Black Sea Commodities Ltd v Lemarc Agromond Pte Ltd

This has been the hearing of an application under s 67 of the Arbitration Act 1996 (the 1996 Act) by the Claimant, Black Sea Commodities Ltd, to challenge an Arbitration Award, by Arbitrators, Messrs Burneski, Lucas and Sipos, dated 14 April 2020, in favour of the Defendant Lemarc Agromond Pte Ltd as buyer against the Claimant as seller (on jurisdiction and liability, subsequently followed by an Award on quantum dated 8 July 2020, also challenged by the Claimant).

The issue depends upon the question of what occurred between 9 and 14 March 2018, and whether the Arbitrators had jurisdiction to decide that there was a binding arbitration agreement in place between the Claimant and Defendant in respect of the sale of a consignment of Ukrainian corn FOB Odessa (and subsequently to make substantial findings in favour of the Defendant). The Claimant, represented by Mr Oliver Caplin, submits that there was no arbitration agreement and the Defendant, by Mr Francis Hornyold-Strickland, that there was. There was oral evidence before me, since the issue of jurisdiction was reopened, from Mr Heston, the COO of the Claimant and from Mr Gurov, a senior trader of the Defendant. The exchange of communications in March 2018 was carried out through a broker, Mr Gligoric, and there is no dispute about agency or authority.

The Defendant submits, by a late amendment at the hearing which I permitted, to add a case not run by it before the Arbitrators nor pleaded before me, that there was a binding contract of sale on 9 March. which allegedly contained a term implied by trade custom, a GAFTA arbitration clause, alternatively as his original, and still main, case, that between 12 and 14 April an exchange of draft conditions including a GAFTA arbitration clause had led, before a breakdown of communications between the parties on 14 March, to a binding arbitration agreement, whether or not there was a binding sale contract on 9 March, alternatively it constituted a variation of the 9 March agreement so as to include such an arbitration clause.

It was common ground before me that it was not my task to decide whether or not there was a binding sale agreement, but rather whether there was a binding agreement for arbitration. Mr Caplin submitted that if, contrary to his contention, there was a binding contract of sale on 9 March (as the Arbitrators found at paragraph 6.17 of their Award) it did not contain a GAFTA or any arbitration clause, and that although, in the draft conditions which the Claimant's agent sent to the Defendant on 12 March, a GAFTA arbitration clause was included, those draft conditions were never agreed, and there was no consensus ad idem by 14 March when the negotiations broke down. Hence there was no arbitration agreement agreed on 9 March, and none thereafter.

I will reserve to later in this judgment the question of the Defendant's amended case, which only materialised towards the end of the hearing.

The Defendant's primary case is as to there having been a binding agreement on 9 March, varied/supplemented by agreement as to a GAFTA arbitration clause by virtue of the subsequent exchanges. Mr Hornyold-Strickland made it clear at the outset of his opening that his case was not that there was an arbitration clause agreed on 9 March as part of the key terms then agreed, but that subsequently, in the exchange of draft conditions, whereas there was dispute about other terms, there was no dispute as to the GAFTA clause, so that the arbitration agreement became binding, either by virtue of a variation of the contract or as an independent agreement. He relied on the definition of separability of an arbitration agreement in s 7 of the 1996 Act and the obiter dicta of Lord Hoffmann in Fiona Trust v Privalov [2007] Bus LR 1719 at [10], whereby he referred, in the context of construction of contract, to the fact that “businessman frequently do want the question of whether their contract was valid, or came into existence, or has become ineffective, submitted to arbitration and that the law should not place conceptual obstacles in their way”. So here the Court should not place conceptual obstacles arising out of a strict interpretation of offer and acceptance in the way, when, as he submits, the parties intended to be bound by the GAFTA arbitration clause. Hence the pleading by the Defendant that the contract was made between 9 and 14 March.

The Arbitrators concluded, after considering the position of the parties, and the law as to sufficiency and certainty of terms of contract, particularly as set out by Lloyd LJ in Pagnan SpA v Feed Products Ltd [1987] 2 Lloyds 601 at [619], as follows at paragraph 6.17: “ The Tribunal finds that there was a valid contract entered into through the broker which was binding on those parties as of 9 March." But they then continued at paragraph 6.18 “The Tribunal having found that there was a valid contract which included ….the GAFTA .. arbitration rules… explicitly referenced in the written Contract agreed...finds that it has jurisdiction in this matter.”

