriter’s reputation and, importantly, what was published by the managing agency about the syndicate, including its results and underwriter’s report.

This conclusion is entirely in line with the Feltrim decision and is supported by the defendants’ expert, Mr. Outhwaite during cross-examination on his, now, famous paper delivered in April, 1988, in which he wrote about the LMX market:

[Q.] Then you say:

Thirdly, an underwriter may accept that a substantial amount of the liabilities he is accepting are not protected by reinsurance on the basis that it is unlikely that a loss of sufficient magnitude would happen at all, and if it did it is acceptable that his Names or shareholders could suffer a loss. That is a perfectly reasonably approach, provided that the Names and shareholders are aware that this is the case.

Pausing there, that was something you believed, was it not, that it was a reasonable approach, provided that the Names and shareholders were aware that that was the case? [A.] I think it may have been acceptable.

[Q.] Provided the Names and shareholders were aware that it was the case? [A.] Yes, there are all sorts of aspects to this, of course, because you can tell people that this is exactly what the position is. Whether they accept that in the event of a loss or not is quite a different matter.

[Q.] I see. Unless people are told it is the case, you are making the point that it would not be acceptable? [A.] It seems to me that syndicates or people on a syndicate or the agents ought to know the characteristics of the syndicate.

[Q.] If, in fact, the underwriting policy of the syndicate is to retain for net account a PML exposure or liabilities that are not protected by reinsurance, that is perfectly acceptable or reasonable, provided that the Names and agents supporting that syndicate are told, that is what you are saying, is it not? [A.] It would be.

[Q.] That is what you are saying? [A.] Yes.

I turn, therefore, to consider what information was available about the syndicate.

In force at the relevant time was Byelaw No. 1 of 1985 made by the Council of Lloyd’s on Mar. 11, 1985 pursuant to s. 6(2) of the Lloyd’s Act, 1982. This byelaw prescribed the form of agreements pursuant to which, and no other, all underwriting business was to be undertaken after Dec. 31, 1986. It was those forms of agreement which were considered by the House of Lords in Merrett. Under the sub-agency agreement between the managing agent of the syndicate [the sub- agent] and the members’ agent, cl. 10 provided that:

The sub-agent shall furnish to the Council of Lloyd’s such reports, accounts, certificates, figures and information in relation to the Syndicate as the Council may from time to time require, and the sub-agent shall supply to the Agent such reports, accounts and certificates as the Council may from time to time require, together with such figures, information and particulars concerning the Syndicate underwriting business as may be necessary for the Agent or the Agent’s auditors to have for the purpose of discharging (*479) the Agent’s obligations to the Council and to the Agent’s Names.

Under the agreement between the Name and his members’ agent, there was a termination provision [cl. 12]. Sub-clause (c) provided that:

The agency shall be determinable either wholly or in respect of any Syndicate on the 31st December of any year by the Name or the Agent giving to the other not less than six calender months previous written notice. Provided that in any year in which the annual report of any Syndicate as at the preceding 31st December is received by the Name later than 1st June, the said notice period of six calendar months shall not apply in relation to such Syndicate, and in lieu thereof the agency in respect of such Syndicate shall be determinable by the Name on the 31st December of that year by written notice given at any time within thirty days of the receipt of the said report.

It follows, I think, that it was the anticipation of the market that a syndicate’s annual report, which included the underwriter’s report, was the specified formal method by which managing agents informed their members’ agents about the syndicate’s business and on which, in turn, the members’ agent gave advice to the Name. Hence the requirement placed upon the managing agent to prepare such reports, and hence also the link between notice to come off a syndicate and the delivery of such reports.

Mr. Bullen’s two most relevant reports provided the following information to Names, and I quote extracts, only:

A. Underwriter’s Report as at 31 December 1986 [published May 19, 1987]

For 1987 the Syndicate has doubled in size to a Gross Premium Income Limit of 23 million. We have a good trading position on the ground floor of the new Lloyd’s building and we have been shown and are therefore able to write considerably more, what I believe to be, profitable business.. . .Increased amounts of Cargo, Hull, and Hull related business have been written and numerically we have considerably reduced the difference in the number of risks between the direct and reinsurance accounts. The gap however has widened substantially when premium income figures are compared between direct and reinsurance business. Thus on a premium income the account is split 75% reinsurance and 25% direct business. The underwriting approach has, however, remained unchanged. My policy remains to underwrite fairly modest lines on a wide spread of business. As and when possible, I intend to look for ways to further spread our account on profitable business but without upsetting the balance we have built up over the past few years. For 1987 the Syndicate’s own excess of loss protections have been strengthened both upwards - we now have various layers up to a limit of US$64 million - and also sideways. These protections are placed mainly in the Lloyd’s market or with London Insurance Companies. We have no problems regarding bad debts on any of our protections.

