Classic Maritime Inc v Limbungan Makmur SDN BHD & Anor

This is an appeal from a judgment of Teare J in which he held that the charterer under a contract of affreightment was not entitled to rely upon an exceptions clause referring to “accidents at the mine” because it would not have been ready and willing to provide cargoes for shipment even if the accident had not occurred, and was therefore in breach of an absolute duty to provide such cargoes; but that nevertheless the shipowner was not entitled to recover substantial damages because this would put it in a better financial position than it would have been in if the charterer had been ready and willing to provide cargoes. As a result the judge awarded nominal damages of US $1 for each shipment in issue.

The shipowner appeals on the issue of damages. The charterer and its ultimate parent company (which acted as a guarantor) cross appeal on the issue of liability.

I shall refer to the appellant as the shipowner and to the respondents together as the charterer. There is no need to distinguish between the two respondents. It is accepted that if the first respondent, the actual charterer, is liable, the second respondent is also liable on its guarantee.

At the trial there were a number of factual issues but, as a result of the judge’s clear findings, which are not challenged on appeal, the facts can be shortly stated. They raise starkly two issues of law.

The contract between the parties was a long-term contract of affreightment providing for shipments of iron ore pellets from Tubarao or Ponta Ubu in Brazil to Port Kelang or Labuan in Malaysia on tonnage to be provided by the shipowner. The contract (in fact an addendum to a previous contract) was dated 29 June 2009. The present action was concerned with seven shipments which should have taken place between July 2015 and June 2016. In the case of the first two shipments during that period, the charterer failed to provide a cargo and has no defence. It accepts that it is liable to pay substantial damages in respect of the first shipment, which should have taken place between July and October 2015, although only nominal damages are payable in respect of the second shipment as the shipowner suffered no loss. The trial was, and this appeal is, concerned with what should have been the third to seventh shipments.

The issues arise out of a dam burst which occurred on 5 November 2015. As the judge described it:

“1.

On 5 November 2015, the Fundao dam, in the industrial complex of Germano in Brazil where iron ore is mined, burst. According to one iron ore expert who knows this area well the slurry went right down to the ocean, villages were swamped and people lost their lives. The bursting of the dam also stopped production at the iron ore mine and it is that event which has fuelled this litigation between a shipowner and a charterer.”

The mine in question was operated by Samarco Mineracao SA, which was the sole supplier of iron ore pellets for shipment from Ponta Ubu. Since August 2011 all shipments under the COA had been of pellets shipped from Ponta Ubu, although in fact no shipments had taken place during the second half of 2015 due to a collapse in demand at the Malaysian steel mills for which the cargoes were intended.

As a result of the catastrophe it was impossible to ship iron ore pellets from Ponta Ubu between November 2015 and June 2016, the period during which the five shipments in issue in this action should have taken place. Shipments from Tubarao, where Vale SA was the supplier, were not affected by the dam burst, but the judge found that Vale was unwilling to supply pellets from Tubarao to the group of which the charterer was a part, and would have been unwilling even if the charterer had made reasonable efforts to persuade it (which it did not). Thus he concluded at [132] that “it is more likely than not Vale would not have agreed to supply iron ore pellets to Antara (and hence Limbungan) to enable the five shipments in question to have taken place”.

The judge found also, at [110], that even if the dam burst had not occurred, it was “more likely than not that Limbungan would not have been able and willing to supply cargoes for shipment pursuant to the Classic COA in the months between November 2015 and June 2016”.

Accordingly, on the facts found by the judge:

The charterer’s defence was that it was protected from liability for breach of what was otherwise an absolute duty to supply cargoes by clause 32 of the contract. This provided:

“EXCEPTIONS

Neither the Vessel, her Master or Owners, nor the Charterers, Shippers or Receivers shall be Responsible for loss or damage to, or failure to supply, load, discharge or deliver the cargo resulting From: Act of God, act of war, act of public enemies, pirates or assailing thieves; arrest or restraint of princes, rulers or people; embargoes; seizure under legal process, provided bond is promptly Furnished to release the Vessel or cargo; floods; frosts; fogs; fires; epidemics; quarantine; Intervention of sanitary, customs or other constituted authorities; Blockades; Blockages; riots; insurrections; civil commotions; political disturbances; earthquakes; Landslips; explosions; collisions; strandings, and accidents of navigation; accidents at the mine or Production facility or to machinery or to loading equipment; accidents at the Receivers’ works, Port, wharf or facility; or any other causes beyond the Owners’, Charterers’, Shippers’ or Receivers’ Control; always provided that any such events directly affect the performance of either party under This Charter Party. If any time is lost due to such events or causes such time shall not count as Laytime or demurrage (unless the Vessel is already on demurrage in which case only half time to count).”

I have set out the clause with its somewhat idiosyncratic capitalisation, but as it was not suggested that this has any significance, I can ignore it for the remainder of this judgment.

It was common ground that the dam burst qualified as an “accident at the mine” and the judge found at [135] that it was “beyond the … Shippers’ control”. That finding is not challenged. The main focus of the argument on appeal has been on the words “resulting from” and the concluding proviso, “always provided that such events directly affect the performance of either party under this Charter Party”, although the final sentence, dealing with the impact on laytime of time lost due to such events or causes is also relevant.

