Royal & Sun Alliance Insurance Plc & Ors v Rolls-Royce Plc (Rev 1)

This is an application by the defendant Rolls-Royce plc to stay proceedings brought against the company in the English court by claimant insurers, either generally, or at least pending determination by the US court of its jurisdiction in proceedings brought by Rolls-Royce against the insurers there. In substance, the issue between the parties is whether an insurance claim by Rolls-Royce in connection with the manufacture by the company or its subsidiaries of a marine propulsion system known as a Mermaid Pod should be litigated in the English court or in the United States District Court for the Southern District of Florida (Miami Division). The claimant insurers argue that the proceedings should take place in the English Court, because both Rolls Royce and its primary insurers are (predominantly) English companies and the contracts of insurance in issue are expressly governed by English law. Rolls-Royce argues that its claims under the insurance policies arise from legal actions brought against it (mainly) in the Florida courts, and that Miami is the most appropriate forum for the dispute. In the English court, Rolls-Royce must however also surmount the objection taken by the insurers that since Rolls-Royce is an English company, the English court’s jurisdiction is mandatory under Council Regulation (EC) No 44/2001 of 22 December 2000 on jurisdiction and the recognition and enforcement of judgments in civil and commercial matters (“the Regulation”). This, it is submitted by the insurers, follows from the decision of the Court of Justice of the European Communities (“ECJ”) in Case C-281/02 Owusu v. Jackson [2005] QB 801. There is, the insurers say, no discretion to stay on the grounds of parallel proceedings in a non-Member State (that is to say a state like the United States that is not a member of the European Union).

The dispute concerns Rolls-Royce’s claim for an indemnity under a policy of insurance (“the Ledger Policy”) which covers it and its subsidiary and associated companies in respect of, inter alia, General Liability risks, Construction All Risks and Professional Indemnity risks for a period from 1 May 1997 to 30 April 2002. The insurers under the Ledger Policy are referred to in this judgment as the primary insurers, and are the claimants in action 2009 Folio No. 822. The primary insurers participate on all coverage sections subject to their limits of indemnity. Rolls-Royce and its subsidiary and associated companies are also insured under three excess layer insurance policies which covered the years 1 May 2000 to 30 April 2001 and 1 May 2001 to 30 April 2002 and provided additional layers of cover in respect of the General Liability section of the Ledger Policy. The second and third excess layers exclude coverage for certain financial losses for occurrences in respect of which claims are brought in the United States. The excess insurers are the claimants in the actions 2008 Folio Nos. 828 and 849, and together the insurers make common cause between them in opposing the application for a stay.

The claim arises in this way. The Mermaid Pod is a self-contained marine propulsion device containing principally a propeller shaft and an electric motor which drives the propeller. These devices are attached to the bottom of ships, the electrical power for the motor being provided by the ship’s engines. The Mermaid Pod can be fitted so as to rotate allowing it to steer the ship as well as to propel it. Rolls-Royce explains that the device was originally developed by a consortium made up of a Swedish subsidiary of (at that time) the Vickers Group called Kamewa AB (now “Rolls-Royce AB”) and a French company in the Alstom group, Alstom Power Conversion SA (now called Converteam SAS). Rolls-Royce purchased the Vickers Group of companies in November 1999.

From about 1998, the Mermaid Pod was fitted to a number of large cruise ships by a French shipyard, Chantiers de l’Atlantique. As Rolls-Royce puts it, shortly after the first installation, operational issues began to emerge, the focus being on the operation of various bearings contained within the device. As a result, vessels had to be dry-docked for evaluation and repair. The upshot of the problems has been litigation against Rolls-Royce. In summary, this was as follows:

Rolls-Royce’s case is that the claims which it (and its subsidiary/associated companies) make on the insurance policies issued by the claimants relate to the amounts payable in these actions and the costs of the actions as well as costs incurred in performing required repair work on the Mermaid Pods on the RCCL vessels and also amounts incurred by the Rolls-Royce companies in the comprehensive engineering investigation successfully undertaken to rectify the operational issues and to mitigate against additional losses.

The question of notification is a matter of contention between the parties, but in summary, in March and April 2002, Rolls-Royce’s brokers, Marsh UK Ltd, notified the primary lead insurer (Royal & Sun Alliance Insurance plc) of incidents affecting the Mermaid Pods, identifying seven vessels, and notifying claims “in respect of these marine vessels and any others not yet identified”. Following the original notifications, Royal & Sun Alliance engaged in a process of investigation. The evidence is to the effect that by early 2009, the insurers had indicated that they would not meet the claim. It seems that specific notification in relation to the QM2 claim was not made until September 2009, and the insurers say that there is plainly no coverage in respect of that claim. I will need to come to the arguments that arise on the claim in more detail later, but for the present it is sufficient to say that the principal issues that are likely to arise between the parties have been identified by Mr Robert Miles QC for Rolls-Royce as falling into three groups being, first issues as to the interpretation of the contracts of insurance, second factual and technical issues, and third notification issues, and this is broadly common ground, though as will be seen the emphasis that the parties place on each is different.

