Has Promoting Commercial Certainty Made the Law More Uncertain?

August 11, 2008



I.         Introduction


In Transfield Shipping Inc v Mercator Shipping Inc [2008] UKHL 48, the House of Lords considered the issue of remoteness of damages in contract. Although the case arose in the specialised context of shipping law, the principles apply equally to all contracts and should be of interest to all commercial litigators. In the context of IT disputes, the issue will arise in relation to the recoverability of losses where, for instance, software or a system is not delivered in accordance with the terms of the parties’ contract – be it late or not in accordance with the specification.


The issue addressed by the House was defined by Lord Hoffmann (at [9]):


The case … raises a fundamental point of principle in the law of contractual damages: is the rule that a party may recover losses which were foreseeable (“not unlikely”) an external rule of law, imposed upon the parties to every contract in default of express provision to the contrary, or is it a prima facie assumption about what the parties may be taken to have intended, no doubt applicable in the great majority of cases but capable of rebuttal in cases in which the context, surrounding circumstances or general understanding in the relevant market shows that a party would not reasonably have been regarded as assuming responsibility for such losses?


Not all of the House chose to address the issue in this way or answer this question head on but a majority did; in doing so, they have, at the very least, added an important gloss on the existing law.


This article explains this development of the law and explores its implications. As our title suggests, not all of the implications are necessarily positive or easy to predict.


II.        Background


The Appeal related to a vessel called The Achilleas which had been chartered by Mercator to Transfield for a five/seven month charter with an option to extend for a further similar period, which was in the event exercised. The charter rate under the extended charter was US$16,450 per day. On the facts, the latest day for redelivery of the vessel was midnight on 2 May 2004. Transfield gave notices of redelivery, including a ten-day definite notice on 20 April, ie that the vessel would be redelivered within ten days of the date of notice. In anticipation of such redelivery, on 21 April 2004, Mercator fixed the vessel for a follow on charter of four/six months with Cargill at the rate of US$39,500 per day with a delivery period of 28 April to 8 May (‘the Cargill Fixture’).  If the vessel was not delivered within that period, Cargill was not obliged to take delivery of the vessel but could cancel the charter.


In the event, the vessel did not load its final cargo until 24 April and was not redelivered by Transfield to Mercator until 11 May. In the intervening period there was a substantial fall in the dry charter market, in other words the hire rates that Mercator could obtain had dropped substantially. On 5 May, when it became apparent to Mercator that the vessel was not going to make the cancelling date under the Cargill Fixture, Mercator approached Cargill to obtain an extension of the cancelling date.  Cargill were prepared to agree to this but only in return for a US$8,000 per day reduction in the hire rate, a reduction which reflected the market at that time. 


The vessel completed her last voyage and was redelivered by Transfield to Mercator, and, in turn, delivered by Mercator to Cargill on 11 May.


Mercator claimed for their loss of profit for the entire period of the Cargill Fixture for the breach by Transfield in failing to redeliver the vessel by 2 May, ie the reduction in the hire rate which were they were forced to agree to in order to secure the charter for the duration of the Cargill Fixture. Transfield argued that, although they were in breach of charter, they should only be liable for the difference between the market rate and charter rate during the nine day overrun from the date that the vessel should have been redelivered to the date when it was in fact redelivered. The parties agreed quantum so that if Mercator’s case was accepted the damages payable would be about US$1,365,000 net, whereas on Mercator’s case an amount of about US$158,000 would be payable.  The approach to remoteness of damages clearly had significance consequences to the level of quantum that Mercator could recover.


III.      The Decisions of the Arbitral Tribunal and the Lower Court


The Arbitrators


The Majority of the Tribunal (one arbitrator dissented) made the following findings:


(1)      That ‘in today’s market’ in the event of late redelivery of a vessel, the loss of the next fixture by Mercator due to the cancellation date of the next fixture being missed was ‘the kind of result which the parties would have had in mind’ when they concluded the extension of the charterparty.


(2)      That the fact that the market rate for tonnage may go up and down was well known in the market and would have been known by Mercator.


(3)      The type of loss claimed by Mercator was readily identifiable to those with ‘even a minimum of experience of the shipping industry’.


(4)      Because the type of loss claimed was foreseeable, in the sense indicated above, it did not matter that the precise extent of the loss was not foreseeable.


(5)      The claim fell within what the Majority described as the first rule of Hadley v Baxendale (1854) 9 Ex 341 and Mercator’s claim was allowed.