The first issue addressed by the judge, described as the “no relevant arrangements” issue, was whether the charterer was entitled to rely on this clause in circumstances where it had made no arrangements to provide cargoes for the shipments in question at either of the two load ports. Ultimately, the judge’s conclusion at [66] was that this was essentially part of the next issue, described as the “but for” issue, so that if the charterer had in fact made no arrangements to provide cargo at either load port, “that circumstance may make it more difficult for Limbungan to establish that its failure to supply cargoes for the disputed 5 voyages resulted from the dam burst”. That was, as already noted, the factual finding which the judge went on to make.

However, the “but for” issue was whether it was necessary for the charterer to prove that, but for the dam burst, it could and would have performed the contract in accordance with its terms. The shipowner submitted that it was. The charterer disputed this, submitting that while the clause imposed a causation requirement in the sense that it had to be shown that the dam burst rendered performance of the charterer’s obligations impossible, it was not necessary for it to show also that, but for the dam burst, it would have performed its obligations. The judge accepted the shipowner’s submission, holding at [71] that:

“All must depend upon the wording of the clause. In this case clause 32 imports a causation requirement by the use of the words ‘resulting from’ and by the requirement that the force majeure must directly affect the performance of Limbungan’s obligations.”

He concluded at [73] that a reasonable and realistic businessman “would see the broad common sense of saying that if, but for the dam burst, Limbungan would not have performed its obligations, its failure to perform cannot fairly be said to have ‘resulted from’ the dam burst and the dam burst cannot fairly be said to have ‘directly affected’ the performance of Limbungan’s obligations”.

However, Mr Simon Rainey QC for the charterer relied before the judge (as he has done before us) on a line of authority stemming from the decision of the House of Lords in Bremer Handelsgesellschaft mbH. v Vanden Avenne-Izegem PVBA [1978] 2 Lloyd’s Rep 109 and dealing with the prohibition clause in the then current version of GAFTA 100. This was said to establish that it is unnecessary for a party seeking to rely on force majeure to show that it would have performed its obligations but for the force majeure. As to this, the judge accepted at [78] that “in a case of a ‘contractual frustration’ clause which is indistinguishable from GAFTA 100, clause 21, the ‘but for’ test does not have to be satisfied”, and went on to define the question for decision in the present case as follows:

“79.

The question which arises for decision in this case is whether the ‘but for’ test has to be satisfied in a force majeure or exceptions clause which does not cancel the contract for the future, like frustration, but provides a defence to a claim in damages for breach of the contract.”

The judge concluded that the “but for” test did have to be satisfied. His reasoning was that there is an important difference between a “contractual frustration” clause and an “exceptions” clause:

“80.

Mr Rainey’s argument derives support from the circumstance that the words to be construed in a contractual frustration case are in essence the same as the words to be construed in an exceptions clause. Thus, in the present case the court must construe the words ‘resulting from’ and ‘directly affect’. If those words appeared in a contractual frustration [clause] the court would give them the effect explained by Donaldson LJ in Bremer Handelsgesellschaft v Westzucker [1981] 2 Lloyd’s Rep 130. If those words appear in an exceptions clause, as they do in clause 32 in the present case, Mr Rainey submitted that they should be given the same effect. Thus Limbungan must show that the dam burst made performance of its obligations impossible so that its failure to perform ‘resulted from’ or was ‘directly affected’ by the dam burst but need not show that but for the dam burst it would have performed its obligations.”

Pausing there, “the effect explained by Donaldson LJ in Bremer v Westzucker” appears to be a reference to page 135 rhc of the report where Donaldson LJ explained that in the light of the House of Lords decision in Bremer v Vanden Avenne the words

“preventing fulfilment” in clause 21 of GAFTA 100 should be read as meaning that fulfilment was prevented by the embargo on the assumption that the seller would otherwise have been in a position to fulfil the contract. The judge continued:

“81.

I find myself unable to accept Mr Rainey’s argument. There appears to me to be an important difference between a contractual frustration clause and an exceptions clause. A contractual frustration clause, like the doctrine of frustration, is concerned with the effect of an event upon a contract for the future. It operates to bring the contract, or what remains of it, to an end so that thereafter the parties have no obligations to perform. An exceptions clause is concerned with whether or not a party is exempted from liability for a breach of contract at a time when the contract remained in existence and was the source of contractual obligations. It is understandable that a contractual frustration clause should be construed as not requiring satisfaction of the ‘but for’ test because that is not required in a case of frustration. …

As the judge’s factual finding was that the cargo would not have been supplied, the charterer was “unable to rely upon [clause 32] to excuse its failure to supply cargoes for the 5 shipments in question” (see [111]).

Finally, the judge addressed the issue of damages. The charterer submitted (see [138]) “that no substantial damages are recoverable because, even if Limbungan had been able and willing to ship the cargoes but for the dam burst, Classic would not have been entitled to substantial damages because the dam burst would in fact have prevented Limbungan from shipping any iron ore pellets”.

The judge accepted, citing among other cases Bunge SA v Nidera BV [2015] UKSC

He held, however, that a straightforward comparison between the position which the shipowner was in as a result of the breach and the position it would have been in if the charterer had supplied the cargoes was unrealistic. Instead, the correct application of the compensatory principle in this case needed to take account of the reason why the charterer was in breach of its duty to supply the cargoes:

“143.