On 8 April 2009, Rolls-Royce and other companies in the group filed a complaint in Broward County Court, Florida, seeking declaratory relief that the various insurance policies are responsive and that they are entitled to an indemnity. These other companies in the group had each been sued in at least one of the various Floridian proceeding brought by the shipping companies, and are Rolls-Royce AB, which as mentioned is a Swedish company, Rolls-Royce Commercial Marine, Inc, a company incorporated in Delaware and operating in Louisiana and Florida, and Rolls-Royce North America Holdings, Inc, a company incorporated in Delaware and operating in Virginia. These three companies are all subsidiary or associated companies of Rolls-Royce, and are, Rolls-Royce says (and I do not think it is in dispute) covered by the Ledger Policy and the excess layer policies. The originally-named defendants in the Florida proceedings were the insurance companies named in the policies as those providing the relevant insurance cover, but since the policies were written a number of the insurers have changed their names and/or transferred their business to other group companies, and the complaint was in due course amended to reflect the current position. The proceedings were served on the insurers in late July 2009. They were removed from the State Court to the US District Court on diversity grounds on 29 August 2009.

Prior to being served, the insurers learned about the Florida proceedings, and began their own actions in the Commercial Court in London. The claims were issued on 18 June, 19 June and 25 June 2009 respectively, and service was effected on Rolls-Royce (which is the only defendant) on 22, 25 and 29 June 2009 respectively. By the claims, the insurers assert that they are not liable to indemnify the defendant or its subsidiary or associated companies. Rolls-Royce acknowledged service on 3rd July 2009. The issues in the two sets of proceedings are, therefore, substantially the same.

Thereafter the parties began negotiations, and both sets of proceedings were stayed by consent—neither of them has got very far. After negotiations broke down, the insurers challenged the jurisdiction of the District Court and a timetable has been set down by that court for discovery on jurisdictional issues, and the production of briefs on whether the court has personal jurisdiction over the insurers and, if so, whether the proceedings should be stayed on forum non conveniens grounds. The briefing will be complete by November 2010, no date for the hearing having yet been set. On 14 May 2010, Rolls-Royce filed a Second Amended Complaint so as to correctly name the insurer defendants. Also on 14 May 2010, Rolls-Royce applied (unsuccessfully) to the District Court for an anti-suit injunction against the insurers to prevent the continuance of the English proceedings. Meanwhile, on 12 April 2010 Rolls-Royce brought its application in this court to stay the English proceedings either generally, or at least pending determination by the US District Court of its jurisdiction in the proceedings before it. This is the application that is before the court for decision. A considerable amount of material was filed in support, but in the event the parties were able to restrict the material to an agreed reading list, for which I am grateful.

The claims brought in England were served on Rolls-Royce at its registered office in England. The English court therefore has personal jurisdiction, and also subject matter jurisdiction, namely an English law insurance policy written in London. The question is whether this court is entitled to, and if so whether it should, decline to entertain the claims on the grounds that a claim under the policy is already proceeding in Florida. It is common ground that the burden of establishing this is on Rolls-Royce. In that regard, the primary insurers point out that Rolls-Royce’s application notices dated 12 April 2010 were based only on the ground that the “Florida Proceedings and these proceedings are between the same parties and are in respect of the same cause of action and the Florida Court was first seised of the dispute between the parties in respect of that cause of action”. In other words, it was limited to an assertion of lis pendens. In fact, the question of which proceedings came first turns out to be (as is so often the case) a contentious one. In any case, the argument that Rolls-Royce has advanced at the hearing has been more broadly based and centres on forum non conveniens grounds.

As I indicated, the insurers maintain that as a matter of law, this court has no discretion to stay the English proceedings on the grounds of parallel proceedings in a non-Member State. If they are correct, this is a complete answer to Rolls-Royce’s application regardless of the merits. The argument for the primary insurers, supported by the excess insurers, is as follows. The insurers’ claims against Rolls-Royce are governed by Council Regulation (EC) No 44/2001 because Rolls-Royce is an English-domiciled company. The Regulation sets out (at Section 3) a number of special jurisdictional rules applicable to matters relating to insurance. In this case, the relevant provision is Article 12(1), which stipulates that: “Without prejudice to Article 11(3), an insurer may bring proceedings only in the courts of the Member State in which the defendant [i.e. Rolls-Royce] is domiciled, irrespective of whether he is the policyholder, the insured or a beneficiary”. The purpose of that rule is to protect the insured who is perceived to be the weaker party in the commercial relationship and, for that reason, it is applied inflexibly. As Rolls-Royce is an English-domiciled company, this court is therefore the only Member State court which has jurisdiction over the insurers’ claims against it. In Owusu v Jackson (ibid), the ECJ held that it was mandatory for the English court to exercise the jurisdiction afforded to it by Article 2 (which was the basis for asserting English jurisdiction in that case) unless the case fell within one of the express exceptions enumerated in the predecessor to the Regulation, namely the 1968 Brussels Convention (which it did not). It was accordingly not open to the English court to decline to exercise its jurisdiction on forum non conveniens grounds. There is no difference in principle it is said between jurisdiction founded on Article 2 and jurisdiction founded on Article 12(1). Accordingly, in the light of Owusu, it is not simply that Article 12(1) confers jurisdiction on the English courts over the claim against Rolls-Royce: it is also mandatory for the English court to exercise that jurisdiction, for all the reasons explained by the ECJ.