B. Underwriter’s Report as at 31 December 1987 [published May 19, 1988]

We now write a direct consisting of hulls drilling rigs, cargo, liabilities (not primary), war (including confiscation risks), frustration risks and building risks. These risks have been written either facultatively or by way of treaties or covers and they provide about 20% of our income. The remaining 80% still comes from the excess of loss account which has been pruned and expanded over the years. Foreign accounts make up about 30% of this account.

It has always been my policy to write fairly modest lines on broker’s slips thus producing as wide a spread as possible. Naturally, especially on the excess of loss book, losses aggregate and therefore a constant eye is kept on the potential large loss situation to make sure that our own reinsurance arrangements are adequate. Being a fast growing Syndicate has meant that these have had to be changed annually. In 1987 we purchased protections on a whole account basis. . .As well as this, some specific reinsurance arrangements were made to protect the hull cargo and drilling rig accounts.

For 1988 we have increased our stamp capacity to 28,000,000. It is my intention to write an annual income of some 18 million. With much more marine capacity around this might prove difficult. The underwriting policy is unchanged except I have decided to write a small book. . .Of selected non-marine business away from the working areas. . .Our general protections have been suitably changed to cater for the increased capacity with protections up to 39,500,000.

On the face of these reports, Mr. Bullen was representing himself as an underwriter who wrote modest lines with a wide spread of business; that a careful eye was kept on the risk of aggregation and that he had ensured that the reinsurance arrangements were adequate. There was no warning to Names in the reports to indicate that he was deliberately exposing them to any loss. On the basis of the published information, there would have been no grounds for believing that this syndicate was deliberately exposing its Names to a risk of loss equal to its stamp capacity. (*480)

The next source of information would be, first, the inquiries which a members’ agent ought to have carried out of this particular syndicate and, second, the perception of LMX syndicates more generally.

In the Feltrim action there had been a dispute between Mr. Outhwaite and Mr. Wilshaw as to the duties of a members’ agent. As I have already said, and as will have become apparent from repeated references in this judgment to them, this case was, in a sense, something of a “re-run” of the Gooda Walker and Feltrim actions. Thus, experts who had given evidence in Feltrim, as had both Mr. Outhwaite and Mr. Wilshaw, were cross-examined about what they had said on previous occasions in reports to the Court or in evidence. A bank of material is being accumulated. One of the curiosities of the expert evidence is that the plaintiffs wish to rely on what Mr. Outhwaite, a defence expert, has said; and the defendants, in turn, wish to rely on what Mr. Wilshaw, a plaintiffs’ expert, has said in other cases, including Feltrim and a portfolio selection case. I have tried to distinguish between evidence which I received, and evidence given on other occasions, but I must confess that this has not always been easy.

Mr. Wilshaw had said, in a report he prepared for the Feltrim action:

The meeting between underwriter and Members’ Agent was the occasion for the underwriter to explain his underwriting philosophy and his general approach to life. I would have expected the underwriter to go through all the classes of business which he wrote and to underline those classes where there was a particular hazard or where there was some unusual feature. It was up to the Members’ Agent to probe in those areas where he felt uncomfortable or uncertain and at the end of it all it was up to the Members’ Agent to decide how comfortable he was with what he had been told. If he did not like what he had been told then he would not support the syndicate. It was not a question of the Members’ Agent trying to reunderwrite the account or influence the underwriter or even to second guess the underwriter. If these meetings were to be of any use for the Members’ Agent, then the Members’ Agent had to have knowledge about the classes of business concerned.

The passage was put to Mr. Outhwaite in cross examination, in these proceedings, together with Mr. Outhwaite’s own response to that paragraph:

[Q.] Mr Wilshaw’s report has been put in evidence. It was in bundle W and there was a certain amount of discussion around it when he gave evidence. What I would like you to do, if you would, is go to page 2 of this report and look at paragraph 3. [A.] Yes. [Q.] Where you say:

I have considerable experience from an underwriters perspective or what Members’ Agent were asking in 1989 and what their concerns were. I can recall only one agent who, by the end of 1989, asked me specifically about our aggregate exposures and reinsurances effected. Mr Wilshaw appears to be suggesting that Members’ Agents should have pursued a level of inquiries and depth of investigation which they did not do at that time. Equally, I do not recall any Members’ Agents going through each of the classes of business underwritten and asking questions about them as Mr Wilshaw suggests in paragraph 14 of his report. What Mr Wilshaw there says exaggerates and formalises what, in my experience, occurred at meeting held between underwriters and Members’ Agents. The conversations that took place between underwriters and Members’ Agents were either of a much more general nature, or else they were to discuss some specific matter or unusual feature which had arisen in the market or to discuss some problem specific to the syndicate.

Do you see that? [A.] Yes.

(4.00 pm)

[Q.] If you want to place it in context, we can go back to bundle W, if you could be provided with that, which is Mr Wilshaw’s evidence in Feltrim and the portfolio selection cases. At pages 156 and 159 of this bundle, we will see the sections of his report on which you were focusing. [A.] Yes.

