See Law Reports version at [1995] 1 W.L.R. 1206



COUNSEL: Andrew Smith Q.C. and David Lord for the plaintiffs.

Bernard Eder Q.C. and Simon Bryan for the defendants.


SOLICITORS: Wilde Sapte; Elborne Mitchell.


JUDGE: Phillips J.


DATES: 1995 March 27, 28, 29, 30; April 6



[*1208] Cur. adv. vult.



6 April. Phillips J. handed down the following judgment. This judgment deals with a number of issues of principle that fall to be resolved in relation to the assessment of damages.


Is it appropriate to assess damages at this stage?


Mr. Eder submitted that, having particular regard to comments recently made by Saville L.J. when this and other Lloyd’s proceedings came before the Court of Appeal (unreported), 3 March 1995, it is not appropriate to embark on the assessment of damages at this stage. Saville L.J. said:


“I think the only policy, as I recall it, when I set up the management system was that the point generally about first past the post was indeed adumbrated. I in fact raised it and I think I did point out to everybody that there was at least a possibility that victory first time round did not necessarily mean victory in total, that there would very likely indeed be a second round of litigation in which the question as to who scooped the pool (as we called it) or pools would arise and I simply gave that warning. If my memory is correct, in order not to pre-empt that question the orders and directions I gave, now nearly two years ago, were all orders or directions for the hearing of trials which would not end on my directions in final money judgments.


These comments do not give, and clearly were not intended to give, any guidance as to what course should be followed by judges who are now faced with the implications of competing claims to limited errors and omissions (“E. & O.”) recoveries. When dealing with an application for interim payment in this action on 14 February 1995 and faced with a similar argument advanced by Mr. Eder, I ruled that it was not appropriate to be influenced by the effect that making or refusing the order sought would have on claims to E. & O. funds. I remain of that view and accordingly, subject to the next point raised by Mr. Eder, I propose to proceed with the assessment of damages in this action.


The effect of Society of Lloyd’s v. Clementson: the broad point


Mr. Eder submits that I should not proceed to assess damages in this case until judgment has been given in Society of Lloyd’s v. Clementson, 1992 Folio No. 1820. In that action Lloyd’s seek to recover from Mr. Clementson moneys paid out of the central fund pursuant to paragraph 7(a) of the Central Fund Byelaw (No. 4 of 1986, 14 July 1986) in order to make good defaults by Mr. Clementson in meeting his obligations to policyholders under contracts of insurance written on his behalf. Mr. Clementson is a plaintiff in this action. The damages he claims in this action include liabilities to policyholders which have been discharged by part of the payments from the central fund which Lloyd’s [*1209] are seeking to recover from Mr. Clementson. In this action it is asserted by Mr. Clementson, and other Names who have had liabilities discharged by payments out of the central fund, that they are liable to reimburse the central fund in respect of such payments. That is not the stance taken by Mr. Clementson in the action brought against him by Lloyd’s. By his amended points of defence and counterclaim Mr. Clementson pleaded that the following contravened article 85 of the E.E.C. Treaty (Cmnd. 5179-II) and were void: the decision of the Society of Lloyd’s to adopt the Central Fund Byelaw; the decision to make payments out of the central fund in respect of alleged debts of the Names under paragraph 7 of the Central Fund Byelaw; the decision to sue the Names for reimbursement under paragraph 10 of the Byelaw; the underwriting agency agreements in the standard form required by Lloyd’s Agency Agreement Byelaw (No. 1 of 1985); and that in consequence the payments out of the central fund were not recoverable. The Court of Appeal in Society of Lloyd’s v. Clementson [1995] C.L.C. 117 has held that the defence based by Mr. Clementson on the E.E.C. Treaty is not unarguable.