Rolls-Royce responds that Owusu only decides that, where Article 2 of the Regulation confers personal jurisdiction on a court of a Member State by reason of the defendant’s domicile in that Member State, the court cannot refuse to hear the case because there is a more appropriate forum abroad. It does not however, provide an answer to the question in this particular case because in Owusu the ECJ declined to answer the question whether the actual existence of proceedings in a non-Member State provided a basis for a stay (see paragraphs 47-52 of the judgment). The ECJ was concerned in that case with the position where there were proceedings in England and no parallel proceedings elsewhere. The reasoning which underpins the decision in Owusu does not translate across to a case where the relevant defendant has already commenced his own proceedings in a non-Member State. The absence of an ability in a court of a Member State to stay proceedings before it in favour of earlier proceedings in a non-Member State gives rise to the obvious risk of parallel proceedings and ultimately of competing or inconsistent judgments. Article 12 of the Regulation is no bar, Rolls-Royce submits, the measure being one which provides protection for a defendant as opposed to dealing with issues of the appropriate forum. Even within the Member States, the rule is not all encompassing. Article 12(2) provides that an insurer can bring a counterclaim in a claim pending in another Member State. Such a counterclaim is available in the Florida Proceedings

If Rolls-Royce is correct that a stay is available, the question arises as to the principles on which the court should decide whether or not to grant a stay. It submits that in cases where there are earlier proceedings in a non-Member State, Article 27 of the Regulation can be applied reflexively, in other words by analogy. Article 27 provides that where proceedings involving the same cause of action and between the same parties are brought in the courts of different Member States, any court other than the court first seised shall of its own motion stay its proceedings until such time as the jurisdiction of the court first seised is established, and where it is established, shall decline jurisdiction in favour of that court. (Related actions are dealt with in Article 28 and are not relevant in the present application.) As is well known, the question of how to determine which court is “first seised” led to considerable argument under the 1968 Brussels Convention. Recital (15) of the Regulation says that there must be a clear and effective mechanism for obviating problems flowing from national differences as to the determination of the time when a case is regarded as pending, and that “For the purposes of this Regulation that time should be defined autonomously”. Article 30 of the Regulation accordingly provides (so far as presently relevant) that, for the purposes of Section 9 (which deals with lis pendens), a court shall be deemed to be seised at the time when the document instituting the proceedings is lodged with the court. Applying these provisions reflexively, Rolls-Royce contends that the Court must look at the time of issue of the proceedings rather than service, and that in the present case this means that the Florida court was first seised.

The insurers respond that the regime in Articles 27 to 30 is designed to make provision for a rigid system of priority as between EU Member State courts. Giving reflexive effect to Articles 27 to 30 does not (and cannot) involve applying those Articles in a case involving a non-Member State court because, by their terms, the Articles govern the position only as between “the courts of different Member States”. Because there is no system of uniformity and reciprocity with non-EC Member State courts, the idea of a rigid priority rule would be contrary to principle and impracticable. The insurers submit that the closest relevant analogy to seisin in Florida State law is the rule that between two State Courts, or between a Florida and a foreign court, priority is accorded on the basis of which proceedings are first served, not first filed. Although the Florida proceedings were filed in April 2009, they were not served until the end of July, by which time the English proceedings had been both issued and served. Accordingly, it is said, the English proceedings came first.

In fact, it is to be noted that Rolls-Royce does not argue for a rigid priority rule, accepting that, on the premise that the previous stages in its argument are upheld, the granting of a stay is a matter of discretion to be exercised in accordance with the principles derived from the English forum non conveniens cases, the leading case being Spiliada Maritime Corp v Cansulex Ltd [1987] AC 460. It submits that when taken as a whole, Florida is clearly the appropriate and available forum for the resolution of the dispute between Rolls-Royce, its subsidiaries and the insurers. The insurers on the other hand submit that the English court is overwhelmingly the more appropriate forum. That is the main factual area of dispute, though (understandably) much of the oral argument focused on the Owusu point.

It is clear that the full ramifications of the Owusu decision have yet to be worked out. The impact appears to arise particularly as regards common law Member States like England, which recognise the doctrine of forum non conveniens. It was held in Winnetka Trading Corp v Julius Baer International Ltd [2009] Bus L.R. 109 by Norris J that Owusu did not prevent the stay of proceedings in England where there is a jurisdiction clause in favour of a non-Member State. There is no exclusive jurisdiction clause in the insurance policy in the present case. Though judgment was delivered by the ECJ in Owusu on 1 March 2005, first instance decisions on the question at issue on this application go both ways, and the question appears to remain undecided at the appellate level. In Catalyst Investment Group Ltd v Lewinsohn [2010] 2 WLR 839 Barling J held that Article 27 of the Regulation could not be applied reflexively and that there was no jurisdiction to stay proceedings in a Member State in favour of prior proceedings in a non-Member State. A contrary conclusion was reached in JKN v JCN [2010] EWHC 843 (Fam) by Deputy Judge Theis QC. In Goshawk Dedicated Ltd v. Life Receivables Ireland Ltd [2008] IEHC 90 Clarke J decided that a court in Ireland retains and must exercise a jurisdiction conferred upon it by Article 2, notwithstanding the fact that there may be proceedings in a non-Member State which are first in time. On appeal, this question was referred to the ECJ ([2009] IESC 7), but the case was resolved prior to the reference being heard.