[Q.] In particular, if you look at page 158, paragraph 13, Mr Wilshaw is then setting out what he believes a competent Members’ Agent should have asked about gross net aggregate exposure as a percentage of syndicate stamp? [A.] Yes.

[Q.] What you are saying, Mr Outhwaite, is that: “Well, I cannot recall any Members'' Agent asking me those questions until late in 1989”? [A.] Yes. Neither did our agency ask those questions of other syndicates.

[Q.] Correct. In paragraph 14, Mr Wilshaw then says what he believes a Members’ Agent should have asked at the standard or typical meeting between the Members’ Agent and the active underwriter where three lines down he says:

I would have expected the underwriter to go through all the classes of business which he wrote and to underline those classes where there was a particular hazard or where there was some unusual feature. It was up to the Members’ Agent to probe in those areas where (*481) he felt uncomfortable or uncertain and at the end of it all, it was up to the Members’ Agent to decide how comfortable he was with what he had been told. If he did not like what he had been told, then he could not support the Syndicate. It is not a question of Members’ Agent trying to re-underwrite the account or to influence the underwriter or even to second-guess the underwriter. If these meetings were to be of any use for the Members’ Agent then the Members’ Agent had to have knowledge about the classes of business concerned.

You say, we can go back to your report, that is paragraph 14:

Equally, I do not recall any Members’ Agent going through each of the classes of business underwritten and asking questions about them as Mr Wilshaw says in paragraph 14 of his report.

What you are saying, in a word, is this requiring far too much of the Members’ Agent in early 1989? [A.] What I am saying is that it was not my experience of the way meetings went between Members’ Agents and ourselves at that time.

[Q.] Yes. You are saying that as far as you were concerned, this is requiring rather too much? [A.] Yes.

[Q.] That was - [A.] I do not say it is requiring too much, in hindsight it would be fine, but at the time it did not happen.

[Q.] Yes. If we then go back to your report at paragraph 5 on page 3 you say that:

A greater degree of hindsight is, I believe, being applied by Mr Wilshaw in what he suggests was or should have been the level of awareness of Members’ Agents as to how LMX business was underwritten and, indeed, as to the nature of the LMX spiral. After 87J and Piper Alpha there was a gradual learning curve for Members’ Agents and they began increasingly to appreciate how the LMX spiral operated in practice.

That is an opinion which you would stand-by today, is it not, Mr Outhwaite? [A.] Yes, it is.

[Q.] Paragraph 6:

It was obvious to those who were aware of the nature of the LMX market and it was recognised by excess of loss underwriters after Alicia in 87J that upper layers were vulnerable to losses as were other layers and that the spiral could increase the gross loss by many times.

Pausing there, what was it about Alicia that caused those who were operating in the market and aware of it, to become aware that upper layers were vulnerable to loss? [A.] In 1988?

[Q.] Yes. [A.] Claims were beginning to be advised in the marine market which were going up to quite high levels.

[Q.] Right, If I could ask you then to drop down to the last four lines in that paragraph 6, do you see four lines from the bottom:

“Further, if any such information. . .”

We better take if a few lines further up, otherwise we will not know what you are talking about.

Mr. Justice Morison: “Generally speaking . . . “

Mr. Crane: “Generally speaking Members’ Agents did not ask underwriters about their calculation of the PML and the level of aggregate exposure. If they ever did ask for such information it would have been in the most generalised manner. In addition, this was information that underwriters would not have made readily available, as is recognised by Mr Wilshaw in the third sentence of paragraph 13 of his report. Further, if I any such information had been provided, a Members’ Agent could not possibly superimpose his own judgment on an underwriter’s perception of a Syndicate’s aggregate potential, nor could he assess the adequacy of the reinsurance programme to any given event”.

Is this the position: basically the Members’ Agent could not superimpose his own judgment on the underwriter’s perception of the Syndicate’s aggregate exposure to loss or the adequacy of its reinsurance? [A.] Yes.

[Q.] If the underwriter of a syndicate were to tell a Members’ Agent: “My reinsurance is, in my judgment, adequate for the potential large loss”, the Members’ Agent would have to believe that unless he had come to doubt either the veracity or the competence of the underwriter, that is right, is it not? [A.] If he tells the agent in specific terms that his reinsurance is large enough so that it covers effectively his PML in the parlance which we have been using -

[Q.] They would not use that language, that is the point you are making? [A.] No, he would not use that language.

[Q.] He would say his potential large loss, would he not? [A.] His potential large loss, yes, he may. If his reinsurance is at least as large as his potential large loss, then the agent could not possibly query that.

[Q.] That is right. That agent would have to infer that in the judgment of this underwriter, (*482) he had covered by reinsurance what we now call his PML exposure? [A.] He would from that.

