(back to preceding text)

    63.  The point at issue in this appeal has never been considered by your Lordships' House and remains open for decision. Lord Bingham has placed strong emphasis in paragraph 23 of his opinion on the importance of certainty in commercial transactions. I do not wish to cast any doubt upon that, but I have come to the conclusion that Langley J and the Court of Appeal were right in holding that the contingency of the outbreak of war, which had occurred before the damages fell to be considered in the arbitration, could be taken into account. I find myself in agreement with Lord Mance when he said that considerations of certainty and finality have in this case to yield to the greater importance of achieving an accurate assessment of the damages based on the loss actually incurred.

    64.  The duration of the charter may in a case such as the present be affected by the contingency of the occurrence of an event which is in the contemplation of the parties and catered for in the terms of the charterparty. While the rate at which the hypothetical new charter is arranged on repudiation of the original one is for good reasons taken to be fixed at the time when the injured party could go into the market to negotiate a replacement, the same considerations do not apply to determination of the duration. The damages can be assessed at the date of repudiation by valuing the chance that the contingency would occur and that the charter would be cancelled, an approach accepted by Lord Mance at paragraph 23 of his judgment. That value might lie anywhere on the scale between extreme unlikelihood, which would give the deduction a minimal value, to virtual certainty, which would mean that it would be assessed at a figure very close to that which would be reached if one made the definite assumption that the contingency would occur. This approach is well known and recognised in other areas of the law. It is commonplace in the assessment of damages for personal injuries to award a sum which reflects the chance that a condition such as osteoarthritis may set in. A clear example of the technique may be found in Kitchen v Royal Air Force Association [1958] 1 WLR 563, where in a claim against solicitors for damages for professional negligence in failing to bring an action on for trial the court awarded the plaintiff a sum representing her prospects of success in the action, being a proportion of the damages which she would have received if one could have been certain that it would have been successful: see also the abundant examples discussed in McGregor on Damages, 17th ed (2003), paras 8-024 et seq.

    65.  This is where the principle exemplified by Bwllfa and Merthyr Dare Steam Collieries (1891) Ltd v Pontypridd Waterworks Co [1903] AC 426 operates. The respondent waterworks company in that case exercised its statutory power by notice dated 15 October 1898 to prevent the appellant colliery company from proceeding with its intention to work a seam of coal, notified by the appellant in September 1898 under section 22 of the Waterworks Clauses Act 1847. The coal in the protected area would have been reached in or about June 1900 and would have been worked out in about two years from that date. The appellant sought statutory compensation for this loss and the matter went to an arbitration, which concluded in February 1901, the arbitrator giving his award on 26 April 1901 in the form of a special case. The issue between the parties was whether the compensation was to be assessed, not on the basis of the value of coalfield or the coal in question in October 1898, but on the basis of the amount which the appellant could have made from mining the coal. The House rejected the respondent's contention that the coal should have been valued as if on a sale at the date of the appellant's notice in October 1898 and accepted that the proper basis was the profit which the appellant could have made from mining the coal. That being so, if the arbitration had been held soon after October 1898, the arbitrator could have calculated that by estimating the possible rise or fall in the price of coal over the period in which it would have been mined. As it was not held until a later date, he had available up-to-date evidence of the rise in price over the period. It was held that it was wrong to require him to disregard this, for, as Lord Macnaghten said in characteristic style at page 431:

"Why should he listen to conjecture on a matter which has become an accomplished fact? Why should he guess when he can calculate? With the light before him, why should he shut his eyes and grope in the dark?"

    66.  If the second Gulf War had not broken out by the time the arbitration was held, the arbitrator would have had to estimate the prospect that it might do so and factor into his calculation of the appellants' loss the chance that the charter would be cancelled at some future date under Clause 33. The loss which would have been sustained over the full period of the charter would then have been discounted to an extent which would have reflected the chance, estimated at the time of the assessment, that it would be so terminated. As events happened, however, the arbitrator did not come to assess damages until after the outbreak of war, when, as he found, the respondents would have cancelled the charter. The outbreak of the second Gulf War was then an accomplished fact, which was highly relevant to the amount of damages, and in my opinion the arbitrator was correct to take it into account in assessing the appellants' loss. As Lord Robertson put it in the Bwllfa case at page 432, "estimate and conjecture are superseded by facts." A striking example of the artificiality of failing to have regard to this principle may be seen in paragraph 84 of the opinion of my noble and learned friend Lord Brown of Eaton-under-Heywood.