In Lucasfilm Ltd v Ainsworth [2009] EWCA Civ 1328 on the other hand, Jacob LJ giving the judgment of the Court (with Rix and Patten LJJ) said of Catalyst:

“134.

… The question there was whether the court could stay proceedings in a case where the same point was being litigated between the same parties in the courts of a third country. [Barling J] held that Owusu prevented that, essentially because the lis pendens rule is to some extent a facet of forum non conveniens. We do not have decide whether that was correct, though we note that, if he is right, there is this oddity: that there is a clear lis pendens rule, with associated court first seised rule, for parallel cases within the EU but none for parallel cases where one is running within an EU Member State and one without. …”

The preponderance of academic opinion in this country is in favour of the existence of a discretion to stay (Dicey, Morris & Collins, The Conflict of Laws, 14^th edn, at 12-021, Briggs and Rees, Civil Jurisdiction and Judgments, 5^th edn, at 2.260), though views are not unanimous (Cheshire, North & Fawcett, Private International Law, 14^th edn, at p. 327). There is a full discussion of the issues in Fentiman, International Commercial Litigation at chapter 11.

The question, as Barling J pointed out in Catalyst, is not whether a Member State court should have power to stay where proceedings are extant in the courts of a non-Member State, but whether the reasoning in Owusu permits such a stay. If not, the oddity identified by Jacob LJ in Lucasfilm is demonstrated in this case as well as any. Though the parties have for the purposes of the argument highlighted connections with the respective jurisdictions, the dispute between them is international in nature, and could feasibly be tried in one jurisdiction or the other, but not, as a matter of plain common sense (Dicey at 12-021) in both. That is not to say that the issues are straightforward, again as this case demonstrates. Assuming there to be power to stay, issues include the basis on which it is to be determined, how to determine which proceedings come first in time, the relationship between lis pendens and forum non conveniens, and, if a stay is discretionary, the principles upon which such a discretion is to be exercised. As to the necessity for a discretion if there is a power at all, the excess insurers point out (rightly in my view) that the mechanistic approach of Article 27 functions in the context of the Regulation because each Member State is subject to the terms of the Regulation and applies the same rules to determine whether to accept or decline jurisdiction. By contrast the courts of a non-Member State are not subject to such rules and will decline or accept a case according to their own rules of jurisdiction.

As I have said, Rolls-Royce accepts that as the final stage in its argument, it must show that the alternative forum it contends for is clearly the appropriate and available forum for the resolution of the dispute. The Owusu point is best decided where that can be shown to be the case. With deference to the excellent arguments of the parties, it appears to me to be sensible to take this issue first. If I am against Rolls-Royce on it, the parties are in agreement that the Owusu point does not arise for decision, and that I need not decide it. (A similar approach was adopted in Choudhray v Bhatter [2009] EWCA Civ 1176 and Deutche Bank AG v Sebastian Holding Inc [2009] EWHC 3069, and it is one which I think it right to follow.)

Both parties make complaint about each other. The insurers point out that Rolls-Royce began the Florida proceedings without giving them prior notice. They only learned about the proceedings by chance. Rolls-Royce in turn objects that the English proceedings are tactical, being the mirror image of the Florida proceedings. On the information that I have seen, I do not consider that in this case criticism can be levelled at either party. Rolls-Royce, for reasons which seem good to it, would like the dispute to be tried in the Florida courts. The insurers, for reasons which seem good to them, would like the dispute to be tried in the London court. Since Rolls-Royce is an English company and has been properly served here, the task of this court is to determine whether a stay of the English proceedings should be granted.

There is no dispute about the principles upon which the English court determines whether to grant such a stay. In argument, neither party went beyond citing the leading case, namely Spiliada Maritime Corp v Cansulex Ltd [1987] AC 460, where the principles are set out by Lord Goff at pp 475-478. Each relied on the basic principle, which is that “… a stay will only be granted on the ground of forum non conveniens where the court is satisfied that there is some other available forum having competent jurisdiction, which is the appropriate forum for the trial of the action, i.e. in which the case may be tried more suitably for the interest of all the parties and the ends of justice” [p. 476C]. The insurers emphasise the passage at p. 477E, to the effect that the defendant (Rolls-Royce) must show there is another available forum having competent jurisdiction which is “clearly or distinctly” more appropriate than the English forum for the trial of the action.

The factors that have been advanced by Rolls-Royce (following the sequence in the argument of Mr Miles QC) are as follows.