In Society of Lloyd’s v. Clementson Mr. Clementson has now substituted a more detailed points of defence and counterclaim running to some 50 pages. This alleges that the Lloyd’s arrangements, as therein defined, violate article 85 of the E.E.C. Treaty. The Lloyd’s arrangements include (i) the byelaws promulgated by the Council and Committee of Lloyd’s pursuant to the Lloyd’s Act 1982 and Lloyd’s Act 1911 and (ii) the undertakings given by the members of Lloyd’s to Lloyd’s upon their election to Lloyd’s pursuant to which the members agreed to comply with the provisions of the Lloyd’s Acts 1871 to 1982 (inclusive) and subordinate legislation made or to be made thereunder. The object of Mr. Clementson’s pleading is to demonstrate that he is under no liability to reimburse Lloyd’s for payments made out of the central fund. In order to achieve this end he appears to be attacking the very foundations of Lloyd’s. I am told that Mr. Clementson has the support of other members of Lloyd’s who have joined together to form what is commonly known as the “Writs Response Group.” They almost certainly include other plaintiffs in this action.


No issue arises in this action as to the legality of the Lloyd’s arrangements. The plaintiffs (“the Names”) contend that those names whose liabilities have been discharged out of the central fund are legally bound to reimburse Lloyd’s in respect of such payments. The defendants do not challenge this contention. If the transactions with which this action is concerned were, on their face, manifestly illegal, I should be bound to take account of that of my own motion. Mr. Eder does not suggest that this could be the case here. He submits, however, that, were “persuasive and comprehensive” evidence of illegality to emerge, I should have to take notice of it. He suggests that, in circumstances where Mr. Clementson, supported by other names, is challenging the legality of the Lloyd’s arrangements in other proceedings, I should consider whether the correct course is not, of my own motion, to stay this action until the result of those proceedings is known. Whether Mr. Clementson has a good defence to Lloyd’s claim, as presently constituted in Society of Lloyd’s v. Clementson, on the basis of his plea of illegality will not be known until, at earliest, October. An appeal to the Court of Appeal and a reference to Europe are plainly possible. Even if his defence were to succeed it seems to me unlikely that it would do so on a basis that would require this court to refuse to grant the Names any further relief. Having [*1210] considered Mr. Eder’s invitation to stay these proceedings of my own motion pending the result in Society of Lloyd’s v. Clementson

I have no hesitation in declining to do so.


The narrow point


The Lloyd’s central fund has paid some 102m. in order to discharge obligations of the Names in this action in respect of the syndicate years other than 290/1990. A substantial, though unidentified and perhaps unidentifiable, part of this total will have been paid to the managing agents in order to discharge the liabilities to policyholders which form the subject matter of the relief sought in this action. Mr. Eder makes the following submissions: (1) the Names can make no recovery of liabilities discharged by Lloyd’s from the central fund unless they can demonstrate, on balance of probabilities, that they are obliged to reimburse the central fund; (2) because of the plea raised in Society of Lloyd’s v. Clementson, Names cannot at present establish that they are liable to repay Lloyd’s. In consequence the damages they recover must be reduced to take account of the liabilities discharged by payments out of the central fund. Mr. Smith joins issue with both these contentions.


Is liability to Lloyd’s a precondition of recovery?


The central fund, save in so far as it may include moneys borrowed by Lloyd’s, consists of contributions levied from members of Lloyd’s and investments and income derived from such contributions. Mr. Smith submits that, in these circumstances, the defendants cannot claim credit in diminution of damages for any liabilities of the Names discharged out of the central fund. The fact that the Names may prove to be under no duty to reimburse Lloyd’s for such payments is irrelevant. The payments are collateral under the doctrine in Parry v. Cleaver [1970] A.C. 1. In Parry v. Cleaver Lord Reid considered two situations in which a plaintiff does not have to give credit in favour of the wrongdoer for collateral benefits received from third parties – the proceeds of insurance and sums coming to him by reason of benevolence. He said, at p. 14:


“It would be revolting to the ordinary man’s sense of justice, and therefore contrary to public policy, that the sufferer should have his damages reduced so that he would gain nothing from the benevolence of his friends or relations or of the public at large, and that the only gainer would be the wrongdoer . . . As regards moneys coming to the plaintiff under a contract of insurance, I think that the real and substantial reason for disregarding them is that the plaintiff has brought them and that it would be unjust and unreasonable to hold that the money which he prudently spent on premiums and the benefit from it should enure to the benefit of the tortfeasor.”