    67.  Two features relating to this conclusion are worthy of mention. First, it is a necessary assumption in estimating the damages that the hypothetical new charterparty taken to have been negotiated by the appellants in early 2002 would have contained terms which corresponded as closely as possible to those of the original charterparty, including the cancellation provision in Clause 33. Mr Hamblen accepted in argument that one should regard it as containing such a clause and Lord Mance recorded in paragraph 27 of his judgment that the court was told that bilateral war clauses were invariable in time charters for tankers likely to visit the Gulf. Secondly, Mr Hamblen expressed concern that repudiating parties in future cases might attempt to delay the assessment of damages in order to see if such a suspensive condition might come into operation. I recognise that a risk of that nature may exist, but courts and arbitrators have the ability to prevent such abuse if application is made to them to proceed with dispatch.

    68.  For these reasons and also for those contained in the opinions of my noble and learned friends Lord Scott of Foscote and Lord Brown, I agree with the conclusion reached by Langley J and the Court of Appeal and would dismiss the appeal.

LORD BROWN OF EATON-UNDER-HEYWOOD

My Lords,

    69.  The basic facts of this case could hardly be simpler. On 17 December 2001 the appellant owners accepted the respondent charterers' repudiation of a charterparty which nominally still had nearly four years to run—to 6 December 2005. I say "nominally" because by clause 33 of the charterparty the charterers were entitled to cancel it in the event of war or hostilities breaking out between any two or more of a number of countries including the USA, the UK and Iraq. On 20 March 2003, such a war (the Second Gulf War, hereafter 'the War') did indeed break out and the arbitrator has found as a fact that the charterers would in any event then have cancelled the charterparty.

    70.  Indisputably the owners are entitled to damages for having been deprived of the value of this charterparty for the fifteen months or so up to the outbreak of the War. Are they, however, entitled, as they claim, to be compensated on the basis that the charterparty would have continued for the whole length of its nominal term?

    71.  The owners advance their argument by reference to the familiar principle that damages for breach of contract ordinarily fall to be assessed as at the date of the breach (the breach date rule as it was called in argument). They submit that that principle is applicable here and that the assessment of damages must accordingly ignore the outbreak of the War. That, it is argued, is a subsequent event of no relevance to the proper assessment of the owner's loss in December 2001. Indeed, even had there existed in December 2001 a very substantial risk of imminent war in the Gulf, the owner's principal argument would require it to be ignored: only if it could be shown that by that date war was inevitable—"pre-destined to happen" in the words of Megaw LJ in The Mihalis Angelos [1971] 1 QB 164 (in the passage cited by my noble and learned friend Lord Bingham of Cornhill at para 14)—could it be brought into account to ensure that "the damages which [the owners] can recover are not more than the true value, if any, of the rights which [they have] lost, having regard to those pre-destined events" (again the words of Megaw LJ in the same passage). The charterers submit to the contrary that, whilst certainly (given the availability of a market for the vessel's period chartering) the breach date rule would operate in this case to fix the daily net differential base charter rate, it should not determine either the period for which that loss was suffered or the date for assessing that period.

    72.  A single issue has been formulated by the parties for your Lordships' determination on the appeal:

"Where damages for an accepted repudiation of a contract are claimed, in what circumstances can the party in breach rely on subsequent events to show that the contractual rights which have been lost would have been rendered either less valuable or valueless?"

    73.  It is convenient at this point to notice the way this issue was dealt with by the arbitrator in his Second Declaratory Arbitration Award. His main conclusion—to which he felt reluctantly driven by Timothy Walker J's decision in The Seaflower [2000] 2 Lloyd's Rep 37—was that the outbreak of war "did place a temporal limit on the recoverability of damages … and that no damages are recoverable for the period from 21 March 2003 onwards." But his award considered also what the position would have been if (contrary to his main conclusion) damages had had to be assessed as at December 2001. The owners (relying on The Mihalis Angelos) argued that in that event the charterers would have had to show that as at that date war was inevitable; the charterers submitted to the contrary that all they would have needed to establish was that war was probable. Expert evidence on this issue was adduced respectively from a professor of peace studies and a professor of war studies, one saying that war had been inevitable, the other (as recorded in para 35 of the award) that "its level of probability was not above 50%, as of mid-December 2001." The arbitrator expressed his conclusion at para 59:

"On the evidence, I have concluded that at 17 December 2001, a reasonably well-informed person would have considered war (or large-scale hostilities) between the United States/United Kingdom and Iraq merely a possibility. I do not consider that such a person would have considered it inevitable or even probable but merely a possibility, although I do accept that the degree of probability would have been higher had that person known as much about the prevailing circumstances then as we do today. In view of this finding it is unnecessary for me to consider whether the relevant degree of likelihood is that war was inevitable or that it was likely to occur on the balance of probabilities, although it appears that I am bound by the decision in The Seaflower and that the relevant degree of likelihood required is 'inevitable'."

    74.  Having regard to his main conclusion, of course, this secondary issue was not in any event capable of influencing the result: that was to be determined by the known outcome of events, not some notional earlier forecast. It seems to me, however, important to consider its implications. There was in this charterparty (as my noble and learned friend Lord Carswell points out at para 59 of his opinion) a suspensive condition, clause 33, recognised by the parties to be capable of bringing this contract to a premature end. Could it really be the position that only were cancellation under clause 33 inevitable could it properly be brought into account in calculating the value of this repudiated contract? What, indeed, in this context does inevitability connote? At what point in the remaining four years of the nominal term of this charterparty did its premature cancellation have to be inevitable? It would be one thing if war were adjudged likely to have broken out within a month, quite another if that was not to be for a further three years—and so forth. For my part, I have found The Mihalis Angelos of little assistance on this appeal. The position that arose there could hardly have been more different. It concerned a single voyage charterparty. Once it was recognised that the charterers would have been bound in any event to have cancelled the charterparty before ever it took effect, the owners could establish no loss. It was an all or nothing case. The voyage either would or would not have taken place. Not so here: the question rather was whether and if so when the charterparty would have been ended under clause 33 before the completion of its nominal term. Even had war not broken out by the time damages came to be assessed I can see no reason why that question should not have been addressed in the conventional way, i.e. by making the best possible assessment of the likely course of future events as at the date of assessment. As Lord Denning MR said in The Mihalis Angelos, at p 196: "You must take into account all contingencies which might have reduced or extinguished the loss." It was hardly a novel proposition.

    75.  I realise, of course, that I have yet to address the central question arising for your Lordships' decision: the date by reference to which the overall damages assessment falls to be made. It seems to me, however, essential first to appreciate just what is the exercise the tribunal assessing damages would be embarking upon.

    76.  Not only do I reject the owners' contention that the tribunal would be concerned with contingencies only if, at the relevant date, they were inevitable, but I reject too the charterers' concession, apparently implicit in their competing submission to the arbitrator, that only contingencies which are probable need be brought into account. It seems to me that before the arbitrator everyone was mesmerised by The Mihalis Angelos, a case which, as I have already suggested, affords very little assistance in the very different circumstances arising here. And, indeed, despite their apparent approval of Megaw LJ's formulation of the test in The Mihalis Angelos, I understand both Lord Bingham and my noble and learned friend Lord Walker of Gestingthorpe to accept that account should properly be taken of a contingency which would reduce the value of the contract lost even were the chance of it happening less than 50% (provided always that it was of some real and not just minimal significance)—see particularly para 22 of Lord Bingham's opinion and para 46 of Lord Walker's.

    77.  In this connection I respectfully disagree with Lord Walker's interpretation (in para 45 of his opinion) of the arbitrator's findings. Unlike him, I understand the arbitrator's characterisation (in para 59 of his award) of the prospect of war as at 17 December 2001 as "merely a possibility" to mean precisely "less than a 50% prospect." On the arguments addressed to him, the only issue was whether the correct test was one of probability or inevitability. Once he decided on the facts that neither of those tests was satisfied here, it was unnecessary for him to say more. The expression "merely a possibility" was used simply in contradistinction to both the rival situations being contended for, probability and inevitability. Para 59 was, indeed, an acceptance of Professor Freedman's view, expressed at para 35, "that war was clearly neither pre-destined nor inevitable and, indeed, its level of probability was not above 50%, as of mid-December 2001." Not even Professor Freedman was suggesting that the prospect of war was only an outside possibility, of no real significance whatever. His evidence was no more extreme than that the prospect "was not above 50%." Assuming this were so, whether the prospect was 5% or 45% was, on the arguments addressed to the arbitrator, immaterial.