But application of the compensatory principle in this way appears to me to be unrealistic because it ignores the reason why, on the facts of this case, Limbungan was in breach of its duty. It was in breach of its duty because, had there been no dam burst, it was more likely than not that Limbungan would not have been able or willing to ship the five shipments and so it was unable to rely upon clause 32 to excuse its breach. The realistic comparison, and the one that reflects the facts of this case, is between the position that Classic is in with the position it would have been in had Limbungan been able and willing, but for the dam burst, to supply and ship the five cargoes.

144.

If, but for the dam burst, Limbungan had been able and willing to ship the five cargoes, no cargoes would in fact have been shipped because of the dam burst and the dam burst would, in that event, have excused Limbungan from its failure to make the required shipments.

145.

For that reason, and applying the compensatory principle which determines recoverability of damages, Classic is not entitled to substantial damages for Limbungan’s failure to supply and ship the five cargoes, notwithstanding that Limbungan is unable to excuse its failure by reference to clause 32 of the COA. To award substantial damages on the facts of this case would breach that principle…

146.

This conclusion is not an impermissible sleight of hand, from not being ready to perform the COA when liability was being assessed to being ready to perform when damages are being assessed. When assessing what damages are recoverable it is necessary to compare Classic’s position with the position it would have been in had Limbungan complied with its obligations. It would be contrary to the compensatory principle, when assessing damages, to ignore what Classic’s position would have been had Limbungan been ready and willing to perform its obligations but for the dam burst. Classic cannot be put in a better position than it would have been in had Limbungan been able and willing, but for the dam burst, to ship the required cargoes.”

As a result the judge awarded nominal damages of US $1 for each shipment.

It is now common ground that, if the shipowner is entitled to substantial damages, the total damages to which it is entitled for the five shipments in issue is US $19,869,573, in addition to the damages already awarded in respect of the first shipment which should have taken place before the dam burst occurred.

It is convenient to address the cross appeal first as, if that were to succeed, the appeal on the issue of damages would not arise.

Although the parties’ submissions ranged more widely, they can be shortly summarised as follows.

For the charterer Mr Rainey submitted that clause 32 is a force majeure clause. Like other such clauses, it lists a number of events beyond the parties’ control (stage 1); it requires a causal link between such events and performance of the contract (stage 2); and it provides for the consequences which the event is to have on the parties’ obligations (stage 3). While all will ultimately depend on the terms of the clause, in general a force majeure clause operates to qualify a party’s obligations, unlike an exceptions clause which excludes or limits liability for breach. Force majeure clauses are typically concerned with events which have an impact on a party’s ability to perform, and do not require the party affected to prove that, but for the force majeure event, it would in fact have performed. That was the position here. On the true construction of clause 32, the charterer is not required to prove that it could or would have performed the contract but for the force majeure event; rather, it is sufficient that the force majeure event in fact rendered any performance impossible.

Mr Rainey relied in support of these submissions on the “settled line of authority” stemming from Bremer v Vanden Avenne which is said to establish that it is not necessary for a party to show that, absent the defined force majeure event which prevents the contract or the obligation from being performed, it could and would have performed in any event. He submitted that the distinction between “contractual frustration” and “exceptions” clauses drawn by the judge is unsupported by authority and unprincipled, and that the language used in clause 32 is typical of force majeure clauses including the clause in question in Bremer v Vanden Avenne.

For the shipowner Mr Richard Southern QC submitted that the judge was right for the reasons he gave. The issue was one of construction of clause 32 of the contract, not one of general legal principle. There was no doubt that the parties could have agreed an exceptions clause which only excused non-performance if the party invoking the clause proved that, but for the occurrence of the relevant event, it could and would have performed. The question was whether clause 32 did so provide. The Bremer v Vanden Avenne line of authority was concerned with a very different kind of clause, a “contractual frustration clause”, and was irrelevant.

I would accept, as Mr Rainey submitted, that although clause 32 is referred to as an “Exceptions” clause, it shares some of the features of a typical force majeure clause. That is to say, it lists a number of events or causes beyond the parties’ control; it defines the effect on a party’s performance of the contract which such events must have if the clause is to apply; and it specifies the consequences on the parties’ contractual responsibility when that occurs.

Ultimately, however, even if it were correct to characterise the clause as a force majeure clause, that would take the argument only so far. Undoubtedly the question is one of construction of clause 32 in this contract, and the answer to that question is determined by the language of the clause which the parties have chosen, having regard to the context and purpose of the clause. As Lord Sumption said in The Kos [2012] UKSC 17, [2012] 2 AC 64 at [12]:

“Like all questions of causation, this one is sensitive to the legal context in which it arises. It depends on the intended scope of the indemnity as a matter of construction, which is necessarily informed by its purpose.”

The purpose of clause 32 is, however, to be gathered from its terms.

I accept Mr Southern’s submission that the parties could have agreed a clause which only excused non-performance if the party invoking the clause proved that, but for the occurrence of the relevant event, it could and would have performed. The question is whether clause 32 does so provide.

In my view neither party’s construction would be particularly uncommercial or surprising. Mr Rainey submitted that it would be harsh and strange to hold the charterer liable for failing to supply in circumstances where, whatever its intentions, there was once the dam had burst never any possibility that it would be able to supply any of the five shipments in issue. That may be so but, standing back, the other side of the same coin is that it is hard to see why the dam burst should make any difference to

Accordingly I approach the construction of clause 32 without any predisposition as to the construction which should be adopted and without any need to avoid what are said to be the unfair consequences of adopting one or other of the rival constructions. It is simply a matter of construing the words of the clause.