Rolls-Royce is the only defendant in the English proceedings, and other Rolls-Royce entities which it contends are relevant to the claim are co-plaintiffs in the Florida proceedings. It contends that this makes a practical difference because the result of the Florida proceedings might be different. For the reason given by Mr Murray Rosen QC for the primary insurers, this factor does not in my view carry any weight. The insured under the Ledger Policy is Rolls-Royce plc, albeit the cover extends to its subsidiaries and associated companies, and the claim is the same in both jurisdictions, albeit framed positively in Florida and negatively in England, namely whether the Ledger Policy is responsive or not. I do not see that the presence of other Rolls-Royce companies as plaintiffs in the Florida proceedings makes much if any practical difference.

It certainly does not, in my view, outweigh the fact that all the insurers are either registered in the United Kingdom or wrote the policies from branches in London. All the primary insurers are companies registered in the United Kingdom, with the exception of Zurich Insurance plc which is a company registered in Ireland. Its participation in the Ledger Policy was as the result of a transfer of business which took effect as of 1 January 2009. The original counterparty to the Ledger Policy was Zurich International (UK) Limited, a company registered in England. Rolls-Royce is also registered in England. By contrast, none of the Florida defendants are Florida companies. The question whether the Florida court will assume jurisdiction remains undecided. It is for that reason that Mr Miles QC recognises that a stay of the English proceedings can in any case only be temporary, since the Florida court may in the event decline jurisdiction. In so far as it may be relevant, none of the plaintiffs in the Florida proceedings are incorporated in Florida either. It may not carry much weight, but so far as it does, the location of the parties points to England as the appropriate forum rather than Florida.

The subject of the dispute is a claim on an English insurance policy, but Rolls-Royce emphasises that the issues that arise are likely to fall into different categories. As I have mentioned, it divides these into questions as regards the interpretation of the insurance contracts, factual and technical issues, and notification issues. The significance of the fact that the law of the contract is English law should not be exaggerated, Rolls-Royce submits, since the Florida court can interpret the contractual provisions just as well as the English court. As to the factual and technical issues, it says that these relate to the manufacture and operation of the Mermaid Pods and their component parts, the reasons for the operational issues which arose with the Pods, and the remedial measures taken around the world. It is submitted that these are the issues which will require the bulk of the witness evidence (both factual and expert), which will generate the disclosure, and which are likely to take up the majority of court time at trial. These matters, it is submitted, have no real connection with England. In oral submissions, Mr Miles QC developed these points, identifying as specific factual issues questions as to how the Pods were constructed and operated, the physical mechanisms that led to the damage, dry-docking episodes, so far as quantum is concerned, how much Rolls-Royce was out of pocket, the question whether the Pods were prototypes (which may take them outside cover), the question whether the various problems were caused by negligence, and the question whether Rolls-Royce was under any liability within maintenance periods. As regards notification, the issues are likely to be whether it was in time or out of time, what Rolls-Royce knew at any particular time, and what was the scope of the notifications in question. Such issues, it was submitted, will be highly sensitive to the factual and technical evidence.

The primary insurers on the other hand submit that based on the correspondence between the parties, the following are likely to be key issues as between Rolls-Royce and the primary insurers: (a) issues of late notification (in that only Royal & Sun Alliance was notified during the policy period), (b) issues as to the meaning and effect of the notifications made to Royal & Sun Alliance, and in particular (i) whether those notifications can include within their scope incidents of which Rolls-Royce were already aware but which were not identified specifically in the notifications, (ii) whether those notifications can include within their scope incidents involving bearings which had not yet been installed at the time the notifications were made (a point said to be relevant to fourteen of the twenty five incidents), (iii) whether those notifications, and in particular the blanket notification of 16 April 2002, can amount to a proper notification of circumstances in respect of each of the remaining incidents, (c) technical issues, specifically whether the phenomenon of sub-surface cracking displayed by certain bearings was caused by the use of standard bearings that were not in themselves defective but unfit for the unusually heavy load profiles to which they were subject or by the use of defective steel in the manufacture of the inner raceways, (d) issues as to the meaning and effect of Occurrence 2 under the General Liability coverage section (and similar wording in the CAR coverage section), (e) issues as to the meaning and effect of Special Clause 6 of the General Liability section (and similar wording in the CAR coverage section), (f) issues as to the wording of the financial loss cover provided under Special Condition 8 of the General Liability section, (g) issues as to the meaning and effect of the prototype exclusion applicable to the CAR coverage section, (h) issues as to the availability of coverage under all coverage sections in respect of incidents where bearings were replaced as a precautionary measure, rather than because they were damaged, (i) factual issues specific to particular incidents (by way of example Incident 1 is said to have involved no bearing replacements at all and incident 14 is said to be attributable to a previous, unrelated grounding), (j) issues as to whether there is any coverage at all in relation to the Carnival Proceedings in light of the fact that those proceedings involve a different design of pod which was not installed until some time after the expiry of the policy period. Additional issues, it is submitted, are likely to arise as between Rolls-Royce and excess insurers. Moreover, given that the issues are far from crystallised, it would be unsafe to assume that further issues will not emerge in due course.