It seems to me that I can properly conclude from the expert evidence before me that whilst it was the duty of members’ agencies to advise their Names about the choice of syndicates, they were not required so to question the underwriter of any particular syndicate as to endeavour to second guess the way the risks were written. Further, it would have been unlikely before late 1989 that any members’ agent would have asked specific questions about the underwriter’s PML or whether there was any deliberate exposure. But there probably would, or should, have been a fairly general discussion about the reinsurance protection which the underwriter obtained, and his attitude to its adequacy. There is no significant direct evidence before me as to what any members’ agent would have been or was told about the syndicate. There was some anecdotal evidence from a selection of Names, but I have disregarded this, as I must assume that they were competently advised; and that Mr. Bullen was entitled to assume that any one of his Names was as cognisant of the risks he ran as any other. But what is available tends to show that Mr. Bullen was putting himself forward as a conservative underwriter; he believed, wrongly, that he had obtained adequate reinsurance of his account. As it was put by RTY in their submissions to the Boatman Inquiry.

Mr Bullen and his staff at all material times believed that in the light of the factors referred to above this reinsurance protection was adequate for all foreseeable Market catastrophe losses. Mr Bullen made this quite clear in his underwriting reports.

As I read that statement, Mr. Bullen was saying that he had got reinsurance protection for his PML event. In my judgment, it is unlikely that any specific inquiries of Mr Bullen or of RTY would have revealed that this syndicate was running a substantial amount of exposure net for the Name’s account. This conclusion is reinforced by what Mr. Bullen is recorded as having said to a members’ agent in May, 1988, the relevant part of which reads:

NTWB [Mr Bullen] explained the problem that could be caused by reinsurance spiral but was confident that his reinsurance programme was adequate to cover such an eventuality.

He confirmed to the Court that he saw himself as a conservative underwriter and would say to the Names when he met them. He was never, at the relevant time, asked by RTY what his underwriting policies were. RTY were, therefore, in no position to give any Name any kind of special warning about this syndicate; it did not know, because Mr. Bullenhad not told them [and if he had, it would have beenuntrue] that Names were exposed to 100 per cent.loss.

I, therefore, turn to the last matter, which is the fact that the market, generally, appreciated that syndicates in the LMX market were “high risk”. That was common ground. But what does it mean in terms ofdeliberate exposure to loss? It seems clear from the evidence of the experts that within the LMX “genus” there are many “species”. Mr. Outhwaite accepted that the amount of deliberate exposure which an underwriter in an LMX syndicate might run would vary from syndicate to syndicate. It is significant, I think, that he himself reinsured up to the full amount of his PML, knowing, as he did, of the risks inherent in the spiral.

Mr. Crane. . .Mr. Outhwaite, this underwriter will have, as part of his plan, for the given year of account, an idea of how much of that PML exposure he intends to run net, will he not? [A.] Yes.

[Q.] As you told my Lord 10 or 15 minutes ago, that amount varied from LMX Syndicate to LMX Syndicate? [A.] Yes.

[Q.] In some cases it was nil? [A.] Well, you could not actually run nil, but what I am saying is that they reinsured vertically up to or in excess of their PML.

[Q.] In other words, they covered entirely their PML exposure? [A.] Yes, that does not mean to say they were running no risk.

[Q.] There was the possibility that their judgement of the PML exposure, whether made competently or incompetently would prove wrong? [A.] There was that factor, but of course there was also retentions, there was also the fact that you could not cater for numbers of catastrophes either.

In various passages in his evidence, Mr. Wilshaw agreed that there were differences between syndicates.

The picture which I have, therefore, is of a conservative, but wrong-headed, approach to risk taking in a pre-eminently risky segment of the market. High rewards for high risks. But the reason why the business was risky was largely for the reasons which Mr. Outhwaite gave, which are unrelated to vertical exposure. I have found this a very difficult question to judge. The underwriters’ reports in the Feltrim case were no less “conservative” in feel. On the other hand there has been no evidence to suggest that every LMX syndicate could be expected to run a net exposure of 100 per cent.; indeed, the opposite was the case. There was no evidence that, had inquiries been made of RTY, (*483) Names would have been told of a policy of 100 per cent. exposure. Indeed, the failure by the defendants to call any witness which would have helped on this issue is not without significance. There are documents in the files which show that Mr. Horrex, RTY’s managing director, was himself quite flabbergasted when he came to appreciate, after Piper Alpha, how the business had been written.

At the end of the day, but for quite different reasons to those advanced by them in their pleadings, I have come to the conclusion that for this syndicate an underwriter who was acting competently could not have assumed that a Name had accepted that he would be deliberately exposed to a loss greater than 40 per cent. of stamp. In other words, he could anticipate, purely by reason of them being known to be high risk syndicates operating in the LMX market that such a deliberate exposure was on the cards.

I recognize that this is a different conclusion from that reached in the Feltrim case. But such a difference is justified, in my judgment, because Mr. Justice Phillips had specific evidence that Names’ agents would be told or would know that the syndicates were running a net 100 per cent. exposure. In Feltrim, Mr. Fagan gave this evidence:

Later he emphasised that he planned the reinsurance cover so as to try to limit the potential loss on a PML basis to the names to no more than about 100% of stamp. If asked, he would always tell agents that there was potential for a 100% loss.