As what matters is the language of the clause, the citation of cases dealing with very differently worded clauses is in my judgment of limited if any assistance, although the enthusiasm and diligence of counsel on both sides have demonstrated that it is possible to find examples of decided cases which fall, at least arguably, on one or other side of the line. Accordingly I do not think it would be profitable to examine the cases relied upon (Cowden v Corn Products Co Ltd (1920) 2 Ll L R 344, Brightman & Co v Bunge y Born Ltda SA [1924] 2 KB 619 and [1925] AC 799, Ross T Smyth &

Features of clause 32 which I regard as of particular importance are as follows.

First, it is a general exceptions clause of mutual application. Although this is not in any way conclusive, the heading describes the clause as an “Exceptions” clause, that heading is part of the parties’ contract as there is no clause providing that headings should be ignored, and examination of the terms of the clause shows that the heading is accurate. When it applies, the clause excuses responsibility “for loss or damage to, or failure to supply, load, discharge or deliver the cargo”. These are words which, on their face, exempt a party from responsibility for a breach of an obligation.

I would not accept, therefore, that the construction of the clause should be approached on the basis that the clause is a force majeure as distinct from an exceptions clause, or that it is helpful to import into the construction exercise the kind of general considerations which often apply to force majeure clauses, but which do not apply in the same way to exceptions clauses, on which Mr Rainey relied. Mr Rainey submitted that the purpose of the clause was to act as a force majeure clause, and that the clause should be construed accordingly, but in my judgment that begs the question.

Second, the words “loss or damage to cargo” necessarily refer to a particular cargo which was in fact lost or damaged as a result of one or more of the events listed in the clause. There is no scope for these words to apply unless, but for the event in question, the cargo would not have been lost or damaged.

The same consideration applies to the words “failure to … load, discharge or deliver the cargo”. These must refer to a cargo which, but for the event in question, would have been loaded, discharged or delivered. The use of the definite article (“the cargo”) also suggests that a particular cargo is referred to, namely the cargo which, but for the event in question, would not have been lost or damaged or (as the case may be) would have been loaded, discharged or delivered. The same is not necessarily true of the word “supply” but I see no reason why “failure to supply” should be treated differently, so far as this issue is concerned, from “failure to … load, discharge or deliver the cargo”.

Pausing there, these opening words of the clause, which apply generally to all the events which are then listed, are at least consistent with and in my judgment support the shipowner’s construction.

Third, the clause covers a very wide range of miscellaneous events. It was described by Mr Southern as lacking intellectual coherence and by Mr Rainey as a ragbag. In the case of some of these events, such as “accidents at the mine”, it may be possible to give effect to the clause on either party’s construction – that is to say, regardless of whether the party invoking the clause would have performed the contract if not for the event in question. In other cases, however, the clause can only apply if the party invoking the clause would otherwise have performed. An obvious example is “seizure under legal process, provided bond is promptly furnished to release the Vessel or cargo”. This can only refer to a seizure of the performing vessel or of the actual cargo which was loaded or destined to be loaded. Similarly, “collisions”, “strandings” and “accidents of navigation” must refer to incidents affecting the performing vessel. The fact that some events can only apply when they impact on performance by one or other party which would otherwise have taken place suggests to my mind that this must be true of the clause as a whole. The clause must be construed consistently.

Fourth, I agree with the judge that the words “resulting from” together with the requirement that the events in question “directly affect the performance of either party” import a causation requirement. That is confirmed by the words “any other causes” in the concluding part of the first sentence and the reference to “such events or causes” in the second sentence. These are not merely “events” which happen to have occurred, but “causes” which impact on performance. This is not necessarily conclusive as the causation requirement could be merely that the event renders performance impossible, but the combined effect of these phrases suggests a more demanding requirement. It is a valid use of language to say that a failure to supply the cargo (or even a cargo) does not “result from” an event if in fact the event makes no difference because the charterer was never going to supply a cargo anyway. Similarly, the proviso refers in my view to the performance which would have been rendered if the event or cause had not occurred. If the accident at the mine did not cause the charterer to do anything different because it had no intention of supplying a cargo anyway, it is fair to say that its performance was not “directly affected”. Again, therefore, the words of causation contained in the clause are at least consistent with the shipowner’s construction.

Fifth, the “time lost” provision in the concluding sentence of the clause is significant. Time cannot be lost due to an event or cause listed in the clause unless, but for the event, loading would have taken place during the time in question. Thus in Burnett Steamship Co Ltd v Danube & Black Sea Shipping Agencies [1933] 2 KB 438, the clause provided that “should any time be lost whilst steamer is in a loading berth owing to work being impossible through rain … the amount of actual time so lost during which it is impossible to work owing to rain … to be added to the loading time”. The Court of Appeal held that the clause would only apply if the charterer had cargo which was ready to be loaded during the period of rain. Greer LJ said:

“I think those words mean this: there are two things that he has to prove in order to entitle him to that extension of time, and if he fails to prove either of those two things he fails to establish his right to an extension of time. He has to prove that work became impossible through rain, and that in consequence of that he lost time in loading; unless he proves both those circumstances he does not bring himself within the clause. He did prove that there were hours of time, amounting in all to two days, in which work was impossible through rain, but he did not prove that that resulted in any loss of time by him, because on the facts as found he was not there ready to utilize the time, and therefore he cannot say that he has established that it was the impossibility of loading that caused him to lose that time.”

