[Home] [Databases] [World Law] [Multidatabase Search] [Help] [Feedback]
	England and Wales High Court (Commercial Court) Decisions
You are here: BAILII >> Databases >> England and Wales High Court (Commercial Court) Decisions >> Tyson International Company Ltd v Partner Reinsurance Europe SE (Rev1) [2023] EWHC 3243 (Comm) (15 December 2023) URL: http://www.bailii.org/ew/cases/EWHC/Comm/2023/3243.html Cite as: [2023] EWHC 3243 (Comm)

[New search] [Printable PDF version] [Help]

IN THE HIGH COURT OF JUSTICE
BUSINESS AND PROPERTY COURTS OF ENGLAND AND WALES
COMMERCIAL COURT (KBD

		The Rolls Building 7 Rolls Buildings Fetter Lane London EC4A 1NL
		15th December 2023

STEPHEN HOUSEMAN KC :

Introduction

1. It has been observed at the highest level that anti-suit injunctions and mandatory stays for arbitration are "opposite and complementary sides of the same coin" in the context of enforcing an arbitration agreement. Both remedies, one equitable, the other procedural, enforce a party's promise to arbitrate, therefore not to litigate a particular dispute. A stay is granted by the court which is substantively seised of the relevant dispute, in this case pursuant to section 9 of the Arbitration Act 1996, reflecting Article II(3) of the New York Convention. An anti-suit injunction ("ASI" for short) may be granted if this court has personal jurisdiction over the defendant in respect of such claim, whether or not it happens also to be the curial court, i.e. the forum of chosen arbitral seat. The source of such power is section 37 of the Senior Courts Act 1981.
2. The interplay between these converse remedies is illustrated by the present case, albeit through unusual circumstances. The unusual feature of this situation is that there are two distinct contracts covering the same legal relationship, each providing for different applicable law and dispute resolution forum. This means that the choice between converse remedies is a direct product of which contract prevails. Only one court is involved. In contrast, the normal position involves one contract and two different courts.
3. There are two applications before the court. They were heard together in a full day of argument on Wednesday, 13th December. The court sat in private for 25 minutes or so of the hearing, so that matters relating to an arbitral process could be discussed without compromising their confidential status.
4. As to the two applications, first, there is an application by the defendant, who I refer to simply as "D", dated 24th May 2023, seeking a stay under section 9 of the 1996 Act that the claim commenced in this court by the claimant, who I refer to as "C", three weeks earlier on 3rd May 2023. The basis for this application, which I refer to as the stay application, is that the action comprises a "matter" or "matters" which the parties have agreed to refer exclusively to arbitration in New York, pursuant to clause 13 of a reinsurance contract executed on 8th July 2021. The second application was made by C on 3rd November 2023 and at a time when this one-day hearing had been listed for some months to be heard over one day on 13th December.
5. By this cross-application C seeks an ASI to restrain or stop the further pursuit of arbitration proceedings commenced by D (as arbitral claimant) against C (as arbitral respondent) in New York almost six months earlier on 4th May 2023. (I refer to this arbitration neutrally as one that is contractually effective to avoid using the words "alleged" or "purported" or "putative" or their adverbial equivalents on each occasion).
6. The basis of this injunctive application is that the relevant dispute was agreed to be referred exclusively to the jurisdiction of the English court in a term appearing on the fourth page of a reinsurance contract dated 30th June 2021. Since C seeks to restrain pursuit of foreign-seated arbitral proceedings as distinct from first-seised foreign court proceedings, it is, technically speaking, an anti-arbitration injunction. The parties nevertheless refer to it as an ASI. Whether anything turns on this is addressed further below.
7. It might be thought at this point that the two reinsurance contracts cover different risks or subject matter, or perhaps cover different parties or periods. This is not so. Both contracts cover precisely the same risk, period and parties. Each of them could or would be a self-standing and self-sufficient contract if viewed in isolation from the other. The reinsurance itself is an excess layer facultative cover in respect of an original policy of insurance written by C as captive insurer of Tyson Foods in the USA, for the year 1st July 2021 to 1st July 2022 (the "2021-2022 year").
8. The terms of the original policy are incorporated into the reinsurance in different ways in each of the two competing reinsurance contracts. Both bear the same policy number taken from that original insurance, namely PRPNA 2103490. The prior year (the "2020-2021 year") used a policy number PRPNA 2003490. The renewal into 2021-2022 was not a like-for-like or literal renewal - the excess layer was materially different, other terms were different.
9. The two reinsurance contracts were agreed just eight days apart in time. A dispute of substance is whether the later contract varied or superseded the earlier contract. More specifically for present purposes the dispute is whether the forum selection clause, New York arbitration in the later contract, replaced or supplanted the corresponding provision, English exclusive jurisdiction in the earlier contract.
10. Quite sensibly, neither side contended for a position in which both forum selection clauses stood with overlapping scope covering the substance of the relevant dispute. For the parties to have agreed an optional and exclusive - hence preclusive - forum selection regime like this would have taken clear language even in a single contract, let alone in or between two separate contracts covering the same commercial subject matter. So this is not a case of who started first. There was no election between English court and New York arbitration.
11. As regards burden of proof or persuasion on this central issue, D has the burden of showing that the New York arbitration agreement prevails in order to invoke section 9, whilst C has the burden of showing that the English jurisdiction clause prevails in order to seek an ASI. Both must prove their case to the final standard in order to be able to get their respective remedy. The case does not, however, come down to who has the burden.