Both of these situations are a long way away from the position where the third party, Lloyd’s makes payments for the benefit of Names on the false premise that the Names will be bound to reimburse Lloyd’s. I do not find my sense of justice outraged at the possibility that the defendants might benefit from such payments. None the less, on balance, I consider that Mr. Smith is correct to contend that such payments would not relieve the defendants of their liability in damages. I have dealt with this difficult point in a cursory manner because I do not consider that it is the primary answer to Mr. Eder’s contention. [*1211]


Are Lloyd’s entitled to reimbursement?


On the premise that the payments from the central fund have indeed discharged liabilities of the Names to policyholders, Mr. Smith has submitted that Lloyd’s have a basis for claiming reimbursement which is an alternative to their pleaded reliance upon the Lloyd’s arrangements. I can do no better than to quote from his skeleton argument:


“It is a general principle of equity that where by an unauthorised act of an agent (the managing agent) money of a third party (Lloyd’s) is obtained and applied for the benefit of the principal (the Name) the principal is liable to restore such money to the extent that it is so applied – Bowstead on Agency, 15th ed. (1995), art. 100, p. 407, Rolled Steel Products (Holdings) Ltd. v. British Steel Corporation [1986] Ch. 246, 300C, 307F, Bannatyne v. MacIver [1906] 1 K.B. 103 and Reversion Fund and Insurance Co. Ltd. v. Maison Cosway Ltd. [1913] 1 K.B. 364. (This is not strictly a matter of subrogation but an application of the principle of equity exemplified in B. Liggett (Liverpool) Ltd. v. Barclays Bank Ltd. [1928] 1 K.B. 48, 60.)”


I do not find that the facts of this case lie comfortably within the principle upon which Mr. Smith relies. None the less I consider that, if the Names recover damages in respect of liabilities discharged by Lloyd’s in an action where they assert that they are liable to indemnify Lloyd’s, it is unlikely that the principles of agency and of restitution will not provide Lloyd’s with a route to recovery that dispenses with the need to rely upon the Lloyd’s arrangements.


Once again I have dealt with a difficult point in a cursory fashion, for I now come to what seems to me to be the most simple and compelling answer to Mr. Eder’s submission. In this action no issue of illegality is raised. The defendants have no answer to the Names’ contention that they are under an obligation to reimburse the central fund pursuant to paragraph 10 of the Central Fund Byelaw. The fact that the Court of Appeal has ruled that there is an arguable defence in Society of Lloyd’s v. Clementson is no bar to the Names establishing in this action, where that defence is not asserted, that there is a liability to reimburse the payments made from the central fund. It follows that the payments made out of the central fund do not reduce the losses in respect of which the Names are entitled to claim damages.


The approach to damages


The point to which I now turn has been raised in respect of the primary losses claimed by the Names, that is liability to policyholders. This head of damage falls into two parts, claims that have been paid and claims that the Names anticipate that they will have to pay in the future. The Names contend that they are entitled at this stage to a final award of damages, covering both the paid claims and the anticipated claims. The defendants contend that the appropriate course is that my award should be restricted to the paid claims and that the assessment of damages in relation to anticipated claims should be deferred, subject possibly to a declaration that the Names are entitled to an appropriate indemnity in respect of such claims. In support of their proposed approach the defendants have referred me to two precedents.