It is to be noted that the issues have not yet been identified in the pleadings in either jurisdiction, because of the early stages of the proceedings—see paragraphs 64-68 of the Second Amended Complaint in the Florida proceedings which set out Rolls-Royce’s case as to repudiation by the insurers of their policy obligations. So far as formal pleadings are concerned, the allegations in the English proceedings have not got beyond the Claim Form. To form a view as to the issues which are likely to arise in the litigation is not therefore entirely straightforward. I would accept Rolls-Royce’s submission to the extent that technical and factual issues are clearly likely to be important in deciding the claim. Such issues will have an impact on interpretational issues, whether as to notice, or as to the policy. Nevertheless, as the excess insurers rightly put it, the policy was written in the London market, through an English broker. The claims under the policies have been presented in London. Discussions relating to the policy dispute have been taking place in London (between Rolls-Royce, brokers and insurers, and later insurers’ lawyers) since about 2002. The underlying failures of the Mermaid Pods (which were manufactured in France and Sweden) took place all over the world. That latter point (which is not in dispute) is the one which relates most clearly to the technical and factual issues identified by Mr Miles QC in oral argument.

So far as issues are concerned, in my judgment, the dispute has no real connection with Florida, except that the claims by the shipping companies though not the shipyard were brought in the Florida Courts. (The claims in respect of vessels also included two cruise-liners operated by a company called Festival Crociere SpA which does not appear to have resulted in litigation.) As I shall make clear, I accept Rolls-Royce’s submission that the fact that claims were brought in Florida is a significant factor pointing to Florida to be weighed in the balance. I also accept that the domicile of Rolls-Royce in England is unlikely of itself to matter much in this case. However, even allowing for the diverse nature of the issues that the court is likely to have to resolve, it is plain in my view that whilst technical and factual issues will have a world-wide ambit, the matters relating specifically to the insurance arise in England. Considerable weight must be given to the fact that, as the primary insurers put it, at its most basic level, this a dispute between an English insured and (predominantly) English insurers under an insurance policy placed by London market brokers.

The Ledger Policy is governed by English law, but Rolls-Royce submits that the significance of this should not be exaggerated. There is no evidence that the Florida Court would adopt a different approach to interpretation than the English Court. Questions of interpretation, it is submitted, will be heavily informed by the factual context. The insurers, on the other hand, submit that both issues as to notification, and issues as to the construction of various provisions of the Ledger Policy, raise potentially complex issues of English law, which are better resolved in the English Commercial Court with its specific experience in such matters. In addition, the excess insurers submit (with some force in my view) that there may also be questions, if liability is established, as to how the primary and excess layers should be stacked, given the different scope of cover of the different excess layers (in particular in relation to financial loss cover).

Rolls-Royce accepts that the fact that the contract of insurance is governed by English law is a factor pointing towards England as the most appropriate forum. Whilst I agree that this factor should not be exaggerated, nevertheless its importance in the case of insurance and reinsurance has been recognised in the case law: see E.I. Du Pont de Nemours v Agnew [1987] 2 Lloyd’s Rep 585, where a stay was refused notwithstanding parallel proceedings which had been issued in Illinois on the main ground that English law applied to the policy (see p. 594, Bingham LJ), CGUInternational Insurance plc v Szabo [2002] Lloyd’s Rep IR 196 at [52], Toulson J), and Tryg Baltica International (UK) Ltd v Boston Compania de Seguros SA [2005] Lloyd’s IR 40 at [48] to [50], Cooke J. Furthermore, I have set out in paragraph 26 above the issues identified by the primary insurers where governing law may be relevant to the disposition of the case. It is not suggested that these issues may not arise, and on the face of it, they concern what could be points of substance of English law, going well beyond a straightforward reading of the provisions in the contract. I do consider that this factor is a significant one pointing to England as the appropriate forum.

In the course of argument, Rolls-Royce put forward as its key point the fact that it has been involved in legal proceedings about these Mermaid Pods for some seven years in Florida, and that in the course of these claims, its lawyers have accumulated a great deal of knowledge and experience in connection with the dispute. By way of example, I was taken to the parties’ respective exhibit lists in the RCCL litigation, both dated 22 December 2009. Mr Miles QC submits that anyone who has undertaken complex litigation will know, as he puts it, the step change that exists between advising and litigating. There is a steep learning curve involved, and if proceedings are to proceed in England rather than Florida, Rolls-Royce will have to start all over again with a new legal team. In this regard, it stresses the importance of the technical and factual background to the case, and the fact that this is liability insurance, and that the focus will be accordingly be upon the claims made against Rolls-Royce, which, with the exception of the proceedings in Paris, have been pursued in Florida. In an international case, Rolls-Royce submits, this is the one clear factor that tips the balance.