Relevant Background

12. A great deal of evidence has been served by the parties, including eight witness statements in total, which in places seeks to show that there is an applicable market custom or practice as to use of each type of contract involved in this placement. I am not persuaded that this evidence ultimately matters when ascertaining the common intention of the parties as a matter of English law. The essential inquiry is what the parties agreed as their mandatory forum selection clause: exclusive English jurisdiction or New York arbitration?
13. Having said that, I adopt the market form terminology to describe each of the two contracts. I refer to the first contract as MRC, standing for Market Reform Contract as widely known and used in the London market. I refer to the second contract as MURA, standing for Market Uniform Reinsurance Agreement as widely known and used in the US market.
14. Nothing turns for present purposes on the nature or scope of the underlying dispute. D has purported to avoid the reinsurance. This is contested by C. Both the substantive claim in this court and in New York arbitration concern that avoidance dispute.
15. The MRC is a 35-page document which identifies itself as Policy Number PRPNA 2103490. It provides, "All Risks of Direct Physical Loss Or Damage" reinsurance in respect of identified original insurance for the period 1st July 2021 to 1st July 2022. It follows the format of a slip or slip policy with bold upper case headings on the left-hand side, then a typed-in provision to the right of each heading. There are four pages relating to the key terms of reinsurance (pages 2-5; page 1 being a title page) and a further 30 pages of additional terms and details concerning the original insurance all following the same slip layout. Many standard London market clauses are set out in full in these pages.
16. The fourth page includes a side heading "CHOICE OF LAW & JURISDICTION", which reads as follows:

"This Reinsurance shall be governed by and construed according to the Laws of England and Wales. The Courts of England and Wales shall have exclusive jurisdiction of the parties hereto on all matters relating to this insurance."

17. The fifth page includes a side heading "INSURER CONTRACT DOCUMENTATION", which reads:

"This document details the contract terms entered into by the (re)insurer(s) and constitutes the contract document.

Any further documentation changing this contract, agreed in accordance with the contract change provisions set out in this contract, shall form the evidence of such change..."

18. The phrase "contract change provisions" is a reference to a later clause on page 24 of the MRC, headed "BASIS OF AGREEMENT TO CONTRACT CHANGES". The first two paragraphs of that provision read as follows:

"All changes to be managed and agreed in accordance with the General Underwriters Agreement (version 2.0) February 2014 and the GUA Non-Marine Schedule (October 2001). Non bureaux markets to follow the agreement of the slip leader unless otherwise stated.

As regards Contract Change Endorsements where full market approval is deemed not necessary within the provisions of the GUA then, when required Lockton Companies LLP may be permitted to utilise email facilities to supply the 'follow' Underwriters with scanned copies of such Contract Change Endorsements for their records."