In Brown v. KMR Services Ltd. [1994] 3 All E.R. 385 Gatehouse J. gave judgment in respect of what he described as the “present liabilities” of the plaintiffs and an indemnity against future losses. The nature of that [*1212] indemnity is not clear from the report of the judgment. Trans Trust S.P.R.L. v. Danubian Trading Co. Ltd. [1952] 2 Q.B. 297 was a case where buyers had wrongfully repudiated a contract for the purchase of steel. The sellers claimed that the buyers’ breach of contract had rendered them subject to the risk of a claim for damages from their own suppliers. At first instance [1952] 1 K.B. 285 McNair J. granted a declaration that the sellers were entitled to be indemnified in respect of any damages that they might have to pay to their suppliers. On appeal this head of claim was held to be too remote, but the court expressed opinions as to the appropriate procedure had this not been the case. Somervell L.J. said, at p. 303:


“The judge, in making the declaration which he did, followed with modification a declaration made in somewhat similar circumstances by Lewis J. in Household Machines Ltd. v. Cosmos Exporters Ltd. [1947] K.B. 217. The problem can be shortly stated. B sues C for breach of contract. The court holds that B is entitled as against C to recover damages in respect of B’s liability to A arising out of C’s breach of contract. At the time of the hearing B is not in a position to call evidence to quantify this damage. There may be some cases in which the court can state a principle which makes the subsequent quantification of this damage simple. On the other hand, difficult questions may arise, depending, for example, (1) on any variation of the terms of the contract between B and C as between B and A, (2) on the question whether A took the steps which should have been taken to mitigate damage. No declarations ought to prejudice or preclude a proper determination of these issues, on which the defendants should be entitled to be heard. It might, as it seems to me, be more satisfactory if there were liberty to apply for directions as to the determination of these issues, if any, and quantification of damages under this head as between plaintiffs and defendants, should disputes arise. Some order in this form, at any rate, in some cases, might be more satisfactory than a declaration in the form ordered.”


Denning L.J. added, at p. 307:


“If the liability of the sellers to a third party were within the contemplation of the parties, but had not yet been assessed, then the proper course for the judge was to reserve that head of damages. Judgment could be entered for the damages already ascertained, leaving the rest to be ascertained later by the same or another judge.”


Mr. Smith submitted that the course proposed by Denning L.J. was one which the court has no jurisdiction to follow and I must first deal with this point.




Mr. Smith referred me to relevant statements in two cases in the House of Lords. In Murphy v. Stone-Wallwork (Charlton) Ltd. [1969] 1 W.L.R. 1023, 1027, Lord Pearce said:


“Our courts have adopted the principle that damages are assessed at the trial once and for all. If later the plaintiff suffers greater loss from an accident than was anticipated at the trial, he cannot come back for more. Nor can the defendant come back if the loss is less than was anticipated. Thus, the assessment of damages for the future is necessarily compounded of prophecy and calculation. The court [*1213] must do the best it can to reach what seems to be the right figure on a reasonable balance of the probabilities, avoiding undue optimism and undue pessimism. Although periodic payments and a right of recourse whenever circumstances change might seem an attractive solution of the difficulty, yet they, too, have serious drawbacks such as an unending possibility of litigation which, in the view of the law, have hitherto been held to outweigh the disadvantages of an assessment of damages once and for all. The present case is a classic example of the latter disadvantages if no remedy is available to the appellant.”


In Mulholland v. Mitchell [1971] A.C. 666, 674, Lord Hodson made a similar observation:


“By our law, unlike that of many other countries, the maximum interest reipublicae ut sit finis litium is, in the usual case, strictly followed. Damages are accordingly, assessed once for all at the time of the trial notwithstanding that in many cases, and this applies especially to cases of personal injury, uncertain matters have to be taken into account. The court has to make the best estimate it can as to the future life of the injured person, not only as to his prospects of recovery or improvement but also, as in this case, as to the cost of caring for him either in his own home or in an institution suitably equipped to deal with his condition. This is the function of the court.”


Both of these cases were dealing with the question of whether, upon a change of circumstances, fresh evidence should be admitted on appeal in support of an application to vary the measure of damages awarded. The cases were not dealing with the question of whether the court could, in appropriate circumstances, make an award in relation to one part of the claim while deferring adjudication on another. Mr. Eder submitted that the power to do this is conferred by the following provisions of R.S.C., Ord. 33:


“3. The court may order any question or issue arising in a cause or matter, whether of fact or law or partly of fact and partly of law, and whether raised by the pleadings or otherwise, to be tried before, at or after the trial of the cause or matter, and may give directions as to the manner in which the question or issue shall be stated.