    78.  In the end, I should add, nothing to my mind turns on this difference between Lord Walker and myself. My more fundamental conclusion, as I shall shortly explain, is that the breach date rule does not require contingencies—such as the likely effect of a suspensive condition—to be judged prior to the date when damages finally come to be assessed. All I am presently concerned to demonstrate is that, whatever the appropriate date by reference to which the assessment must be made, the taking into account of a contingency is likely to be an altogether more complex exercise than it was in The Mihalis Angelos and that it will not generally be possible, as it was there, simply to decide one way or the other (in that case it mattered not whether at the point of breach or later) whether or not the whole contract was inevitably doomed to fail. The point is sufficiently made if the arbitrator here might have found, say, a 30% or 40% prospect of war as at December 2001. The next question would have been: war when? Could it really be thought in the interests of certainty to address these questions as of some date prior to that on which damages are in fact being assessed? To my mind, not. Must the judge really shut his eyes to the known facts and speculate how matters might have looked at some earlier date? Again, not without compelling reason and none appears to me. Lord Bingham (at para 12) and Lord Carswell (at para 65) have already explained the "Bwllfa principle": Bwllfa & Merthyr Dare Steam Collieries (1891) Ltd v Pontypridd Waterworks Co [1903] AC 426. There is no need to repeat it. Suffice it to say that I see no good reason to depart from it here.

    79.  It is time finally to address the breach date rule directly. Its most obvious manifestation is to be found in section 51(3) of the Sale of Goods Act 1893:

"Where there is an available market for the goods in question the measure of damages is prima facie to be ascertained by the difference between the contract price and the market or current price of the goods at the time or times when they ought to have been delivered, or, if no time was fixed, then at the time of the refusal to deliver."

    But the rule is by no means confined to the sale of goods context and, as Lord Bingham explains, has been applied by analogy to a variety of other situations. Essentially it applies whenever there is an available market for whatever has been lost and its explanation is that the injured party should ordinarily go out into that market to make a substitute contract to mitigate (and generally thereby crystallise) his loss. Market prices move, both up and down. If the injured party delays unjustifiably in re-entering the market, he does so at his own risk: future speculation is to his account—"the buyer's decision is (in the vernacular) down to him" (Bingham LJ, as he then was, in Kaines (UK) Ltd v Osterreichische [1993] 2 Lloyd's Reports 1, 11).

    80.  The rule is easy to apply where, for example, goods or shares are traded: if it is the seller who is injured by non-acceptance, he must as soon as possible re-sell the goods or shares at the then available market price; if the buyer, he must similarly buy in substitute goods or shares. But undoubtedly the rule can be applied in more complex situations, for example, to building or repairing contracts and, most relevantly for present purposes, to breached charterparties—see particularly The Mihalis Angelos, The Elena d'Amico [1980] 1 Lloyd's Rep 75, The Wave [1981] 1 Lloyd's Rep 521, The Noel Bay [1989] 1 Lloyd's Rep 361 and The Seaflower [2000] 2 Lloyd's Rep 37. Where goods or shares are sold, the breach date rule is at its strictest. In other cases, however, time may well be needed before the injured party can reasonably be required to re-enter the market. Radford v De Froberville [1977] 1 WLR 1262 concerned a defendant's contractual failure to build a wall on the plaintiff's land. In a much quoted judgment Oliver J at p 1285 said this:

"It is sometimes said that the ordinary rule is that damages for breach of contract fall to be assessed at the date of the breach. That, however, is not a universal principle and the rationale behind it appears to me to lie in the inquiry—at what date could the plaintiff reasonably have been expected to mitigate the damages by seeking an alternative to performance of the contractual obligation?"

    Or take Dodd Properties (Kent) Ltd v Canterbury City Council [1981] 1 WLR 433, a case of contractual liability for serious structural damage, where Megaw LJ (at p 451) said this:

"The true rule is that, where there is a material difference between the cost of repair at the date of the wrongful act and the cost of repair when the repairs can, having regard to all relevant circumstances, first reasonably be undertaken, it is the latter time by reference to which the cost of repair is to be taken in assessing damages. That rule conforms with the broad and fundamental principle as to damages, as stated in Lord Blackburn's speech in Livingston v Rawyards Coal Co (1880) 5 App Cas.25, 39, where he said that the measure of damages is:

' … 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.'