Commercial Court Judgment ( [2006] EWHC 3030 (Comm))


Mr Justice Christopher Clarke held as follows:


(1)      The judgment in Hadley v Baxendale does not lay down two separate and distinct rules but a single rule with two (often overlapping) branches (see [49]-[53]).


(2)      In the light of the Majority Arbitrators’ findings, the loss claimed was not too remote and could be regarded as naturally arising from the breach especially given the findings as to what the parties had contemplated at the relevant time (see [55]-[56]).


(3)      There was no additional requirement to show that Transfield had assumed responsibility for the loss in a case, as here, where the type of loss is within the reasonable contemplation of the parties at the relevant time as a not unlikely consequence of the breach (see [59]-[65]).


(4)      The fact that in some cases there might be difficulties in assessing loss of a follow-on fixture is not an adequate reason to confine the recovery to a sum markedly less than the actual loss. In any case, no such difficulties arose in the present case ([99]).


IV.       Court of Appeal Judgment


The case was heard in the Court of Appeal by Ward, Tuckey and Rix LJJ ([2007] EWCA Civ 901). Lord Justice Rix gave the lead judgment and his main points were as follows.


(1)      There is no fixed rule, and no binding authority, that damages for late redelivery of a time chartered vessel are limited to the overrun period measure ([118]-[119]).


(2)      That Hadley v Baxendale does not contain two rules but two means by which a defendant may possess the knowledge necessary to make his liability a fair one. That knowledge may either arise from the ‘usual course of things’, or from the communication of special circumstances ([88]).


(3)      The refixing of the vessel at the end of Mercator’s charter was not merely ‘not unlikely’, it was in truth highly probable (barring other possibilities). The nature of the chartering market was at all times an open book to Mercator: it was their own business and it was a market which went up and down and could be volatile ([96]).


(4)      If parties within the market required greater certainty, this could be achieved by the use of a contractual term to achieve this result ([112] and [120]).


(5)      To restrict damages to the difference between contract and market rates for the period of overrun would be both undesirable because it put owners too much at the mercy of their charterers and uncommercial because charterers already know that a new fixture, in all probability fixed at or around the time of redelivery, will follow on from their own charter ([119]).


V.        The Judgments of the House of Lords


The Appellate Committee of the House of Lords which heard the appeal was made up of Lords Hoffmann, Hope, Rodger and Walker and Baroness Hale. All five members of the Appellate Committee gave speeches allowing the appeal (in the case of Baroness Hale, with real reluctance) and finding in favour of Mercator. 


The reasoning of their Lords was not the same. It is important to consider the speeches in turn. However before doing so it is worth stressing that the bases for their decision were not ones which had been argued at any stage by the Mercator. Nor were they ones which had been debated in argument at the hearing. The hearing was concerned with questions of whether there was a market-based rule which required the appeal to be allowed – this argument is hardly addressed and had failed below.


The result is that some of the implications of the reasoning are perhaps not as carefully worked through as they might have been and there is little or any attempt by their Lordships to address how their reasoning fits with previous decisions.


Remoteness test according to prior orthodox thinking


To put the speeches in context it is useful first to set out the law as it was understood to be before this decision. The classic statement of the test for remoteness is set out in Hadley v Baxendale (1854) 9 Ex. 341 at p 354:


‘Where two parties have made a contract which one of them has broken, the damages which the other party ought to receive in respect of such breach of contract should be such as may fairly and reasonably be considered as either arising naturally, ie according to the usual course of things, from such breach of contract itself, or such as may reasonably be supposed to have been in the contemplation of both parties, at the time they made the contract, as the probable result of the breach of it.’


Lord Reid’s formulation of the test of remoteness in The Heron II [1969] 1 AC 350 received subsequent approval. His approach was to ask whether the loss in question was:


‘of a kind which the defendant, when he made the contract, ought to have realised was not unlikely to result from the breach … the words “not unlikely” …  denoting a degree of probability considerably less than an even chance but nevertheless not very unusual and easily foreseeable.’ (at pp 382 – 383).


An example of the application of that test is The ‘Rio Claro [1987] 2 Lloyd’s Rep 173 where the judge stated (at p 175) that in order for a loss arising from a breach of contract to be recoverable:


‘It must be such as the contract-breaker should reasonably have contemplated as not unlikely to result.  To that direction must be added the point that the precise nature of the loss does not have to be in his contemplation.  It is sufficient that he should have contemplated loss of the same type or kind as that which in fact occurred.  There is no need to contemplate the precise concatenation of circumstances which brought it about.’