On their part, the insurers point out that their London lawyers have also been fully engaged in the claim since 2003. That was not just a matter, it is said, of answering letters. However they rely in particular on the fact that Rolls-Royce’s current lawyers were only involved in two of the three sets of Floridian proceedings (RCCL and Carnival) and then only from 2008 onwards—it is common ground that the company was represented before then by a different Florida law firm. In any case, it is submitted that Rolls-Royce must establish that it was liable to the underlying claimants on a basis which also attracts indemnity under the Ledger Policy. It is not simply a matter of analysing the claims that were made against the company. Further, it is submitted that the Carnival proceedings (which are coming on for trial in Florida shortly) must be left out of account because any loss in respect of Carnival is not on any view covered by the Ledger Policy.

In Spiliada at pages 485-6, Lord Goff refers to the legitimate weight that can be given to the accumulated experience of lawyers who have been acting for one of the parties in associated proceedings (the so called “Cambridgeshire factor” after the name of the associated proceedings in that case). He says that, “anyone who has been involved, as counsel, in very heavy litigation of this kind, with a number of experts on both sides and difficult scientific questions involved, knows only too well what the learning curve is like; how much information and knowledge has to be, and is, absorbed, not only by the lawyers but really by the whole team, including both lawyers and experts, as they learn about the interrelation of law, fact and scientific knowledge, having regard to the contentions advanced by both sides in the case, and identify in their minds the crucial matters on which attention has to be focused, why these are the crucial matters, and how they are to be assessed” (p.485G). In my view, Rolls-Royce is right to say that this is a factor pointing towards Florida as the appropriate forum in this case.

However, the weight to be accorded to the “Cambridgeshire factor” will plainly vary from case to case. In Spiliada, both actions concerned damage alleged to have been caused to vessels by cargos of sulphur, and the overlap has to be seen in the light of the equivalence of issues. The present case however involves, on the one hand, proceedings against Rolls-Royce concerning complaints about the Mermaid Pods, and on the other, a claim by Rolls-Royce under its insurance. I accept that much of its Florida lawyers’ knowledge may be relevant, but the context of the insurance dispute is different. Furthermore, the point made by the insurers on the particular facts appears to me to be important. Although the litigation in Florida has been proceeding for some time, Rolls-Royce’s present lawyers have only been involved since 2008. This diminishes the weight to be given to this factor. From the evidence, it seems that Mr Larry Stumpf in particular has had the conduct of the matter on behalf of Rolls-Royce since that time. It seems to me that there will inevitably be some duplication between lawyers whether the proceedings take place in England or in Florida. Mr Stumpf’s expertise remains available to Rolls-Royce even if the proceedings take place in England. I would conclude that this is a factor on which Rolls-Royce is right to rely as pointing to Florida as the appropriate forum. However, it does not appear to me to carry decisive weight in this case.

Rolls-Royce submits that the principal witnesses are going to be those who will deal with the underlying issues about the Mermaid Pods. These, it submits, have limited connections with England, and will include experts in relation to the engineering and production of the Mermaid Pods and the remedial works carried out. Individuals from the cruise ship companies will also be required, it is submitted, and these are principally domiciled in Florida. Their evidence would be relevant to the nature of the claims made against the Rolls-Royce companies which are the basis for the claims on the insurance policies and to the issues which affected the Mermaid Pods. In so far as Rolls-Royce seeks to adduce evidence from these witnesses, it may have to have recourse to the procedures for obtaining evidence under compulsion. Whilst it is possible that this can be done from England, it will be easier, cheaper and quicker, if the starting point is a court in the US seeking to compel a witness in the US. It is accepted that individuals from the Rolls-Royce companies including in-house engineers will be witnesses, but they are principally domiciled in Sweden, reflecting the fact that the most relevant Rolls-Royce company is Rolls-Royce AB. Further, to the extent that there are disputes about the underlying Florida proceedings and whether any settlements reached are reasonable, the relevant lawyers are in Florida, the settlement agreements are governed by Florida law and the Florida Court is best placed to understand and resolve any such disputes. By contrast, it is submitted that whilst the insurers have identified a wide-range of individuals who they say are resident in the UK and would be witnesses in the English proceedings, in fact, the majority of these people are either representatives of the insurers who were involved in the original entering into of the Ledger Policy or were involved in the notification process after the Mermaid Pod incidents started. It is unclear why individuals in either class would or could provide any relevant evidence, since there is no suggestion of rectification and they would not be able to give admissible evidence about their intentions.

The insurers submit that despite Rolls-Royce’s assertions, it is plain that most of the relevant witnesses and documentary evidence is in (or closest to) England and could easily be brought to this jurisdiction. It is likely, it is submitted, that factual matrix evidence will be required as the Ledger Policy is not in standard form. There is, moreover, likely to be an issue as to the terms of the Ledger Policy itself given that the wording of the Financial Loss extension relied upon by Mr Stumpf appears to differ in material respects from the wording understood by insurers to be in place. All of the relevant witnesses (Rolls-Royce representatives, brokers and insurers) are based in London. It is also likely that factual evidence will be required to explain the history behind the Mermaid Pod failures. That evidence has links to a variety of jurisdictions (Sweden, France, England and the location of the dry-dockings in a number of countries) but would logically (it is said) be given by one of the representatives of Rolls-Royce in England who have overseen the Mermaid Pod issues from start to finish. It is also likely that technical evidence will be required from the parties’ respective experts as regards the technical cause of the failures. Primary insurers’ appointed experts, who have been engaged for some time in relation to this dispute, are all based in England. Rolls-Royce’s engineering experts are all said to be based in Sweden (that point I do not think is in dispute). It is accepted that evidence concerning the basis for Rolls-Royce’s liability to the French shipyard and the shipping companies will be required. However it is submitted that what is required is not simply a re-litigation of the underlying claims, rather, the focus will be on the facts relevant to the insurance claim and, while some evidence in relation to those facts may be located in Florida, some will also be located in Sweden, France and England. Wherever the evidence is, it will be readily transportable.