Although the wording of the clause in that case was not identical to the wording of the final sentence of clause 32, I consider that the same principle applies. Mr Rainey accepted this. He submitted, however, that the construction of the second sentence should not affect the construction of the first sentence. Dogs and tails were mentioned. However, I disagree. It would be surprising if a different causation requirement applied to precisely the same “events or causes” depending on whether the first or second sentence of the clause was being applied. Plainly clause 32 must be construed as a whole.

These features of the clause taken in combination persuade me that the judge’s acceptance of the shipowner’s construction was correct. I agree with his conclusion at [73] that a reasonable and realistic businessman “would see the broad common sense of saying that if, but for the dam burst, [the charterer] would not have performed its obligations, its failure to perform cannot fairly be said to have ‘resulted from’ the dam burst and the dam burst cannot fairly be said to have ‘directly affected’ the performance of [the charterer’s] obligations”.

Since, as the judge also rightly said at [71], “All must depend upon the wording of the particular clause”, this should have been enough to determine the causation issue in favour of the shipowner. However, the judge necessarily went on to examine the charterer’s submission that a different conclusion was mandated by the Bremer v Vanden Avenne line of authority, and we must do the same.

Tempting as it is to take an extended trip down memory lane, to the Mississippi floods of 1973 and the arbitral flotsam and jetsam which was washed up in the English courts for the next ten years as a result, this submission can be disposed of without requiring a detailed examination of all the GAFTA cases. The clause with which those cases were concerned, clause 21 of the then current version of GAFTA 100, was a prohibition of export clause which provided as follows:

“Prohibition

In case of prohibition of export, blockade or hostilities or in case of any executive or legislative act done by or on behalf of the Government of the country of origin or of the territory where the port or ports of shipment named herein is/are situate, preventing fulfilment, this contract or any unfulfilled portion thereof so affected shall be cancelled. In the event of shipment proving impossible during the contract period by reason of any of the causes enumerated herein, sellers shall advise buyers of the reasons therefor. If required, sellers must produce proof to justify their claim for cancellation.”

The issue was whether a seller invoking this clause had to prove, not merely that there was a prohibition of export, but also that it had goods ready to ship within the contract period and, but for the prohibition, would have performed the contract. This depended primarily on the meaning of two words, “preventing fulfilment”. Considered in isolation those words were susceptible of either meaning, although it may be that their meaning in clause 21 was coloured by the second sentence of the clause, where all that the seller had to advise the buyer was the reason why shipment had proved to be impossible.

The House of Lords held that cancellation of the unfulfilled portion of the contract occurred automatically once the prohibition occurred, with no need for the seller to prove that it had goods ready to ship within the contract period and a ship to carry them. Lord Wilberforce described clause 21 as a “contractual frustration clause” at page 112, a description which appears to have originated in the argument of Mr Anthony Hallgarten, counsel for the sellers and undoubtedly the doyen of the GAFTA cases. It is apparent that what Lord Wilberforce understood by this was a clause which cancels the contract (or part of the contract) for the future without liability on either side. This is the same sense in which the expression was used by the judge in this case.

Lord Wilberforce dealt with the issue of causation shortly at page 114:

“The clause applies ‘in case of prohibition of export … preventing fulfilment’ – so that a question may arise of causation. Was it the prohibition that prevented fulfilment or something else? This question may be phrased more specifically by asking whether the seller must prove that he had the goods ready to ship within the contract period, and a ship to carry them. The answer to it, in my clear opinion, is in the negative. The occurrence of a ‘frustrating’ event – in this case the prohibition of export – immediately and automatically cancels the contract, or the portion of it affected by the prohibition.

This, in general, is the effect of the authorities. …”

As I read Lord Wilberforce’s speech, his reason for saying that the seller did not have to prove that it had the goods ready to ship was because the effect of the clause was immediately and automatically to cancel the contract or the affected portion of it.

Mr Rainey submitted that this could not be what Lord Wilberforce had meant because it confused causation (stage 2 of the analysis: see [28] above) with consequences (stage 3), and that it was only after deciding that the event had occurred and had the requisite effect on performance that it would be relevant to consider the consequences of those conclusions on the parties’ obligations.

However, I would not accept this. As already indicated, any clause must be construed as a whole. A rigid demarcation between Mr Rainey’s three stages whereby stage 3 cannot even be considered until a final answer has been given at stages 1 and 2 runs counter to the modern iterative approach to questions of construction encouraged by cases such as Wood v Capita Insurance Services Ltd [2017] UKSC 24, [2017] AC 1173, which requires the court to check its provisional conclusions against the terms of the contract as a whole and the commercial consequences of the proposed construction.

It seems to me that where the effect of a clause is to discharge the parties from an obligation to perform in the future, as distinct from to relieve them from liability to pay damages for a past breach, that may well have at least a bearing on the nature of the causative effect on a party’s performance which an event is required to have. In such a case, both parties need to know at once when the event occurs whether they are under any continuing obligation. There is, therefore, much to be said for a simple and straightforward causation requirement (an embargo which makes it impossible for any goods to be shipped) without requiring investigation of matters known only to one party, such as whether it was able and willing to perform if the event had not occurred. Such a consideration has much less force after the event, when the time for performance is over and the only question is whether a party is liable to pay damages. Although none of this is spelled out in Lord Wilberforce’s speech, it provides in my judgment a compelling justification for understanding his reasoning as I would understand it and as Kerr LJ in Tradax v Cook did understand it. So too, it would appear, did Donaldson LJ in Bremer v Westzucker.