The Claimant submits however that, whether there was or was not a binding contract on 9 March, there was no arbitration agreement contained in it, as there was no mention or discussion of the GAFTA or any arbitration clause on 9 March, and no consensus ad idem by virtue of the subsequent exchanges of draft conditions (including the GAFTA clause), which never led to agreement.

I was referred by the parties to a large number of valuable authorities in addition to Pagnan as to whether and when there is a sufficient and complete contract, particularly where there is a perceived need to have subsequent documentation. The cases included Perry v Suffields [1916] 2 Ch 187, May & Butcher Ltd v the King (Note) [1934] 2 KB 17, The Bay Ridge [1999] 2 Lloyds 227, Mamidoil Jetoil Greek Petroleum Co SA v Okta Crude Oil Refinery AD [2001] 2 AER (Comm) 193, Cheverney Consulting v Whitehead Mann Ltd [2006] EWCA Civ 1303, RTS Flexible Systems Ltd V Molkerei Alois Muller GmbH [2010] 1 WLR 753, Immingham Storage Co Ltd v Clear plc [2011] EWCA Civ 89, Air Studios (Lyndhurst) Ltd v Lombard North Central plc [2013] 1 Lloyds 63 and Proton Energy Group SA v Orlen Lietuva [2014] 1 Lloyds 100.

The facts in these cases tend to fall into two categories (though they are sometimes both present):

Whether there was an understanding, express (e.g. 'subject to contract‘ – not the case here) or implied, that there was no binding agreement until documents were complete, or some other condition precedent to being bound such as the effecting of a letter of credit – 'condition precedent cases‘, and – but often overlapping

Whether there was objectively agreement as to the essential/cardinal terms or sufficient terms as allow it to be concluded objectively that there was a complete and binding contract even if negotiations continued as to lesser terms, or documents had to be signed or agreed – 'consensus ad idem’ cases.

The parties here were agreed by the end of 9 March as to the following terms:

Seller – to be one of two associated companies – Mr Caplin submits that this uncertainty was material, but I am not persuaded –and in any event there was clarification between the broker and the Defendant by WhatsApp on 9 March:

Buyer:

Quantity of Ukrainian corn - 50,000 MT:

Either in one or two shipments/’bottoms’: v) Quality:

The parties were not then agreed as to other terms, which were set out in the draft conditions sent by the broker , after approval by the Claimant, to the Defendant on 12th March, dated 9 March, (subject to an extra condition sent on by the Claimant via the broker later in the day). These included the following:

GAFTA terms including the arbitration clause: ii) Loading instructions and fumigation procedures:

By 14th March after three exchanges of drafts using 3 colours, red, yellow and blue for changes and comments, there was one issue left in dispute between the parties, namely the NOR spread, to which the Defendant would not agree.

If it were for me to decide whether there was a binding agreement on 9 March, I would agree with the Arbitrators' decision that there was, in their paragraph 6.17:–

The words used by the Defendant and the broker on 9 March – "firm offer"/"firm counter-offer"/“Pls book it" "Ok book it" "Booked on 205.20" – "Super”) - are suggestive of a concluded agreement:

It is apparent that there was “speed of the market” (as in Proton Energy at [39]), in that the Defendant and the broker gave each other variously 15 minutes and 5 minutes to respond. Mr Heston agreed that at any rate the price would have been agreed and fixed because of the volatility of the market.

In my judgment, by reference to the terms which were agreed, set out in paragraph 11 above, objectively there was sufficient agreement of the essential/cardinal terms, even if, there were, as Mr Hornyold-Strickland put it, “ancillary” terms left to be considered later, such as those in paragraph 12, and even though the Claimant explained the importance to it of the NOR Spread (originally described by the broker as a minor amendment).

Hence I would reach the same conclusion as the Arbitrators in paragraph 6.17. However they gave no roadmap to their conclusion, which immediately followed, in paragraph 6.18, also referred to in paragraph 7 above, and I can see no basis for it. If there was a binding agreement on 9 March there is no sign in the communications between the parties that it included a GAFTA arbitration clause, and there were of course no draft conditions on 9 March making reference to it.

After 9 March:

The broker sent to the Defendant the draft conditions in the form approved by the Claimant at 11:19 on 9 March. There then begins what Mr HornyoldStrickland called the ping-pong.

The Defendant sent them back amended at 16.05, with a number of changes in red. Not all were accepted by the Claimant, and on 13 March the Claimant sent back to the broker, who sent on to the Defendant, a copy of the conditions, setting out which of the proposed changes were acceptable and which were not, with the additional term as to NOR Spread.