19. Reference is made here to the General Underwriters Agreement ("GUA" for short) to which I return below.
20. As noted already, this contract was in the nature of a Market Reform Contract. There is no dispute that, without more, it was and would have remained a legally binding and full contract of reinsurance. On this basis C commenced proceedings in this jurisdiction and now seeks ASI relief.
21. The MURA was entered into just over a week later, on 8th July 2021. It is dated 1st July 2021, i.e. the same date as the MRC. This document was foreshadowed by the brokers as forthcoming. The cover was placed and spoken of as a facultative certificate or "fac cert" for short. The premium would not be administered until this later document was executed according to those contemporary placing communications.
22. This contract is a 10-page documented entitled "AGREEMENT OF FACULTATIVE INSURANCE", which also bears the reference "AGREEMENT No: PRPNA2103490".
23. The first two pages are headed "DECLARATIONS", and contain the primary features of the reinsurance. Pages 3-9 comprise printed "TERMS AND CONDITIONS", which set out 30 numbered clauses. Clause 13 is headed "Arbitration". It provides as follows:

"a. As a condition precedent to any right of action hereunder, any dispute arising out of the interpretation, performance or breach of this Agreement; including the formation or validity thereof, shall be submitted for decision to a panel of three arbitrators. Note requesting arbitration will be in writing and sent certified or registered mail, return receipt requested.

b. Each Party shall choose one arbitrator and the two arbitrators shall, before instituting the hearing, choose an impartial third arbitrator who shall preside at the hearing. If either Party fails to appoint its arbitrator within thirty (30) days after being requested to do so by the other Party, the latter, after ten (10) days note by certified or registered mail of its intention to do so, may appoint the second arbitrator.

c. If the two arbitrators are unable to agree upon the third arbitrator within thirty (30) days of their appointment the arbitrators shall implement the ARIAS-US umpire appointment procedure to select the arbitrator.

d. All arbitrators shall have at least ten (10) years of insurance or reinsurance experience, be disinterested and active or former officers of insurance or reinsurance companies with knowledge about the lines of business at issue.

e. Within thirty (30) days after notice of appointment of all arbitrators, the panel shall meet and determine timely periods for briefs, discovery procedures and schedules of hearings.

f. The panel shall be relieved of all judicial formality and shall not be bound by the strict rules of procedure and evidence. Unless the panel agrees otherwise, arbitration shall take place in New York, but the venue may be changed when deemed by the panel to be in the best interest of the arbitration proceeding. Insofar as the arbitration panel looks to substantive law, it shall follow the law of New York in accordance with the dictates of the Governing Law Clause. The decision of any two arbitrators when rendered in writing shall be final and binding. The panel is empowered to grant interim relief as it may deem appropriate.

g. The panel shall interpret this Agreement as an honorable engagement rather than as merely a legal obligation and shall make its decision considering the custom and practice of the applicable insurance and reinsurance business as promptly as possible following the termination of the hearings. Judgment upon the award may be entered in any court having jurisdiction thereof.

h. Each Party shall bear the expense of its own arbitrator and shall jointly and equally bear with the other Party the cost of the third arbitrator. The panel shall allocate the remaining costs of the arbitration. The panel may, at its discretion, award such further costs and expense as it considers appropriate, including but not limited to attorneys fees, to the extent permitted by law."

24. Clause 17 is headed "Governing Law and Jurisdiction". It provides as follows:

"Insofar as the panel looks to the law of a jurisdiction as governing law, it will apply the substantive law of the State of New York without reference to that state's choice or conflict of laws rules; provided, however, that the substantive law of the State of New York shall not be used to supplant or override underlying court or other judicial body final decisions concerning the claim(s) at issue."

25. Clause 26 is headed "Entire Agreement". It provides as follows:

"This Agreement including any duly executed written amendments and endorsements thereto, and appendices, schedules or other attachments made part thereof or expressly incorporated by reference, and the Policy and any written endorsements, modifications, alterations and cancellations thereto, and waivers and interpretations thereto but only with respect to the claim in dispute, all as permitted under Reinsurance Agreement Clause 2 and Reinsurance Accepted Clause 3, shall constitute the entire agreement between the Parties and shall supersede all contemporaneous or prior agreements and understandings, both written and oral, between the Parties with respect to the subject matter hereof provided, however, that this Clause 26 shall not override or take precedence over Clause 3 hereof."