It is immediately obvious that clause 21 of GAFTA 100 is very different from clause

In fact, the force majeure clause in the then current version of GAFTA 100 was clause 22 and not clause 21. Those who are interested can find the terms of clause 22, broken down into its eight distinct sentences by Mocatta J, in Bremer v Vanden Avenne at first instance ([1977] 1 Lloyd’s Rep 133 at page 150). It was debated before us whether the House of Lords had decided that, in the case of clause 22 as well as clause 21, a seller invoking the clause did not have to prove that it would have performed but for the force majeure event. We were taken to, among other cases, the decisions of this court in Bremer Handelsgesellschaft mbH v C. Mackprang Jr [1981] 1 Lloyd’s Rep 292 and Tradax Export SA v Cook Industries Inc [1982] 1 Lloyd’s Rep 385. There are statements in the former case, albeit obiter, suggesting that this is what the House of Lords decided, while Kerr LJ in the latter case appears to have thought that the point was still open, a view with which Robert Goff J appears to have agreed (Bremer v Mackprang [1980] 1 Lloyd’s Rep 210 at page 222 rhc and Tradax v Cook [1981] 1 Lloyd’s Rep 236 at page 242, both at first instance).

I do not propose to enter upon this debate. It is some way removed from anything we need to decide. Certainly, however, if it is right that the reason why the House of Lords concluded in the case of clause 21 that the seller did not need to prove that it would have performed was that clause 21 was a clause which immediately and automatically cancelled the contract, that reasoning could not have applied to clause 22.

Like clause 21 of GAFTA 100, clause 32 of the contract in the present case operates automatically. Its operation does not, for example, depend on any notice being given.

Clause 32 does not provide for any automatic cancellation of the contract (or of individual shipments) for the future and, accordingly, the reasoning of the House of Lords in Bremer v Vanden Avenne cannot apply to it. It was not a “contractual frustration” clause. Ultimately, however, in deciding whether the charterer can rely on clause 32 in circumstances where it would not have performed its obligation anyway, what matters is not whether the clause is labelled a contractual frustration clause, a force majeure clause or an exceptions clause, but the language of the clause. As with most things, what matters is not the label but the content of the tin.

I turn now to the appeal on the issue of damages. This arises on the basis (which I have held to be correct) that clause 32 does not provide the charterer with a defence. The issue is whether the judge misapplied the compensatory principle.

For the shipowner Mr Southern submitted that he did. The relevant question, hallowed by long usage and authoritative exposition, required a comparison in financial terms between (1) the innocent party’s actual position as a result of the breach and (2) the position it would have been in if the contract had been performed. Here, the performance required was the supply of cargoes and, if that performance had been provided, the shipowner would have been able to earn its freight, resulting in damages (agreed as a figure) of US $19,869,573 in respect of the five shipments. The judge had made the wrong comparison. Instead of asking what would have happened if the contract had been performed, he had asked what would have happened if the charterer had been ready and willing to perform the contract, and had concluded that (in that event) it would have had a defence under clause 32. Paradoxically, therefore, having held that clause 32 did not provide the charterer with a defence, he had concluded that the effect of clause 32 was nevertheless to relieve the charterer from having to pay substantial damages.

For the charterer Mr Rainey supported the judge’s reasoning. He submitted that, on the judge’s unchallenged findings of fact, the contract could never have been performed as a result of the dam burst which was beyond the charterer’s and shipper’s control, which meant that the shipowner could never have earned freight in respect of the five shipments affected. Accordingly the relevant comparison was between the actual position in which the party in breach was unwilling to perform and the “nonbreach hypothetical counter-factual” position in which it is assumed that the party in breach was able and willing to perform (or as Mr Rainey also put it, seeking to perform) but for the force majeure event.

The compensatory principle which applies to the assessment of damages for breach of contract involves putting the innocent party in the position it would have been in if the contract had been performed. It was formulated by Parke B in Robinson v Harman (1848) 1 Exch 850, although this was no more than a statement of a principle already well established:

“The rule of the common law is, 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.”

Since then, this principle has been restated and applied many times. It is sufficient at this stage to cite two recent cases at the highest level. In The Golden Victory [2007] UKHL 12, [2007] 2 AC 353 the issue was whether, in assessing damages for anticipatory breach of a long-term charter party, it was necessary to take account of the fact that, by the time the damages came to be assessed, it was known that if performance of the contract had been continued, the contract would not have run its full term, but would have been lawfully cancelled as a result of the outbreak of the

Lord Bingham (in a dissenting judgment with which Lord Walker agreed) stated at

Lord Scott (in the majority) said:

“29.

… The fundamental principle governing the quantum of damages for breach of contract is long established and not in dispute. The damages should compensate the victim of the breach for the loss of his contractual bargain. The principle was succinctly stated by Parke B in Robinson v Harman (1848) 1 Exch

850, 855 and remains as valid now as it was then …”

After quoting Parke B, Lord Scott continued:

“If the contract is a contract for performance over a period, whether for the performance of personal services, or for supply of goods, or, as here, a time charter, the assessment of damages for breach must proceed on the same principle, namely, the victim of the breach should be placed, so far as damages can do it, in the position he would have been in had the contract been performed.”

After citing further authority, he added:

“32.

… The underlying principle is that the victim of a breach of contract is entitled to damages representing the value of the contractual benefit to which he was entitled but of which he has been deprived. He is entitled to be put in the same position, so far as money can do it, as if the contract had been performed.