Feltrim’s chairman [Mr. Andrew] gave evidence to this effect:

He said that there was a policy of the Board with regard to running net amounts of exposure and that agents would know that Names were exposed to an unreinsured loss of 100% of stamp.

In my judgment, that evidence was important to the conclusions which the Judge reached. The evidence in this case was to a very different effect, as I have indicated.

I cannot accept the defendant’s contention that the true figure is 100 per cent. [or 120 per cent. to take account of the profit which could have been expected]. That submission was founded partly on the basis that I should not depart from what Mr. Justice Phillips had concluded, and partly upon the evidence of Mr. Outhwaite. I justify departing from that decision on the basis that he received different evidence from me. As to Mr. Outhwaite’s evidence I have looked at it with some care.

In his report the said that:

The amount of PML that an underwriter chooses to run is a matter of subjective judgment. It cannot be a matter of precise calculation because the factors against which it is set are not fixed, but are themselves subjective. It is not possible for me as an Expert Witness, to say in a particular case that it was 150% or 200% of capacity. Nevertheless, taking all the points I have discussed into account, I believe that an Excess of Loss Syndicate cannot be criticised for retention of at least 120% (of capacity) in the event of one major occurrence based on its PML estimates.

This passage in the report, taken on its own, does no more than justify the categorization of an LMX syndicate as high risk. What it does not address is what a Name could be said to have consented to when he or she joined, or remained on, the syndicate for the relevant years. That, it seems to me is a different question and can only be answered in the light of the factors to which I have referred. Most importantly, what a Name can be taken to have consented to must relate to the syndicate in question in the light of the way it was presented; otherwise, Mr. Outhwaite must be saying that regardless of what any LMX syndicate actually did or said about itself, a Name must always expect the underwriter to retain a net exposure of 100 per cent. of stamp. That cannot be right, and was not a position adopted by him in his oral evidence. It may very well be that every Name can expect that his overall loss potential for any year of account is 100 per cent. of stamp, simply because it is a high risk category. Such losses may occur, and in this case may do so even after damages have been paid, even if there was only a limited deliberate net exposure. Once it is accepted that it is a question of “consent” and that different LMX syndicates had different net exposures, general statements about the loss-making potential of such syndicates is not determinative of the answer.

However, Mr. Outhwaite justifies his conclusion by reference to the “points” he has made earlier in the report. As I understand it, the thinking behind his conclusion relates to the expectation of a Name that on a high risk syndicate he will receive high profits in the good years which should be “spread” over a bad year. He gave some figures showing, in very rough terms, how this might work. The basis upon which he expresses this opinion is, essentially, related to the concept of “pay-back”. Lloyd’s syndicates are so structured that they are not able to create “reserves”, in the sense of putting aside a portion of profit earned in one year, for setting against losses on business written in a subsequent year. Thus, the syndicates cannot “spread” their results over a period of time. For this reason, I accept that Mr. Outhwaite that a properly advised Name would stay on such a syndicate for a number of years. On the basis that catastrophes do not happen every year, an LMX syndicate will tend to (*484) experience ups and downs. To some extent, the downs are compensated by another aspect of “pay-back”, namely an increase in rates in the reinsurance market immediately after a catastrophe has occurred, such as happened after Piper Alpha.

It seems to me, therefore, that it is an inherent feature of the LMX market that an underwriter will or may plan his account so as to expose his Names deliberately to the risk of loss, relying on good years to create a “balance” or “spread” in his results. The period of time over which a Lloyd’s syndicate would be entitled to seek to balance its profits and losses is open to question. I see no need to depart from the period accepted by Mr. Justice Phillips in Feltrim and I, therefore, take the period as about five years, but it could be as high as 10, which is the period taken by Mr. Outhwaite.

Mr. Outhwaite gives some figures which are designed to justify his conclusion, based upon the average profit a Name might expect from participation in a high risk syndicate, the probability of a catastrophe occurring and the average return of all Names in the Lloyd’s market. Accepting that his methodology is somewhat crude and could provide for a wide variety of possible results, I have found it instructive to apply it to the facts in this case.

The results of the syndicate showed, over a seven year period from 1980 to 1986 inclusive, a profit to the name per share of just under 11 per cent. which is marginally above the Lloyd’s average return for that period. Mr. Outhwaite said that he would anticipate a 20 per cent. return for an LMX syndicate. A properly advised Name would be entitled to conclude, I think, that the lower levels of profit on Syndicate 255 were consistent with Mr. Bullen’s conservative underwriting. Using Mr. Outhwaite’s methodology, and allowing for the fact that the syndicate started in 1980 and, thus, the earliest years may not be a reliable point from which to extrapolate, one might assume an annual return for a Name of 12 per cent. reducing to, say, an average of 8 per cent. [Mr. Outhwaite’s figure] after a catastrophe. Again, taking Mr. Outhwaite’s hypothesis of a catastrophe occurring, say, every 10 years, the maximum deliberate exposure to loss would be of the order of 25 per cent. [10 years at 8 per cent = 80 per cent.; 9 years at 12 per cent. = 108 per cent.; 108 per cent.- 80 per cent. = 28 per cent.]. As he said, the possible variation in the equations are endless. However, his methodology does give me a sufficiently clear means by which I can reach my conclusion. 40 per cent is consistent with Mr. Wilshaw’s evidence and represents, in my judgment, a reasonable and fair conclusion.