“4 . . . (2) [In every action begun by writ] different questions or issues may be ordered to be tried at different places or by different modes of trial and one or more questions or issues may be ordered to be tried before the others.”


In my judgment these provisions enable the court to make an award of damages in relation to part of a claim while deferring for adjudication another part. In reaching this conclusion I have been influenced by the following consideration. R.S.C., Ord 14, r. 3 empowers the court to give summary judgment in respect of part of a claim, leaving the balance of the claim to go for trial. It would be illogical if the court were not able, after determining liability in a split trial, to give immediate judgment for heads of damage no longer in dispute while reserving for assessment heads of damage in respect of which an inquiry was necessary. For these reasons I have concluded that I have jurisdiction to make an award in respect of claims that have been paid while reserving for future determination that part of the Names’ claim that relates to anticipated claims. A similar, [*1214] although not identical, effect could be achieved by ordering an interim payment to reflect the paid claims.


The more difficult question is whether it is appropriate to defer the assessment of that part of the Names’ claim that relates to anticipated claims. In this context the authorities relied upon by Mr. Smith, to which I referred earlier, are germane. The desirability of bringing an end to litigation will normally make it appropriate for the court to make a single award of damages which includes the best assessment possible of future loss. This will not always be the case, however. In a personal injury case it may well be in the interests of justice that a final award of damages should not be made until sufficient time has elapsed for a reliable prognosis of the plaintiff’s medical condition. Thus in Hawkins v. New Mendip Engineering Ltd. [1966] 1 W.L.R. 1341, 1347 Winn L.J. suggested that the defendants should have


“asked the judge, in the interests of justice, to postpone the trial on the issue of damages, or to adjourn the case for consideration of that issue, until some five years after the accident, either by himself or another judge; . . .”


Since that decision the rules have been amended to permit the court to make a provisional award of damages for personal injuries, thus recognising in that context a situation where the interests of justice can call for the litigation to be kept alive rather than to be brought to an end.


Personal injury actions have special features which can justify the deferral of the assessment of damages. I have concluded that the current Lloyd’s litigation also has special features which can justify this exceptional course. Those features are (1) the nature of the loss; (2) the difficulties of assessing the loss; (3) the consequences of once and for all assessment of damages.


The nature of the loss


In this action, as in most of the other Lloyd’s actions, the principal complaint made by the Names is that their agents subjected them to excessive risk of liability to third party assureds. Damage is suffered as and when third party claims are made. Whether a particular risk written will result in a claim cannot be predicted with certainty, having regard to the capricious way in which the spiral operates. Where a plaintiff seeks relief in respect of potential third party liability which is uncertain, it seems to me, as it seemed to the Court of Appeal in Trans Trust S.P.R.L. v. Danubian Trading Co. Ltd. [1952] 2 Q.B. 297, that the appropriate course will usually be to defer dealing with that head of damage until the extent of the plaintiff’s liability, if any, has been determined.


The difficulties of assessing the loss


If the future claims pattern could be predicted with reasonable confidence, uncertainty would not constitute a valid objection to making an overall assessment of damages at this stage. I do not, however, consider that the future loss position is clear. The approach adopted by the Names’ experts to the assessment of future losses involves the application of graphical techniques to past claim experience. Mr. Jewell’s explanation includes the following passage:


“Graphical methods are used to monitor the factual position against the current estimate of the ultimate gross loss. If the factual experience [*1215] does not appear to support the estimate it will be reviewed. In addition, various statistical techniques are used to further analyse the tracked paid loss data in order to estimate the development of the loss to extinction.”