…”

Putting the same point in a striking image, he said at [36] that:

“The lodestar is that the damages should represent the value of the contractual benefits of which the claimant had been deprived by the breach of contract, no less but also no more.”

Lord Carswell and Lord Brown were the other members of the Committee in the majority. Lord Carswell said:

“57.

Damages for breach of contract are a compensation to the claimant for the loss of his bargain: McGregor on Damages, 17^th ed, para 2-002. He is entitled to be placed, as far as money can do it, in the position which he would have occupied if the contract had been performed: Wertheim v Chicoutimi Pulp Co [1911] AC 301, 307, per Lord Atkinson. …”

Lord Brown at [85] expressed his full agreement with Lord Scott and Lord Carswell, as well as adding some comments of his own.

The majority held that, applying this principle, it was necessary to take account of the outbreak of war even though, at the time of the shipowner’s acceptance of the charterer’s renunciation, the outbreak of war could not have been foreseen as more than a possibility. That was because, if the contract had been performed, it would have been terminated prematurely once the war began. Putting the same point another way, the value of the contractual rights of which the shipowner had been deprived by the charterer’s repudiation was the hire which would have been earned at the charter rate up until the date of the outbreak of war, but no further. Hire beyond that date would never have been earned.

In Bunge SA v Nidera BV [2015] UKSC 43, [2015] 2 Lloyd’s Rep 469 there was a sale of Russian milling wheat on the terms of GAFTA 49, which included a prohibition of export clause providing for cancellation of the contract without liability in the event of a prohibition of export by the Russian government. Such a prohibition was announced and the seller declared the contract cancelled. However, this was premature, and therefore repudiatory, because it remained possible that the prohibition would be lifted by the time when performance was due, although in the event that did not happen and the prohibition remained in force. The buyer seized on the seller’s notice of cancellation, which it accepted as a repudiation, and claimed damages. There were two issues, the first being as to the measure of damages at common law and the second being whether that measure was excluded by the default clause of GAFTA 49. Dealing with the first issue, the Supreme Court once again applied the fundamental principle stated by Parke B:

Lord Sumption said:

“14.

The fundamental principle of the common law of damages is the compensatory principle, which requires that the injured party 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’: Robinson v Harman (1848) 1 Exch 850 at 855, (1848) 154 ER 363 at 365 (Parke B).”

Lord Toulson added:

“85.

The fundamental compensatory principle makes it axiomatic that any method of assessment of damages must reflect the nature of the bargain which the innocent party has lost as a result of the repudiation.”

Applying The Golden Victory, if there been had no repudiation and the contract had continued, the prohibition of export clause would have resulted in the contract being cancelled without liability on the part of the seller. Accordingly the value of the buyer’s contractual rights was nil and the buyer was not entitled to substantial damages. This common law measure was not excluded by the GAFTA default clause.

Both The Golden Victory and Bunge v Nidera were concerned with the assessment of damages for an anticipatory breach by renunciation which required the court to value the innocent party’s right to future performance, in the former case the right to performance of what was in effect an instalment contract with monthly hire payments

The present case is not concerned with an anticipatory breach, but with actual breaches as a result of the charterer’s failure to supply cargoes for each of the five shipments in issue. It is common ground that, subject only to clause 32, the charterer’s obligation to supply cargoes was an absolute obligation (see The Nikmary [2003] EWCA Civ 1715, [2004] 1 Lloyd’s Rep 55). Thus the performance to which the shipowner was entitled, once it was determined that clause 32 did not provide the charterer with a defence, was the supply of cargoes. The value of that performance was the freights which the shipowner would have earned if the cargoes had been supplied less the cost of earning them. In principle, therefore, the comparison which application of the compensatory principle required was between (1) the freights which the shipowner would have earned less the cost of earning them and (2) the actual position in which the shipowner found itself as a result of the breach. It is now agreed that this comparison would result in a damages award of over US$19 million.

The comparison which the judge carried out was different. It was between the shipowner’s position if the charterer had been ready and willing to perform and the shipowner’s actual position. The judge said at [146] that undertaking this comparison did not involve “an impermissible sleight of hand” but I do not agree. The character’s obligation was not to be ready and willing to supply a cargo in each case, but actually to supply one. The charterer was not in breach because it was unwilling to perform, but because it failed to do so, even if the reason why it failed to do so was because it was unwilling.

In the case of an anticipatory breach (i.e. a renunciation in advance of the time for performance), a party repudiates a contract if it demonstrates an unwillingness to perform, in which case (as in The Golden Victory and Bunge v Nidera) it may be necessary to consider whether, if it had not demonstrated that unwillingness, it would nevertheless have been excused from performance by later events. If so, that will affect the value of the rights which the innocent party has lost. But that is not so in the case of an actual breach, as in the present case. In the present case, where there is an absolute obligation to supply a cargo, whether the charterer was ready and willing to supply is neither here nor there. Nor is it relevant whether performance is impossible as (in the absence of a defence such as frustration or illegality) impossibility is not a defence: Taylor v Caldwell (1863) 3 B&S 826 at page 833. The simple fact is that the charterer failed to do what it had promised to do and is thereby in breach.

Although the judge described his approach as an application of the compensatory principle which was realistic because it took account of the reason why the charterer was in breach of its duty to supply the cargoes, this was in my judgment an irrelevant consideration in the assessment of damages. There is no case, or at any rate none which was cited to us, in which the reason why a party is in breach of contract has been held to justify, let alone require, a different approach to the compensatory principle.