26. The 8th July 2021 agreement is based on a US standard form known as the Market Uniform Reinsurance Agreement. I am satisfied that, viewed in isolation, this agreement would be a binding contract of reinsurance, governed by and therefore according to New York law. The arbitration agreement in clause 13 would, therefore, be governed by New York law, according to either English or New York private international law principles, although nothing seems to turn on this.
27. D contends that the MURA varied or superseded the MRC such that any relevant dispute is governed by New York law and subject to New York arbitration. On this basis D commenced arbitration against C on 4th May this year, the day after C started these proceedings against D. In that arbitration, C has objected to arbitral jurisdiction from the outset.
28. A three-person tribunal was convoked or constituted on 28th August 2023. D sought a temporary stay pending the outcome of D's pending stay application in this court, which had in the meantime been listed for 13th December. The arbitrators refused such stay in mid-September. They have not formally ruled on their substantive jurisdiction but they have stated that in their view the New York arbitration agreement in the MURA confers jurisdiction upon them to resolve the relevant dispute. The stated basis for this is that the MURA is a binding contract of reinsurance which is later in time than the MRC, i.e. it constitutes the parties' final word on forum selection.
29. For its part, D says that the MURA is merely a ministerial instrument or piece of administrative or recapitulative paperwork which is habitually or customarily issued by or agreed with a reinsurer; but which does not, without more, alter the contractual terms of the reinsurance already written pursuant to the MRC.
30. If C is correct on the threshold contractual / forum selection issue, a question arises as to whether C should be granted ASI relief after having waited until 3rd November to apply for it. I address that separately below. A final merits hearing is scheduled in the arbitration for October 2024.