The insurers submit that while Rolls-Royce has claimed that evidence will be required from Florida-based individuals most of whom are from the cruise lines who claimed against them, those individuals are, under English law, very unlikely to have relevant evidence to give. Six potential witnesses are identified as senior executives of RCCL, Regent and Carnival. Apart from the undisputed fact that those entities made claims against Rolls-Royce, their evidence it is said will not be relevant to the insurance claim. The scope of the claims made against Rolls-Royce is set out in the relevant court papers. It is unclear how evidence from lawyers who represented the shipping companies will be relevant. The remaining individuals are Rolls-Royce’s legal representatives in Florida, but it is submitted that their evidence is likely to be of limited assistance.

My conclusions on this issue are as follows. It is by no means easy to discern which witnesses are likely to be required to give evidence, given the fact that the proceedings are at such an early stage. I am satisfied that there will be relevant witness evidence from Florida as regards the third party claims, and I consider that the insurers’ give too little weight to that aspect of the evidence. I doubt that it will simply be a matter of looking at the relevant court papers and the settlement agreements (where reached). On the other hand, there is also likely to be evidence from witnesses in England as to the making of the policies, even allowing for the points made by Mr Miles QC about the limits in respect of the relevance and the admissibility of such evidence. There may be some evidence as regards issues such as dry-docking, though as the insurers point out, this does not necessarily point to either forum, since the dry-docking happened in a number of different countries, though not, it should be noted, in either Florida or in England. There may also very likely be technical expert evidence required. In this regard, the insurer’s experts are based in England. As I understand it, Rolls-Royce’s expert is based in Sweden where the Mermaid Pods were manufactured in part.

Overall, the issues as regards witnesses appear to me to be relatively finely balanced, and do not point decisively in favour of either jurisdiction. I should state that there is (or was) a separate point advanced in the evidence as regards the location of documentation in Florida, but that has not been pursued in oral argument. Documentation can plainly be copied as required. The point has also been made that Rolls-Royce’s Florida lawyers know their way around that documentation, but whilst I would accept that point, it appears to me to go more to the “Cambridgeshire factor” rather than the issue as to witnesses. In any case, I treat this issue as being broadly neutral.

Rolls-Royce submits that the existence of other proceedings in the relevant appropriate forum is a material consideration. The Rolls-Royce companies have it is submitted “(a) commenced the Florida proceedings in which the very same issues as are raised in the later English Proceedings would (if allowed to continue) be decided, and (b) challenged the jurisdiction of the High Court to hear the English proceedings before any challenge was made to the jurisdiction in the Florida proceedings”.

The italics come from Rolls-Royce’s written submissions, showing that this is essentially a timing point. In E.I. Du Pont de Nemours (see above) Bingham LJ said at p.593:

“… I do not regard this as a case in which the dates of beginning proceedings are significant. As it happens, the English proceedings began first and the Illinois action a month later. It might have been the other way round. I do not think the outcome of these appeals should be affected by what is little more than an accident of timing.”

In my view, the same applies in the present case. The Florida proceedings were filed first, the English proceedings were served first. It is unnecessary to have regard to the evidence of when the Florida proceedings should be regarded as having commenced for present purposes. (I should say however that I do not see that the fact that Rolls-Royce took some time to catch up with the name changes of some of the insurers, or the transfer of business assets in one case, is likely to be material.) Looked at realistically, the claims in England and Florida were begun about the same time, and indeed for the same reason, namely that the respective claimants wished the cases to be tried there. The timing to the challenges to the jurisdiction was equally a matter of chance. I can give no weight to this factor.

The main factor that points in the direction of Florida is the accumulated knowledge and experience in Florida, particularly of Rolls-Royce’s Florida lawyers. But for reasons set out above, though I accept the significance of this factor, I do not regard it as decisive in this case. The earlier proceedings in Florida were in respect of its liability for the failure of the Mermaid Pods, whereas this claim is under Rolls-Royce’s insurance cover. The issues overlap, but they are different, and the accumulated knowledge and experience must be seen in that light. The claim is by an English insured against predominantly English insurers under an insurance policy subject to English law. In my judgment Rolls-Royce has not shown that Florida is the appropriate forum for the trial of the claim. It follows that the application for a stay is refused. In those circumstances, I need not decide the Owusu point. I am grateful to all parties for their assistance.