An even more recent decision of the Supreme Court is One Step (Support) Ltd v Morris-Garner [2018] UKSC 20, [2018] 2 WLR 1353, in which Lord Reed once again affirmed the compensatory principle:

“31.

It is necessary next to consider some basic principles of the law relating to damages for breach of contract … The law of contract, on the other hand, gives effect to consensual agreements entered into by particular individuals in their own interests. Remedies granted by the courts are designed to give effect to what was voluntarily undertaken by the parties. Damages in contract are therefore intended to place the claimant in the same position as he would have been in if the contract had been performed.”

After citing Robinson v Harman and the numerous cases, including The Golden Victory and Bunge v Nidera, in which Parke B’s principle had been endorsed at the highest level, Lord Reed continued:

“34.

The compensatory nature of damages for breach of contract, and the nature of the loss for which they are designed to compensate, were explained by Lord Diplock in Photo Production Ltd v Securicor Transport Ltd [1980] AC 827, 848849. As Lord Diplock stated, a contract is the source of primary legal obligations upon each party to it to procure that whatever he has promised will be done is done. Leaving aside the comparatively rare cases in which the court is able to enforce a primary obligation by decreeing specific performance of it, breaches of primary obligations give rise to ‘substituted or secondary obligations’ on the part of the party in default. Those secondary obligations of the contract breaker arise by implication of law:

‘The contract, however, is just as much the source of secondary obligations as it is of primary obligations … Every failure to perform a primary obligation is a breach of contract. The secondary obligation on the part of the contract breaker to which it gives rise by implication of the common law is to pay monetary compensation to the other party for the loss sustained by him in consequence of the breach’: p. 849.

35.

Damages for breach of contract are in that sense a substitute for performance.

…

36.

… The objective of compensating the claimant for the loss sustained as a result of non-performance (an expression used here in a broad sense, so as to encompass delayed performance and defective performance) makes it necessary to quantify the loss which he sustained as accurately as the circumstances permit. What is crucial is first to identify the loss: the difference between the claimant’s actual situation and the situation in which he would have been if the primary contractual obligation had been performed. Once the loss has been identified, the court then has to quantify it in monetary terms.”

There is no trace here of any need to identify the reason for the breach or to undertake a different exercise depending on that reason. The court’s task is simply to ascertain the value to the claimant of the performance which the defendant should have rendered.

This principle has been repeatedly stated and applied over some 200 years. It is principled, firmly established and straightforward to apply. The nominal damages which the judge awarded did not result from applying the long established and fundamental compensatory principle, but instead (in my respectful view) resulted in a distortion of that principle. It led to the paradoxical result that, even though clause 32 did not provide the charterer with a defence to liability, the clause meant that it was not obliged to pay damages for failing to perform its contract.

Mr Rainey submitted that because the contract could never have been performed as a result of the dam burst, and the shipowner could never have earned freight in respect of the five shipments affected, the relevant counter-factual position for the purpose of assessing damages was that the charterer was ready and willing to perform (or was seeking to perform) but was prevented from doing so by the dam burst. He submitted that “the necessary assumption” was that the charterer was able and willing to perform and would have done so in the absence of any supervening external matter preventing it from doing so. This, however, assumes that the breach is the charterer’s unwillingness to supply a cargo and not (as is correct) the fact that it did not supply a cargo. Here is, with respect, the sleight of hand of which Mr Southern complains. The fact that the charterer was unable to perform even if it had wished to do so is irrelevant to the assessment of damages in circumstances where it undertook an absolute obligation to supply cargoes and clause 32 provides it with no defence. In those circumstances it is unable to perform its primary obligation to supply cargoes and accordingly must perform instead its secondary obligation to pay damages.

For the reasons which I have sought to explain, I would dismiss the cross appeal and allow the appeal, with the result that the shipowner is entitled to damages of US $19,869,573 in respect of the five shipments in issue in this case.

I agree. I was initially more inclined to conclude that Lord Wilberforce’s speech in Bremer v Vanden Avenne established a wider principle that where performance is in fact rendered impossible by an event falling within the scope of a contractual clause, then the party seeking to rely on the clause does not have to show that, but for the event, he would have performed his obligations. Having read the judgment of Males LJ in draft, I am now satisfied that that would state the matter much too broadly.

I do not find the attempt to distinguish between clauses on the basis of the labels “contractual frustration clauses”, “exceptions clauses” and “force majeure clauses” particularly helpful. But that does not detract from the fact that there are important differences between clause 21 from GAFTA 100 discussed in the case law and clause 32 of this contract of affreightment. Clause 21 of GAFTA 100 is a cancellation clause that has the effect that on the happening of the event which makes a shipment impossible, the parties’ obligations in respect of that shipment are immediately and automatically cancelled. The different wording of clause 32, the number and variety of supervening events contemplated together with the other factors set out by Males LJ show, in my judgment, that the parties to this contract did not intend that the other

I also agree with Males LJ that Teare J erred in his application of the compensatory damages principle when holding that the shipowner was entitled only to nominal damages. If the shipper cannot rely on clause 32 to avoid liability for breach, he cannot rely on it either to reduce the value of the shipowners’ contractual rights flowing from the breach.

I therefore agree that the cross appeal must be dismissed and the appeal allowed.

I also agree.