There was no additional requirement of the test in Hadley v Baxendale  that the contract breaker has ‘assumed responsibility for the loss’, save, as the judge and Rix LJ rightly found, where the loss claimed is unusual and recovery of such loss would require special notice of unusual facts or circumstances.  It is only where knowledge of special circumstances is required (cases directed at the so called ‘second rule’ in Hadley v Baxendale) that there ever was such a requirement (see Chitty on Contracts (29th Ed) paragraphs 26–054 and 26–055 and the commentary in McGregor on Damages, 17th Ed (2003) at paras 6–174 to 177).



Lord Hoffmann


Lord Hoffmann rejected the submission that the ‘starting point’ was that damages were designed to put the innocent party, so far as it is possible, in the position as if the contract had been performed (Robinson v Harman (1848) 1 Exch 850, 855) and instead stated that one must first decide whether the loss for which compensation is sought is of a ‘kind’ or ‘type’ for which the contract-breaker ought fairly to be taken to have accepted responsibility. He cited Banque Bruxelles Lambert SA v Eagle Star Insurance Co Ltd [1997] AC 191 at p 211 and contended that the same principle applied to an express contractual duty (to redeliver the ship on the appointed day). He also referred with approval to the approach of the Court of Appeal in Mulvenna v Royal Bank of Scotland plc [2003] EWCA Civ 1112, where a claim for damages for the loss of profits, which the claimant said he would have made if the bank had complied with its agreement to provide him with funds for a property development, was struck out.


He concluded that everything rested upon whether the loss claimed was of a type which the contract-breaker had agreed to assume responsibility. He stated:


‘What is the basis for deciding whether loss is of the same type or a different type? It is not a question of Platonist metaphysics. The distinction must rest upon some principle of the law of contract. In my opinion, the only rational basis for the distinction is that it reflects what would have been reasonable and have been regarded by the contracting party as significant for the purposes of the risk he was undertaking.’


On that basis he concluded that the parties would have considered losses arising from the loss of the following fixture a type or kind of loss for which the charterer was not assuming responsibility. He rejected the submission that the arbitrators’ findings were one of fact because: ‘..the question of whether a given type of loss is one for which a party assumed contractual responsibility involves the interpretation of the contract as a whole against its commercial background, and this, like all questions of interpretation, is a question of law.’


Lord Hope


Lord Hope acknowledged that he was, at first, inclined to find in favour of Mercator, but changed his mind after considering the draft speeches of Lords Hoffmann, Rodger and Walker. Lord Hope considered that the ‘assumption of responsibility’ formed the basis of the law of remoteness of damage in contract and that the key question should be ‘whether the loss was a type of loss for which the party can reasonably be assumed to have assumed responsibility’. 


He recognised that it was within the parties’ contemplation that loss would be suffered generally by reason of late redelivery, and that this would be loss of use at the market rate as compared with the charter rate. He also considered that Transfield could not be expected to know ‘how’ Mercator would deal with the charterers under any subsequent fixture. This, he considered, was something over which Transfield had no control at the time of entering into the contract and was completely unpredictable. As a result, he considered that there could be no presumption that the party in breach had assumed responsibility for any loss.


Lord Walker


Lord Walker also appeared to give support for the concept of ‘assumption of responsibility’. He placed real emphasis on the need for commercial certainty and expressed the relevant question in the following way:


It is also a question of what the contracting parties must be taken to have in mind having regard to the nature and object of their business transaction’.


He rejected the application of Lord Reid’s ‘not unlikely’ test.  Instead, in his view, what mattered was:


whether the common intention of reasonable parties to a charterparty of this sort would have been that in the event of a relatively short delay in redelivery an extraordinary loss, measured over the whole term of a renewed fixture, was, in Lord Reid’s words, … ‘sufficiently likely to result from the breach of contract to make it proper to hold that the loss flowed naturally from the breach or that loss of that kind should have been within ….contemplation’’. 


Lord Rodger


Lord Rodger adopted a different approach. He did not adopt the assumption of responsibility test articulated by Lord Hoffmann and expressly approved by Lords Walker and Hope. Instead he sought to say that the loss was simply outside the ordinary contemplation of these parties. In so doing he appears to have made a fresh finding of fact and rejected the arbitrators’ findings.