Finally, I should add that I do not regard the fact that Lloyd’s institute a Loss Review Inquiry when losses equal or exceed the stamp capacity as being of much assistance. Poor underwriting may take manyforms. If there were a link between a deliberate netexposure and a net loss for the year, one might havethought that Lloyd’s would not wish to commence aninquiry whenever a loss reached 100 per cent., if the Society accepted that such a level of loss was to be expected from every LMX syndicate.

B. The special priority treaty

The application of the principles to this head of claim has caused some controversy.

The plaintiffs have suffered loss as a result of bad debts, namely their inability to recover from some of the participants to the treaty. The contention which the plaintiffs make in their closing submissions was to the following effect:

To justify writing the extra risks which he declared to the Treaty, he needed to be confident that he would not end up retaining more than approximately 20% of the aggregate which they generated. . .as a matter of law, it is submitted that the Plaintiffs are entitled to be compensated for all the bad debts arising on the Special Priority Treaty, notwithstanding that the Plaintiffs accept that a competent underwriter could in principle have properly ceded at least a limited volume of aggregate to Aneco and Korea Foreign. On Mr Bullen’s evidence, the choice was not between retaining approximately 50% of the aggregate from the risks declared to the treaty, or retaining about 20% of that aggregate. Effectively it was between being able to cede away something like 80% of the aggregate, or not writing it in the first place. Since the former was, on the evidence, something which a competent underwriter would have appreciated was not an option, applying orthodox principle [sic] of damages means putting the Plaintiffs into the position they would have been in had the additional risks declared to the [SPT] not been written at all. . .it also follows from Mr Bullen’s evidence, as a matter of law, that the Plaintiffs are entitled to recover the bad debt arising from the [SPT] even in relation to a reinsurer upon whom - in isolation - a competent underwriter could have relied to the extent that Mr Bullen did. Thus, if eg contrary to the Plaintiffs’ submission, the Court were not persuaded that a competent underwriter would have felt unable to cede $6.9m of aggregate to Korea Foreign (or if the claim is statute-barred), the bad debt attributable to that company would still be recoverable on ordinary principles of damages - ie as a loss which foreseeably arose in consequence of Mr Bullen writing the additional aggregate which a competent underwriter would not have written. (*485)

In other words, the SPT business cannot be apportioned between the securities; Mr. Bullen would not have written the extra business had he not had the treaty; ergo the plaintiffs can recover all the losses which are attributable to Mr. Bullen having entered into the treaty with anyone. As I understand it, they are saying that they should be compensated on the basis that this is a no-transaction case: that is to say, on the basis that the additional risks ceded to the treaty had not been written at all. Alternatively, they are to be compensated on the basis that the Names were to be indemnified against bad debts due from competently selected reinsurers.

Mr. Hirst’s first objection to this claim was to suggest that, as an issue, it fell outside the ambit of the issues which I had to try. He said that the only damages issues were those which had been detailed in the order I made in November, and that no other such issues could now be tried. He said he was surprised by the plaintiffs’ contention. He also submitted that their approach was wrong in principle.

It seems to me that this is an issue which falls within the ambit of the issues which I should now determine. The question of liability is inextricably bound up with the question “what would have happened if Mr. Bullen had not been negligent”. Nor do I consider that Mr. Hirst should have been taken by surprise, as I think the point was (just) sufficiently signalled by the opening.

However, in my views he was right to say that the plaintiffs’ case is not open to them on the pleadings.

Before the points of claim were amended, the sole complaint in relation to the SPT related to INX Re. It was and is the plaintiffs’ complaint, which I have upheld, that INX Re was not an organisation which any competent underwriter would have accepted as part of the security for the SPT. In the further and better particulars of their claim, the plaintiffs responded to a request to state how it was that the alleged breach of duty caused the plaintiffs any loss. The reply to that request was:

The Plaintiffs’ loss, therefore, is the loss sustained by reason of the failure of the INX Re to reimburse the Syndicate under the [SPT]

It is to be noted that the plaintiffs did not say that they were entitled to be compensated on the basis that the aggregates ceded under the treaty would not have been written in the first place, nor do they seek to say that the losses arising from other participants in the treaty failing or refusing to pay up are attributable to Mr. Bullen having chosen INX Re.