Forum Selection: Analysis

31. The court faced with a threshold jurisdictional contest of this kind, on a section 9 stay application, need not determine the contractual position finally. It has a range of options open to it. Here, however, the parties have served eight witness statements across the two applications, and counsel have filed skeleton arguments making (at least) full use of the 25-page allowance for a one day hearing. The threshold contractual issue is fully evidenced and fully analysed. I consider it appropriate to determine that question.
32. Having received full argument on this substantive issue, I am satisfied as a matter of English law that the English jurisdiction clause and English choice of law in the MRC was or were replaced by the New York arbitration agreement (clause 13) and New York choice of law (clause 17) in the MURA. The latter contract was expressly contemplated by the parties through their brokers at the time of execution of the former contract. The MURA was proffered for consideration and agreement, and separately signed and agreed on both sides. It describes itself and defines itself as an "Agreement". It contains all the operative terms to be a contract of reinsurance, albeit one governed by New York law.
33. The fact that contractual documentation taking the form of a slip policy and subsequent certificate of cover are habitually used in the London or US reinsurance market cannot dictate the contractual effect of the MRC and MURA as used by the parties in this specific instance. This is a question of common intention, objectively ascertained in context.
34. In so far as it is said that there is a fixed or notorious market custom or practice which alters what would otherwise be the objective ascertainment of common intention in this instance, I reject that submission and make no such finding. The evidence on such matters is contained in witness statements from non-independent people whose role is to inform the court as to matters of fact within their recollection.
35. Mr. Killen, on behalf of C, made a number of submissions in opposition to the conclusion I have summarised, albeit without prejudice to the burden of proof on the stay application. I deal with what I regard to be the five most powerful objections.
36. First and foremost, it is said to have been inherently improbable that the parties would agree to replace such fundamental provisions of their binding reinsurance contract within eight days of it being executed. To do so would require the clearest of language, and that is not found in the wording of the MURA, including its 30 printed terms and conditions ("T&Cs" for short). Nowhere in the MURA does it say, in terms, that it varies or supersedes or otherwise alters anything in the MRC.
37. I am satisfied that, however unlikely or unusual it is for contracting parties to replace one specific agreement with another specific agreement in the space of a week or so, there is nothing problematic with them doing so. It all depends on what they said and did. Here, as summarised above, the contracting parties concluded a new legally binding contract of reinsurance on the terms of the MURA. This was contemplated in advance, albeit different from what they did in the prior year, for what it is worth.
38. Second, and related to the first point above, C contends that such an approach is commercially absurd. Why would parties do such a thing? The answer is that they did here, and there is no binding market custom or practice which precludes them doing so in the sense of imposing an immutable characterisation or status upon the issuance of a MURA, or certificate, following the conclusion of a binding slip policy or MRC.
39. I could have sympathy with an argument that the parties were less likely to have intended to replace key provisions of their binding reinsurance contract within eight days where that might lead to uncertainty or complexity in terms of whether cover exists for a loss that impacts the policy in such period. But no such uncertainty or complexity has been identified. A loss that hits the cover when governed by English law for the first eight days post-inception is unaffected once the cover becomes governed by New York law with either retrospective or prospective effect. Nothing falls between the posts, no duplication or conflict arises.
40. The argument of commercial absurdity, as with the primary argument of inherent improbability, is no more than an appeal to empirical normality. I reject an analysis based on empiricism or normativism. What these contracting parties objectively agree is to be found in what they did and said, albeit viewed in the context of what is normal in their market.
41. The third objection is related to the two above. C identified several clauses, including both extensions and exclusions of cover, which are incorporated into or set out separately in the MRC, but which are not or would not be included in any contract of reinsurance contained in or evidenced solely by the MURA. This was referred to as the "swept away" argument in the hearing.
42. Whether or not each such clause would or would not form part of the reinsurance agreement comprised in the MURA may be a matter of dispute between the parties. In some instances D, or indeed C itself, may seek to contend that such provision did form part of the reinsurance, because (for example) terms of the original insurance policy written by C were incorporated as a matter of New York law. A claim for rectification might in the future arise under New York law. Clause 26 of the MURA may feature in such analysis. These disputes, if they ever arise, would be for arbitration in New York.
43. It is difficult to see how such matters would form part of the current arbitration concerning disputed avoidance. That is a matter for the arbitrators. Such dispute could later arise if the arbitrators conclude in their award that the reinsurance has not been avoided. That would be a matter for a further reference to arbitration, if not agreed to be included in the current reference or otherwise engaging the current tribunal. Come what may, this is all for arbitration in New York.
44. The fact that these matters are left over by this judgment is simply a reflection of the primary conclusion, which is all that is necessary to determine the applications before me, namely the parties chose New York arbitration for all disputes relating to their contract of reinsurance.
45. The fourth main objection to the adoption of the New York arbitration agreement in the MURA in place of the English jurisdiction clause in the MRC is that any variation to the first contract required compliance with the so-called "change of contract" provisions contained in the MRC. I set these provisions out above. C contends that such compliance was mandatory; or, at any rate, non-compliance with such formalities was a powerful indication against the parties having intended to effect such a material change to the reinsurance contract agreed just eight days earlier.
46. Such change requires a contract endorsement which complies with the requirements of the GUA, including the stamp of the underwriter as agent for the following market. The rationale for these provisions is certainty and unity of cover across the lead and following market on the relevant layer of reinsurance. Although there was no following market on this excess layer of facultative reinsurance in the 2021-2022 year, the GUA formalities could nevertheless be applied literally to issue a contract endorsement which thereby effects a "change of contract" in accordance with the express terms of the MRC. This is the argument.