He noted that it requires extremely volatile market conditions to create the situation which occurred and concluded that loss arising out of such volatility was not a kind of loss that could be said to be the ‘not unlikely’ result of the breach.  He also placed reliance on the fact that Mercator’s dealings with Cargill were not known by Mercator. 


Of more general interest, when discussing the speeches in Victoria Laundry and the Heron II, he stated (at [52]) as follows:


‘ .. In any event, amidst a cascade of different expressions, it is important not to lose sight of the basic point that, in the absence of special knowledge, a party entering into a contract can only be supposed to contemplate the losses which are likely to result from the breach in question – in other words, those losses which will generally happen in the ordinary course of things if the breach occurs. Those are the losses for which the party in breach is held responsible – the stated rationale being that, other losses not having been in contemplation, the parties had no opportunity to provide for them’.


Baroness Hale


Baroness Hale was at first inclined to agree with the very full and thoughtful judgments in the courts below and she proceeded to set out her reasons why. After articulating the approach of Lord Rodger and then that of the other Law Lords she states


‘Therefore, if this appeal is to be allowed, as to which I continue to have doubts, I would prefer it to be allowed on the narrower ground identified by Lord Rodger, leaving the wider ground to be fully explored in another case and another context.’


She is critical of the approach of Lord Hoffmann in particular, in the following terms:


‘ it seems to me that it adds an interesting but novel dimension to the way in which the question of remoteness of damage in contract is to be answered, a dimension which does not clearly emerge from the classic authorities. There is scarcely a hint of it in The Heron II, apart perhaps from Lord Reid’s reference, at p.385, to the loss being “sufficiently likely to result from the breach of contract to make it proper to hold that the loss flowed naturally from the breach or that loss of that kind should have been within his contemplation” (emphasis supplied). In general, The Heron II points the other way, as it emphasises that there are no special rules applying to charterparties and that the law of remoteness in contract is not the same as the law of remoteness in tort.


She adds:


The rule in Hadley v Baxendale asks what the parties must be taken to have had in their contemplation, rather than what they actually had in their contemplation, but the criterion by which this is judged is a factual one. Questions of assumption of risk depend upon a wider range of factors and value judgments. This type of reasoning is, as Lord Steyn put it in Aneco Reinsurance Underwriting Ltd v Johnson & Higgins Ltd [2002] 1 Lloyd’s Rep 157, para 186, a ‘deus ex machina’. Although its result in this case may be to bring about certainty and clarity in this particular market, such an imposed limit on liability could easily be at the expense of justice in some future case. It could also introduce much room for argument in other contractual contexts.


VI.       Implications


The relevant test


Only time will tell whether this decision has changed the test for remoteness in contract generally.


The application of the assumption of responsibility test by a majority of the House is, as Baroness Hale points out, both novel and very different to how the test in contract had previously been understood. The majority appear to regard this not as an aspect of the test but ‘the starting point’.


However, it is not at all clear how parties (or courts) will determine what responsibilities parties agreed to assume other than by applying the existing foreseeability question. For that reason, it may be that the new test will not lead to a different outcome in many cases, but on the other hand it may well do so in many cases. It certainly did in the appeal in question.


Indeed questions of assumption of responsibility seem alien to many commercial contracts.


The position is made less clear by the fact that many decisions which appear to contradict the assumption of responsibility approach are simply not addressed or explained by the House. This might in part be a function of the fact that the assumption of responsibility point was never argued before the House, as Baroness Hale notes.


So, it is reasonable to ask how a case such as Jackson v Royal Bank of Scotland [2005] 1 WLR 377, the last case in which remoteness was considered by the House of Lords, and in particular the judgments of Lords Hope and Walker, fit with the Transfield case.


In Jackson the claimant imported dog chews from Thailand and sold them on to a customer in the UK.  The customer knew the identity of the supplier in Thailand, but knew nothing of the commercial arrangements which existed between that supplier and the claimant and nothing of the mark-up on each order. In March 1993 the Bank erroneously sent a completion statement and other documents, including an invoice from the Thai supplier, to the customer who saw the amount of mark-up and decided to cut the claimant out of the equation as a middle man.


The claimant sued the bank for breach of confidence and succeeded and was awarded damages for the loss of opportunity to earn future profits from its trading relationship. The judge held that there was a significant chance that the claimant’s relationship with the customer would have continued for a further four years. Thereafter, the claimant’s chance of obtaining repeat business was considered to be too speculative for the award of damages. The Court of Appeal substantially reduced the level of damages on the basis of remoteness.