When the points of claim were amended to include the allegation that, although Mr. Bullen was entitled to use Aneco and Korea Foreign he should not have ceded as much to them as he did, there wasno withdrawal of the particulars to which I have justreferred. There is nothing in the amended case which, in principle, would give rise to any different measure of damage from that previously particularized.

Based on the pleaded case, the plaintiffs are entitled to be compensated on the basis that had INX Re been a sound participant the syndicate would have been able to recover from it such sums as are due; and in relation to Aneco, they are entitled to recover such sums as are due but unpaid in relation to claims in respect of that aggregate exposure ceded beyond what I have held to be the acceptable limit.

For what it is worth, had the plaintiffs amended their claim to put their case on damages as they have advanced it in their closing submission, I would have rejected it. There are great practical difficulties, as I have said, in seeking to re-write Mr. Bullen’s book of business. In my view, if Mr. Bullen had known of the limitations of INX Re and Aneco, he would have obtained other participants who were sound and to whom he would, hypothetically, have ceded the same business. This was not a “no-transaction” case, on the facts. Further, the alternative claim would have the effect of transferring the risk that a good security might become insolvent from the Names to the defendants. There is no legal principle, of which I am aware, which enables this to be done as part of an award of damages. The losses which flowed from Korea Foreign’s inability or unwillingness to pay were not due to Mr. Bullen’s incompetence but were due to the risks inherent in competently choosing any security. Those losses had nothing to do with this incompetent choice of INX Re or of Aneco.

6. ESTOPPEL

The estoppel point concerns six working Names who are plaintiffs and who are identified in Schedule D to the second defendants’ amended points of defence. One of the persons [Dr. Damant] identified in the Schedule is not in fact a working Name and the estoppel argument was not pursued against him.

The estoppel point is clearly identified in the pleadings at par. 48:

(1) Certain of the Names specifically requested (unprompted by their Members’ Agent) to be placed on the Syndicate. . .

(2) When each Name made such request, he impliedly represented to his Members’ Agent that he knew enough about the Syndicate’s underwriting practices to be able to make an informed judgment that (*486) the syndicate would be an appropriate part of his Portfolio.

(3) In reliance upon such representation, the Members’ Agent placed the Name upon Syndicate 255 and lost the opportunity to place the Name on a different Syndicate or leave a portion of the Name’s premium income limit unallocated.

(4) It is unjust that any of these Names should now allege against his Membes’ Agent that the underwriting practices adopted by the Syndicate were negligent and that his Members’ Agent is contractually responsible to him for this negligence.

(5) In the premises, as against his Members’ Agent each of these Names are [sic] estopped from making the allegations of negligence contained in the points of Claim.

I am bound to say, with great respect to Richard Slade who argued this point with skill, that it seemed to me that this submission were best told to the marines.

There is nothing inconsistent, or unfair or unjust, in a Name choosing to go on a syndicate and thereafter enforcing his rights against the members’ agent under the contract which he must make with the agent. A Name cannot go on a syndicate otherwise than through the agency arrangements to which I have referred. Under the contract, the agent warrants that the syndicate’s business will be written competently and carefully. As a quid pro quo for their contractual obligations the agencies charge the Name commission. By choosing which syndicate he will join, the Name may absolve the agent from his responsibility to advise him about his portfolio selection; but I can see no warrant for the idea that the Name also represents that he will not hold the agent responsible for the underwriter’s incompetence. The agent receives his commission under the contract whether or not the Name chooses the syndicate. Mr. Slade did not feel able to support his alternative plea of volenti non fit injuria, and rightly so. The facts do not establish either that the Names consented to the lack of reasonable care [the volenti defence] nor represented that they knew of the matters giving rise to claims against RTY and the agencies [the estoppel defence]. The agents are liable under the contractual arrangements whether or not they could be blamed for not discovering that a syndicate’s business was being badly managed. If an agent can be liable even if not at fault, I fail to understand why it is injust for him to be held liable when he was not at fault in choosing the syndicate in the first place, whether that choice was made by the agent or by the Name.

7. LIMITATION

I can deal with this point briefly. The unamended points of claim complained about Mr. Bullen having accepted INX Re as a participant in the SPT. There were also a claim made against RTY that they failed to monitor or supervise the adequacy of the protection provided by the SPT “and the level of security represented by the reinsurers subscribing to that treaty.” The claim against RTY was, therefore, in wider terms and referred to all the reinsurers. It seems to me that the original pleading opened up the possibility of an investigation into the facts relating to participants in addition to INX Re. As was accepted in evidence, the exposure ceded to participants had to be looked at collectively. Thus, both as a matter of fact and as a matter of pleading, the adequacy of the security provided by the other participants was likely to be in issue had the pleadings not been amended.

It seems to me clear that there is a real overlap between the amended allegations and the original ones in relation to the SPT, and that what is new arises out of the same or substantially the same facts as were already in issue. To use Mr. Hirst’s phrase, the proof of the pudding is that the defendants have had no difficulty in dealing with the allegations on their merits. That is, I think, because the new and the old were so obviously all part of a piece that it was convenient that they should all be looked at together.