In any case of doubt, it is desirable that the judge, having decided provisionally as to the amount of damages, should, before finally deciding, consider whether the amount conforms with the requirement of Lord Blackburn's fundamental principle."

    A little later, Megaw LJ (at p 453) added:

"I agree with the observations of Oliver J in Radford v De Froberville … as to the relationship between the duty to mitigate and the measure, or amount, of damages in relation to a question such as the question with which we are here concerned."

    81.  Take, indeed, this very case. Whilst it was not disputed before the arbitrator that an available market existed for the chartering of this vessel, the owners contend that a new fixture could only have commenced earning on 1 April 2002 and in the result claim for loss on the spot (rather than the period charter) market for the three and a half months between 17 December 2001 and 1 April 2002. In this case, as in Dodd Properties, therefore, account must necessarily be taken of post-breach events. Why then ignore the outbreak of the War? Say that this had started a year earlier, on 20 March 2002, before a substitute charterparty could have commenced earning? Would the damages assessment really have had to ignore it? Nor, indeed, is the lapse of some three and a half months before a substitute charterparty could have commenced earning the only, or even the main, post-breach event necessarily to be taken into account here. As my Lords have observed, the rate of hire of the Golden Victory consisted only partly of the base charter rate; it consisted also of a share in the charterers' operating profit over and above that rate and this latter element of the claimed loss, as Lord Mance pointed out, would require any court or arbitrator to take account of actual market rates during the relevant period over which the loss is claimed. Yet on the owners' argument that part of the claim too would continue throughout the remaining four years of this charterparty, the intervention of the War notwithstanding.

    82.  None of the earlier charterparty cases to my mind addresses or determines the particular problem arising here. In none of them, as Lord Carswell has observed, was there a suspensive condition which could (and which we know in the event would) have operated to bring the contract to an early end. Any substitute contract here—albeit, as the arbitrator found, for the 4 year balance of the original 7 year term— would have been subject to the self-same conditions as the repudiated contract. And it can be assumed that the hypothetical substitute charterers would similarly have cancelled their contract on the outbreak of the War. The arbitrator surmised that had the owners in December 2001 sold the vessel with a substitute 4 year charter "the value they would have received would surely have been calculated on that basis" i.e. the basis of a 4 year charterparty with war "no more than a possibility." Even were that so, however, no one has ever suggested that the breach date rule operated here so as to require the owners to go out into the market not only to obtain a substitute charter but also then to sell the vessel. The measure of loss did not fall to be crystallised on this basis.

    83.  In my opinion the owners' argument here seeks to extend the effect of the available market rule well beyond its proper scope and to do so, moreover, at the plain expense of Lord Blackburn's fundamental principle: to restore the injured party to the same position he would have been in but for the breach, not substantially to improve upon it. It is one thing to say that the injured party, mitigating his loss as the breach date rule requires him to do, thereby takes any future market movement out of the equation and to that extent crystallises the measure of his loss; it is quite another to say, as the owners do here, that it requires the arbitrator or court when finally determining the damages to ignore subsequent events (save where the defendants can demonstrate that at the date of breach some suspensive condition would inevitably—and immediately—have operated to cancel the contract). There is no warrant for giving the rule so extended an application.

    84.  There is a final point to be made. Shift the facts here and assume that the arbitrator had found, as at December 2001, a probability (or even merely a significant possibility) of (perhaps imminent) war breaking out in the Gulf, but that in fact, by the time damages finally came to be assessed, not only had war not broken out but all risk of it had disappeared—or, indeed, the assessment might not have taken place until the whole nominal term of the charterparty had expired. On the view taken by the minority of your Lordships, the damages award would have had to reflect a risk which never in fact eventuated, a conclusion in the circumstances, greatly to the owner's disadvantage. Yet that inescapably is the logic of the minority's approach. Is such a result compelled in the interests of "certainty" or "finality" or "consistency" or "predictability" (the interests identified by Lord Bingham at para 23 of his opinion), or so that the charterers could have "promptly honoured their secondary obligation to pay damages" (Lord Bingham at para 22)? In my judgment it is not. When market movement can be eliminated from the assessment of damages (as here with regard to the charterparty rate) it should be (by the breach date rule). But not history; the Court need not shut its mind to that.