47. In support of that submission, C points to the way that the equivalent facultative certificate, i.e. MURA, was executed and incorporated in the prior year. That was done by a contract endorsement to or under the MRC, albeit not stamped in accordance with the GUA. However, the fact it was done this way in the prior year does not really assist me in ascertaining the parties' common intentions in the present context. As noted above, there was no like-for-like or literal renewal into the 2021-2022 year. No estoppel or course of dealing with extra-contractual effect is alleged by C. The parties through their brokers did things that way one year but did things another way the next year, so it seems. Tellingly, there was no GUA stamp on the MURA endorsement in the prior year, so on any view the parties appear to have departed from the strict formality of their own "change of contract" wording that year.
48. I do not regard the "change of contract" wording as sufficient to contra-indicate the parties' intention to replace the English jurisdiction clause in the MRC with the New York arbitration agreement in the MURA by execution of the latter agreement eight days later. Whilst it could be said, and was submitted by D, that the MURA itself fulfils all the requirements of an endorsement as described in the GUA, the better view in my judgment is that the parties simply entered into a further contract which was not an attempt to "change" the MRC according to its own terms regulating such things.
49. Contracting parties are sovereign as to the terms and duration of their own bargain. Here, they exercised such sovereignty to replace the MRC with the MURA, at least to the extent of the forum selection and governing law provisions. None of the problems associated with oral variations are engaged in this scenario: cf. MWB Business Exchange Centres Ltd. v. Rock Advertising Ltd. [2018] UKSC 24; [2019] AC 119.
50. On one view, this could have amounted to an implied recission. The execution of the MURA is necessarily inconsistent with the subsistence of the MRC; hence the former replaced the latter by cancelling it as a matter of English contract law analysis: see the recent decision of Frangou v. Frangos [2023] EWCA Civ 1320 at [96]-[99]. It may be conceptually possible for a separable jurisdiction agreement within a main contract to be impliedly rescinded by the parties' subsequent agreement of a fundamentally inconsistent jurisdiction agreement, such as an arbitration clause in a later contract. No authority exists for this narrower proposition; but it seems possible as a matter of English contract law given the separate juridical identity of a jurisdiction agreement.
51. However, D's preferred analysis and all that is needed to reach my determination on the stay application today is that the parties varied the MRC in the terms of the MURA. This is so despite the absence of a contract endorsement bearing a GUA stamp. There is sufficient expression of consensual sovereignty in the terms of the MURA, and how it came to be executed, to establish this contractual position.
52. The fifth point concerns the interplay between standard and bespoke terms, or perhaps between what are sometimes called 'general' and 'specific' terms. C contends that the MRC was a negotiated document brimming with incorporated or replicated conditions, many being standard London market clauses; whereas the MURA is a much shorter, standard document, containing the bare minimum tailored to this reinsurance and a set of 30 standard printed T&Cs. It is said, therefore, that the later standard or general wording should not supplant or modify the earlier bespoke or specific wording. C refers by analogy to the decision of Foxton J in Generali Italia v. Pelagic Fisheries [2020] EWHC 1228 (Comm); [2020] 1 WLR 4211.
53. I derive little assistance from an analogy to the materially different contractual matrix addressed in that case. My task is to ascertain the parties' common intention in the present matrix however unusual that may be in practice. No case was cited to me which bore any close resemblance to the relevant contractual behaviour of these parties.
54. The fact that the MURA is a shorter document, and that it contains 30 standard looking T&Cs, including the New York arbitration agreement and New York choice of law, does not contra-indicate the parties' intention that those two clauses should replace their equivalents contained in the MRC. C's submission on this point may well form a plank of the first point addressed above, namely inherent improbability.
55. My conclusion on this is as follows. However unusual or inherently improbable it was for contracting parties to swap out fundamental terms of a concluded contract by another contract within eight days, that is what they did here, and I am satisfied that was their objective common intention.
56. I reach this conclusion without resort to the so-called entire agreement provision in clause 26 of the MURA. That is a New York law-governed provision. By definition, it depends for its own existence and force on being part of a contract which has come into being so as to contain the parties' reinsurance contract. To resort to that contractual language in order to show that that key provisions in the MRC or that contract as a whole were replaced by the MURA feels somewhat circular to my mind. I can see, however, that the existence of such "entire agreement" language in the MURA might be said to fortify the conclusion I have already reached as to the contractual effect of the parties' execution of the MURA.
57. In summary, I conclude that the parties agreed to refer the relevant dispute or disputes to arbitration in New York, pursuant to the terms of the MURA. On this basis the court must stay the present proceedings under section 9 of the 1996 Act.
58. This makes it unnecessary to consider D's alternative contractual analysis which involves an attempted reconciliation of the English jurisdiction clause in the MRC and the New York arbitration agreement in the MURA, so that they can operate or 'dovetail' together in relation to any given disputed matter. As indicated during the hearing, I find this an extremely challenging submission, given the language of each of the forum selection clauses.
59. The fact that the earlier clause confers "exclusive jurisdiction ... on all matters relating to this insurance" makes it exceedingly awkward - to say the least - to read it as conferring some form of non-exclusive quasi-supervisory jurisdiction over a New York arbitral process. This involves reading down the exclusive English jurisdiction clause as a mere submission to this court's jurisdiction for the purposes of granting ancillary or injunctive relief in aid of a New York arbitration, or perhaps enforcement of any award made in such arbitration. That is not what the clause says. It is, if anything, the opposite of what the clause says.
60. The courts have at times been required to reconcile so-called 'pathological' forum selection choices in a single contract. They have at other times been required to ascertain whether a particular dispute falls within one or more different forum selection clauses agreed by the same or overlapping parties in more than one contract relating to a commercial relationship or set of related transactions, especially in the financing context. Here, however, the parties agreed two successive contracts, covering precisely the same legal relationship. There is not room for both. Even if there somehow were, there is no way of reconciling the two forum selection clauses so as to give the English jurisdiction clause some independent role alongside the New York arbitration agreement. The latter supplants the former altogether, in my judgment. As it happens, D does not need to go this far. I say no more about the alternative analysis based upon reconciliation.