Two of the unagreed amendments were not accepted. iv) After further email negotiations there was a sticking point on the NOR Spread.

Mr Hornyold-Strickland submits that, while there was not agreement on all the amendments in the course of the ping-pong, the GAFTA clause remained un-objected to through the exchanges. He submits that there was thus agreement to an arbitration clause by conduct, by virtue of the fact that it was not objected to, while other clauses remained in contention. Thus, by analogy with Fiona Trust, just as an arbitration contract can survive and indeed resolve issues where the underlying agreement may be, for example, rescinded for fraud or representation, so arbitration, by reference to an arbitration clause, can still be the source of resolution of whether there was or was not consensus to the underlying agreement.

Mr Hornyold-Strickland refers to s 7 of the 1996 Act:

He prays in aid Lord Hoffmann‘s dictum and refers to an authority from which he says he can draw support for his proposition, at any rate by reference to obiter dicta. In UR Power GmbH v Kuok Oils and Grains Pte Ltd [2009] EWHC 1940 (Comm) Gross J found that there was a binding sale contract, and that there was no unfulfilled condition precedent (in that case said to have been the opening of a letter of credit). Considering, obiter, what his conclusion would or might have been had he decided that there was no binding contract, he referred to s 7 and to the possibility that an arbitration agreement may be binding, even though the underlying contract had not come into existence (at [40 (iii)). However: –

In that case the agreement said not to be binding did incorporate a FOSFA arbitration clause:

He made it quite clear that his obiter provisional conclusion was directed to a case where an arbitration clause was contained in a contract said to be subject to a condition precedent and that “it is perhaps to be underlined that the question here went not to the existence of any consensus ad idem but instead to the nature of Kuok's obligation.” (I shall refer to this as his ‘caveat’).

It is not surprising that he made this caveat because, if there was no consensus ad idem to the underlying contract, it would ordinarily be likely that there was no consensus as to the arbitration contract either. Indeed this becomes clear from the two authorities relied upon by Mr Caplin, Pacific Inter-Link SDN BHD v EFKO Food Ingredients Ltd [2011] EWHC 923 (Comm) and Hyundai Merchant Marine Company Ltd v Americas Bulk Transport Ltd [2013] 2 AER (Comm) 649. In both these cases, Gross J’s obiter view, or as referred to by David Steel J at [40 H] in Pacific his “provisional inclination”, is referred to but not regarded as apt. In Hyundai at [35 (c)] Eder J explains that “If there was no consensus at all… the lack of consensus not only prevented any charter from coming into existence but also any arbitration agreement from coming into existence.“ He refers expressly to Gross J's caveat. David Steel J is quite clear that the contention put forward before me by Mr HornyoldStrickland, of partial acceptance of terms not rejected, described by Mr Caplin as his 'pick and mix' case, is not well founded:

I do not consider, quite apart from reference to those authorities, that the 'pick and mix’ approach to offer and acceptance can be justified in law:

Mr Hornyold-Strickland's reference to “rest as per below“ (paragraph 11), relevant to consecutive emails adding terms which are then themselves agreed in addition to the previous ones, is of no support in relation to a case of supply of draft conditions in total which are then either accepted or not in total, and not by way of 'building blocks'.

His 'building block' case of supplementation of an agreement is not supported by any standard application of offer and acceptance, as David Steel J makes clear. He had to depend upon Lord Hoffmann in order to suggest that an arbitration clause is “sui generis". He submitted that an arbitration agreement is unlike any other contractual clause, and the fact that there are other terms still in issue does not determine whether the arbitration agreement is separable: he accepted and asserted that such argument would not apply to a jurisdiction clause or a choice of law clause, because of his case that arbitration clauses require special treatment. But Lord Hoffmann's obiter advice as to an approach to construction cannot support this edifice.

In my judgment, before one can apply s 7 or Lord Hoffmann's obiter to an arbitration clause there has to be an arbitration agreement. That was addressed in terms by the House of Lords in May & Butcher v The King, upholding the decision to that effect of Rowlatt J, as per the words of Viscount Dunedin at [22]:

Accordingly, I am satisfied that if there was no binding agreement on 9 March (as the Claimant contends), there was no subsequent consensus ad idem to an arbitration clause, But in any event, if, as I consider likeliest, there was a binding agreement on 9 March, the GAFTA arbitration clause was not one of the terms set out (paragraph 11 above, and as conceded by Mr Hornyold-Strickland(paragraph 6 above), and it was not agreed then or by virtue, or in the course, of the ensuing abortive exchange of draft conditions between 12 and 14 April, whether by 'pick and mix' or progressive supplementation or silent agreement by conduct.