The defendants’ experts have applied two main methods to estimate future loss development. They are both based on fitting mathematical models to past development and using past and current trends to estimate the future. A brief description of their methods is given in their report:


“Curve fitting to development factors. This method involves fitting four separate families of curves to past development factors which are then extrapolated to estimate future development. The curve families are: exponential, inverse power, power and Weibull. Of these families, the inverse power curve is generally found to be pessimistic.


“Multiple regression modelling of the logarithms of incremental payments. This method allows the fitting of piecewise exponential curves to past claims development and extrapolation of future development. This method also includes a model of the variability of past payments which is used to estimate the likely variability of future payments and in turn the reliability of the estimated reserves. This method allows the use of several powerful statistical checks to assess the validity of the models.”


The Names’ estimates of ultimate losses are substantially higher than the defendants’. It would, of course, be possible for the court to attempt to resolve the differences between them, but having regard to the nature of the methodology that each side has adopted, I do not believe that the court would be able to make confident findings as to future loss development.


The consequence of once and for all assessment of damages


The uncertainty of future losses is not the most compelling consideration that has led me to defer the assessment of this part of the Names’ claim. I think it desirable that a uniform approach should be adopted in the Lloyd’s cases and this gives merit to adopting a similar approach to that of Gatehouse J. More importantly, the unprecedented circumstances of the Lloyd’s litigation make it desirable to consider the overall implications of awarding lump sum damages in respect of future underwriting liabilities before deciding whether this course is appropriate. The following questions arise: (1) is it just that agents or their E. & O. underwriters should be obliged to pay at this stage capital sums to cover losses that the Names have not yet suffered? and, (2) is it just or desirable that the Names should receive at this stage capital sums by way of damages to cover losses that they have not yet suffered?


The burden on defendants or their E. & O. underwriters of having to pay at this stage damages in respect of future losses will be substantial, but as nothing compared to the position of defendants in the “long tail” cases, should they be held liable. If Names receive substantial sums in anticipation of losses that they have not yet sustained, there must exist the risk that in some cases the damages will be dissipated before the claims are made. I am told that Lloyd’s have just introduced a byelaw designed to prevent this – although I have not been referred to this – but I am also told that the validity of this byelaw is to be challenged. [*1216] Mr. Eder touched on these considerations in his submissions, but they have not been fully explored. My overall conclusion is that the special features of the Lloyd’s litigation weigh strongly in favour of awarding damages in respect of underwriting losses when those losses are sustained and not in anticipation of them.


Mr. Smith has made the point that Names are required to meet cash calls that reflect to a degree estimates of future claims. This might justify an application for an interim payment – I do not prejudge that question – but it does not persuade me that it is appropriate to make at this stage a once and for all assessment of damages. In reaching my decision I have not had regard to a factor which Mr. Smith did not mention, but must have had in mind, namely the fact that E. & O. funds are limited. This cuts both ways. Names in syndicates whose actions have not yet come to trial might not unreasonably think it particularly harsh if E. & O. funds were exhausted in meeting claims for future, and less certain, liabilities, leaving them to bear the burden of the losses that they have already suffered.


Declared losses


Mr. Smith has defined as declared losses the losses that the managing agents have shown for each syndicate year in the annual syndicate returns. Under Lloyd’s Solvency and Reporting Byelaw (No. 13 of 1990, 5 December 1990) Names have to provide “eligible assets” sufficient to meet their “relevant liabilities.” Declared losses form part of the relevant liabilities. Mr. Smith submitted that, because the declared losses have this direct impact upon the amount of the eligible assets that a Name must provide in order to demonstrate solvency, the declared losses resulting from the five central catastrophies constitute damages recoverable by the Names. Mr. Eder submitted that this was a non sequitur. I agree. Declared losses fluctuate from one year’s return to the next. They represent the managing agents’ contemporary estimate of the losses that the syndicates are likely to suffer. They are not of themselves, conclusive of the amount of the Names’ losses and such impact as they have on the Names’ solvency position does not have the effect in law of making them recoverable losses.


Damages to be paid for liability for claims already made.


Damages pertaining to future claims deferred.


Applicants to have three-quarters of their costs.