ASI: Remedial Discretion

61. In light of my conclusion on the contractual issue, there is no basis for a court to grant any injunctive relief in respect of the New York arbitration. D has not acted and is not acting unlawfully by commencing and pursuing that arbitral reference. Nor can it be said that D is otherwise acting vexatiously or abusively or unconscionably.
62. If it were otherwise, a separate issue would have arisen as to whether it was "just and convenient" to grant such injunctive relief pursuant to section 37 of the Senior Courts Act 1981. An ASI or anti-arbitration injunction is a discretionary equitable remedy.
63. It is well established that an anti-suit applicant must apply "promptly and before the foreign proceedings are too far advanced": see The Angelic Grace [1995] 1 Lloyd's Rep 427 at page 96 col.2. This formulation involves disjunctive limbs. Delay alone may bar the grant of anti-suit relief even where the first-seised legal process is not far, let alone too far, advanced.
64. Delay can bar the grant of anti-suit relief in a way that operates more acutely or exactingly than injunctive relief in general. The rationale for this is not entirely clear in the authorities but is certainly infused with considerations of international judicial comity: see Ecobank Transnational Inc. v. Tanoh [2015] EWCA Civ 1309; [2016] 1 WLR 2231.
65. Where a defendant to a foreign legal process waits too long to seek a coercive remedy from the English court, this can bar such relief, even though no prejudice is suffered by their adversarial counterparty. In waiting too long, the injunction claimant is more likely to have allowed the foreign court system to commit its own finite resources in entertaining the relevant claim before it. Other users of that court system, including local citizens, may have been made to wait longer for a hearing as a result of the impact of the relevant case on that court's capacity, even if it has not progressed far in the meantime.
66. Delay can bar ASI relief even when there is no submission to the foreign court's process as a matter of private international law analysis. Although no representation akin to estoppel has been made, and hence no legitimate expectation engendered as to non-resort to any coercive remedy in another court, the simple fact of this delay and its pragmatic impact may suffice to bar that coercive relief. The local court system is obliged to deal with the matter before it, thus expending finite public resources. Delay in seeking this unique form of coercive relief can waste foreign judicial and administrative resources and keep others out of justice in that foreign court system. It may also in some instances be misunderstood by judges and court staff, given the potentially terminal effect such an injunction can have on an action pending in their court system. The English court is sensitive to these potential consequences.
67. It is for this reason that the law requires an injunction claimant to provide a good reason to justify or excuse any objective delay in seeking anti-suit relief following the discovery or discoverability of the violation of their jurisdictional rights. A solid explanation for proven or admitted delay can, therefore, overcome the perceived detriment arising from such surplus engagement of the foreign court process at the public expense of the first-seised state. This, in essence, is what the injunction claimant has to justify. It may in some cases motivate, if not compel, them to waive privilege in their local legal advice and any advice they receive as to why they should hold off seeking an ASI from the English court. It all depends on the circumstances.
68. What is not clear on the authorities is whether or how this calculus applies to private arbitration. There is no engagement of a foreign court system in this context, hence no consumption of that country's finite public resources during such seisin, or (by definition) any risk of offence or affront or disrespect to any judge in such country. Arbitrators are, of course, entitled to courtesy and respect by a foreign court in the discharge of their private contractual mandate and functionally judicial responsibilities; but the position is not entirely equivalent. There is no use of public resources, even though the local court system is allocated with curial responsibility for the arbitral process and any awards.
69. With those observations in mind I turn to the present position. The ASI application was made on 3rd November, one day shy of six months after D commenced the New York arbitration. Viewed objectively and in isolation, this is not prompt.
70. C's explanation for waiting until then is tied into what was going on in the arbitration itself. C waited until the three-person tribunal was constituted in late August to request a stay pending the outcome of D's stay application in this court, by then listed for 13th December. I have described above what then occurred in mid-September.
71. C then waited until after what was called an "organizational meeting" in the New York arbitration on 17th October. The outcome of that hearing was the timetable to or fixing of the final merits hearing in the arbitration, including a document production or "fact discovery" phase through to next Easter, and then a final merits hearing next October.