Mr Hornyold-Strickland proposed his amendment in the course of his closing submissions, and in the light of the evidence before me:

At my direction, Mr Hornyold-Strickland served on the day after the hearing an amendment of the defence, setting out what he had orally submitted, namely now pleading that the contract on 9 March “included a term implied by trade/custom that this grain trade contained a provision for GAFTA arbitration”.

The case that there was an arbitration clause on 9 March as part of the binding terms agreed was expressly abjured by Mr Hornyold-Strickland at the outset, but it now became his alternative case, and in the light of my conclusions above it would be the only basis on which he could now succeed.

In the context of this very late amendment, I gave permission to amend, because I could see no prejudice to the Claimant, provided that Mr Caplin was given the opportunity to put in submissions in response, but I made an order for the parties to exchange post-hearing submissions, which they have now done. I have been assisted by those submissions and by the presentation of two bundles of authorities, one from each party, which I have read. Both counsel use as the foundation for their submissions the analysis of terms implied by custom and usage in Chitty on Contracts (33rd Ed) at 14–033–5.

The case now put forward by the Defendant is by reference to a term allegedly implied by virtue of custom/usage of the trade in Ukrainian corn F.O.B. Odessa that GAFTA arbitration would be incorporated in such contracts. As was common ground, it was not a question of prior dealing between the parties, as there had been none.

In order for this to be established, the Defendant must establish a trade custom/usage that is, as per Chitty at 14–033 “an invariable, certain and general usage or custom of any particular trade or place”. Mr Hornyold-Strickland made some play in his submissions with the fact that in my judgment in Durham v BAI (Run off) Ltd [2009] 2 AER 26, upon which (inter alia) Mr Caplin relied, I had, in setting out the requirements for the usage which was relied upon as allegedly being implied in the insurance market, with which I was there dealing, at [181 (iv)], stated that such usage as is alleged and relied upon must be “binding, in the sense that it is not simply a usage which parties choose from time to time to follow”. It is plain that I was using the word ‘binding’ in the sense of binding on the market, as in General Reinsurance Corporation and Others v Forsakringsaktiebolaget Fennia Patria [1983] QB 856, where the words of Kerr LJ are set out at 871: “I do not think that the evidence was sufficient to establish any binding custom“, and Slade LJ at 874 approved the words of Ungoed-Thomas J in Cunliffe-Owen v Teather & Greenwood [1967] 1 WLR 1421 at [1438], another case relied upon by Mr Hornyold-Strickland: “What is necessary is that for a practice to be a recognised usage it should be established as a practice having binding effect.”

This is simply part of establishing on the evidence (of which there was a great deal as to the market in Durham) a custom which is invariable, as is required by the words of Chitty set out above (regularly applied, as for example by the Court of Appeal in

The Defendant did not put forward such a case before the GAFTA arbitrators, who would have had some knowledge as to whether this was so. It did not appear in any evidence for the Defendant, and no evidence such as is normally produced to prove such a trade custom or usage was adduced by the Defendant: no expert witness, no evidence from other traders and no documentation. As Mr Caplin pointed out in his submissions, in Durham there was a good deal of such evidence, but I did not accept it as sufficient.

The evidence now relied upon by the Defendant is that of Mr Heston that he always included a GAFTA arbitration clause in his concluded contracts and from Mr Gurov that he always dealt on that basis, but he additionally asserted that “I have never heard about any single trade on the Ukrainian F.O.B. corn market based not on the GAFTA

As I was not persuaded in Durham as to the existence of the custom, despite considerable evidence as to its existence, I am not so persuaded in this case.

There would be significant other problems for the Defendant in establishing this amended case:

I am not persuaded that a term that there be “provision for GAFTA arbitration” is sufficiently certain. The arbitration clause in the draft conditions here was not as per GAFTA 49.

I am not persuaded that, particularly in the absence of prior dealing, an arbitration agreement can be implied without any reference “in an agreement to a written form of arbitration clause or to a document containing an arbitration clause” as required by s 6 (2) of the 1996 Act. Implication of such a term is doubted in Russell on Arbitration (24th ed) at 2–061–3 and speculated as a possibility in Merkin and Flannery on the Arbitration Act 1996 (6^th ed) at 6.2.3.

I conclude that the Defendant cannot succeed on his new case, and the s 67 application by the Claimant accordingly succeeds.