However, the House of Lords held that it was wrong to limit the period for which damages were recoverable by reference to what was within the reasonable contemplation of the Bank at the time of the breach.  The award which the judge had made on a reducing basis extending over a four-year period was as good an estimate as could now be made. The appeal was therefore allowed and the original judge’s order restored.


It is not easy to see why the Bank could be said to have ‘assumed a responsibility’ for trading loses continuing over a four-year period. In fact this case is only explicable on the basis that the ‘not unlikely’ test is the proper test.


So it is reasonable to ask – what reflects the law: Jackson or Transfield  or both? If both, how do the two decisions sit together? This is especially pertinent, given that Lord Hoffmann identifies the banking market as a particular sector for which the assumption of responsibility test is appropriate.


Furthermore – and perhaps most fundamentally – how does Transfield sit with the Heron II? As Baroness Hale points out, there is hardly a hint of assumption of responsibility in the House of Lords decision widely thought to define the modern test of remoteness. Is it no longer the law? If not, why did the House not say so or at least provide guidance as to when the test set out in that case does apply and when it does not?


Furthermore how does Lord Rodger’s statement that losses are only recoverable if they were likely to result from the breach (at [52]) fit with the express statement by Lord Reid in the Heron II [1967] 3 WLR 1491 at p 1505, which suggests the test is that is that the loss was ‘not unlikely to occur’, which words provide a rather different emphasis and approach.


The upshot is that the legal test for remoteness in relation to contract damages appears a good deal less clear after this appeal than it did before.


Market movements and forseeability


A further area of confusion arises from the minority ratio. This appears to be founded on the proposition that market movements which are volatile are unforeseeable or sufficiently unlikely to fall outside the parties’ contemplation.


There is an immediate objection to this approach as it involves a finding of fact as to what was and what was not in the parties’ contemplation and the House of Lords has no jurisdiction on an appeal from an arbitration to make such a finding. Further, judges have, for many years, taken the view that even volatile movements in the market are not too remote to form the basis of an award of damages because parties are taken to foresee that markets can move in that way: see The Rio Claro [1987] 2 Lloyd’s Rep 173 and North Sea v PTT [1997] 2 Lloyd’s Rep. 418 where Thomas J said at p 438, that, where a claim was based upon a market:


it would generally have been no answer to that claim that the market had moved in an unprecedented way…


It follows that this ratio disturbs a settled understanding of the law that market movements per se, whatever they may be, are foreseeable but it raises the difficult question of when a movement is sufficiently volatile to take its consequences outside the contemplation of the parties.


IT context


There will be a myriad of circumstances when the failure to deliver hardware and/or software on time or on specification will cause significant losses, including losses relating to the operation of the customers’ business. This decision has made the position in relation to the recovery of such losses more uncertain.


It is recognised that many contracts will seek to exclude the recovery of consequential and other losses. However, it is now well settled that the exclusion of consequential losses does not exclude losses which are said to arise in the usual course of things, that is those losses within the so-called first rule of Hadley v Baxendale (see Croudace Construction Ltd v Cawoods Concrete Products Ltd [1978] 2 Lloyd’s Rep 55). Accordingly Transfield may well be important not only in those cases where there is no effective exclusion of such losses but also to define the extent of often-used exclusion clauses.


Applying the approach of the Lords in Jackson, loss of profits caused by an interruption in business might very well be recoverable as losses arising in the usual course of things. The approach of the majority in Transfield might well lead to a different conclusion. Could the supplier be said to have ‘assumed responsibility’ for such losses arising from an interruption in business, especially given that the extent of such losses are unpredictable and potentially open ended? Whichever way the courts end up applying the Transfield decision, it seems certain that remoteness is likely to become a more contentious issue in future disputes.


The overall result of the decision, it is suggested, is the precise opposite of the certainty which the House of Lords appear to value so highly. It is suggested that it is now far from clear what question a court or tribunal should be asking itself in the context of remoteness and hence harder than ever to give clear and certain legal advice.


Simon Croall QC and Yash Kulkarni are commercial barristers who frequently act in IT disputes. Simon was lead counsel for the Owners in Transfield in the House of Lords and in the courts below.  They both practice at Quadrant Chambers, 10 Fleet Street, London, EC4Y 1AU: www.quadrantchambers.com