72. It is not clear how it could have taken C's legal team another 17 days after that arbitral hearing to issue the ASI application. It did not need that long. They took the view that they needed to issue the ASI application in good time to allow the so-called 'heavy application' directions in the Commercial Court Guide to run their course ahead of the listing of the present hearing. They worked back from a date fixed in this court. By this stage in the process, that was a reasonable course to have taken. All eyes were by then on the hearing here, listed for 13th December. The ASI application did not affect the time estimate in a material way so they could be heard together.
73. Working backwards further in the timeline, I can also appreciate why, having sought a stay from the arbitrators, C then awaited their ruling and the clarification of the basis for it in September. It is not clear, however, why C needed to hold off preparation of the ASI application once the tribunal had clarified the basis of its refusal of the stay on 19th September. Waiting and preparing for the arbitral case management hearing in mid-October is not justified in the circumstances. The tribunal had refused to stay its process and was moving towards a hearing that would determine the future course of the arbitral reference. That should have provided a clear impetus to seek anti-suit relief, even if it could not have been determined on an urgent basis prior to 13th December listing in this court.
74. C's delay starts much earlier in the timeline, however. There was no good reason to hold off seeking anti-suit relief whilst the tribunal was being put together in June to August. Nor was there any reason to eschew this coercive option whilst the tribunal was considering C's stay request. This could all have been done in parallel. There is no evidence to infer that seeking ASI relief at that early stage would have 'rocked the boat' or thrown the tribunal offside at the time of its formation.
75. The delay between the commencement of the arbitration on 4th May and at latest D's stay application in this court on 24th May, and issuing the ASI application on 3rd November, has not been adequately justified on an objective basis in my judgment. An anti-suit applicant is required to move much sooner. Whilst the rationale for this may not be as grounded in comity considerations where private arbitration is involved, as discussed above, waiting six months or so to seek coercive relief when it could have been sought at the outset, or in parallel with other solutions or steps within the arbitral process, is inexcusably long.
76. I reject the contention that it was reasonable for C to wait this long because of the intensity or bipolarity of the contractual dispute which underpins the jurisdictional basis for seeking such relief. The basis for seeking such relief is that C's contractual right, i.e. not to have the relevant dispute determined other than in the English court, was violated on 4th May. In that moment, C acquired an equity which it could seek to convert into the equitable remedy of an ASI in this court. C knew that that was the case, not least because it had the day before asserted such right by commencing these proceedings relating to the very same avoidance dispute. It appointed its arbitrator in New York under protest as to arbitral jurisdiction, and it sought a stay from the arbitrators referable to the outcome of its counterparty's NYC-stay application here.
77. There is no requirement for an anti-suit applicant to know that its rights have been violated as an indisputable or even clearly provable legal fact. An apparent violation triggers time. If there is a sustainable - even if ultimately unsuccessful - basis for alleging violation of a jurisdictional right, time runs once it is discovered or reasonably discoverable. That occurred here on 4th May when D commenced arbitration.
78. If I had been required to determine the ASI application on its merits, having found in favour of C on the substantive contractual issue, I would have been minded to refuse to grant relief in the exercise of my remedial discretion under section 37 of the 1981 Act. The relief was sought without adequate justification for having applied six months after the underlying remedial equity was generated on 4th May.

Permission to Appeal

79. I recognise that some issues addressed in this judgment are novel, in terms of absence of precedent and perhaps not so obvious in terms of final resolution. No authority was identified that resembled the contractual matrix presented in this instance. As noted above, I am not aware of an authority dealing with implied recission of a separable jurisdiction agreement. The law on delay in seeking an ASI relief, referable to its rationales and whether they apply with the same force in the case of private arbitration, may also benefit from fresh appellate consideration. For these reasons, I am minded to grant permission to appeal, if sought by C.
80. This concludes my judgment. It was handed down remotely between 1030am and 1113am on Friday 15^th December 2023. A corrected version of the approved judgment was issued on Monday 18^th December 2023.