Finally, the same point arises in relation to the matching reinstatement point. Here, whilst it can be said that there is a substantial difference between horizontal and vertical exposure, in truth the non-matching reinstatement point was simply one small aspect of the claim for general incompetence being made against the defendants. His reinsurance programme was under scrutiny from the word go. Again, it seems to me that there was a sufficient factual overlap to say that the new claim arise out of, or substantially out of the same facts as the old claims.

Accordingly I hold that there is no limitation defence available in relation to the amendments to the points of claim.

8. PRINCIPLES RELATING TO QUANTUM

In the event, only one of the issues which I ordered to be tried remained the subject of contention. But, because the defendants reserved their position on others, it is as well that I set out each of the issues and the way they are to be dealt with.

1. Do the plaintiffs have to give credit for the benefit of any profits made by them in previous years as alleged in the amended points of defence? (*487)

The defendants did not argue this point, in the light of the Court of Appeal’s judgment in Brown v. KMR Services Ltd., but reserved their position “for a higher court”. In my view, there is no merit in the point but I do not decide it.

2. Do the plaintiffs have to give credit for the benefit of any profits made by them across the whole of their portfolio for the two relevant years of account?

This contention was abandoned, in my view rightly.

3. Do the plaintiffs have to give credit for the cost of any additional insurance which Mr. Bullen should have purchased?

The answer is “yes”, as I have already indicated. This point was not argued by the plaintiffs as a separate point from their main submission that the proper measure of damages was not based on the book as written.

4. Do the plaintiffs have to give credit for the benefit of any reinsurance which ought not to have been written in favour of other syndicates on which any of the plaintiffs participated for the years of account in question?

In my view the answer to this question is “No”. In the first place the plaintiffs’ damages are not calculated by re-writing Mr. Bullen’s book of business. The assessment of damages will not involve saying that any reinsurance contract should not have been written. Further, the proposition that Names would have been more exposed to loss on the other syndicates had certain contracts not been written by Mr. Bullen, pre-supposes that the other syndicates would not have obtained the same reinsurance from another source. Such pre- supposition is without foundation. Therefore, there is no factual basis to support any off-set. There is no basis in law for any offset, either. It would be a practical impossibility to have a partial unwinding of the spiral. The law does not require parties to engage in a process which, as here, would be practically impossible to conclude: Jebsen v. East and West India Dock Co., (1875) L.R. 10 C.P. 300. Although a very different case on its facts, the principle is clear. Further, the resulting alleged benefit is to remote, in my view, to be treated as an off-set to the damages. It would be pure happenstance whether a Name was “on” another syndicate which was re-insured by Mr. Bullen who would have been quite unaware of the identity of the Names of any syndicate which was seeking re-insurance protection through Syndicate 255. There is no link, in terms of causation (or foreseeability), between Mr. Bullen’s negligence and the fortunes of 255’s Names on other syndicates. The answer to the question is “No”

5. Do the plaintiffs have to give credit for the benefit of any tax refunds or allowances in respect oflosses suffered by individual Names as a result of their participation on Syndicate 255 for the relevant years of account?

Although the question was answered in the negative by the Court of Appeal on the Gooda Walker appeal, the defendants asked to reserve their position as the matter was going up on a further appeal. The House of Lords has dismissed the appeal. The point is now decided against the defendants and the answer to the question is “No”.

6. Do the plaintiffs have to give credit for the benefits of any stop loss policy insurance they may have received for the relevant years of account?

This question was answered in the negative by Mr. Justice Phillips and Mr. Justice Gatehouse in Gooda Walker and Feltrim and Sword Daniels, respectively. The defendants abandoned it. The answer to the question is “No”.

Finally, in his closing submissions, Mr. Crane invited me to make it clear that, if, as I have, I adopted the approach to the question of damages on the main issue on the basis of the purchase of additional notional re-insurance, that additional reinsurance should fill the gap above the existing cover from the bottom upwards. In Feltrim, as I would hope in this case, the purport of Mr. Justice Phillips original judgment was obvious. Nonetheless, a further hearing was required at which he was compelled to spell out more precisely what must have been obvious in the fire place. The need for this point to be aired further was because the underwriters sought to argue, after his judgment, that the competent underwriter could leave the “100% gap or gaps wherever he chose without being open to a charge of incompetence”. Mr. Hirst said that I should not give a ruling on this point because it was not covered by any of the specific issues on quantum. I reject his submission. This is a point which affects causation rather than quantum; but whatever its nature, it is absurd that I should leave open a doubt about the import of my judgment, so that the litigation between the parties may be unduly protracted. I refuse to leave obscure a conclusion which I have reached. I make it clear (if it is not already clear), as did Mr. Justice Phillips, that:

. . .it is implicit in my judgment that the additional cost of reinsurance should fill the gap above the existing cover from the bottom upwards. . .