AIG Europe SA (formerly AIG Europe Ltd) & Ors. v John Wood Group Plc & Anor.

The issue in these appeals is whether Mr Justice Jacobs was right to grant an anti-suit injunction restraining the defendants from pursuing proceedings in the Court of Queen’s Bench in Alberta, Canada under various policies of excess liability insurance. That depends on whether the policies provide for the exclusive jurisdiction of the English court, so that the pursuit of those proceedings would be a breach of contract by the defendants.

The terms of the various policies are not identical, but they have some features in common. Save where it is necessary to distinguish between them, I can refer to the respondents, who are the insurers under the various policies, as the claimants and to the appellants, who are the insureds, as the defendants.

The policies in issue form part of an “insurance tower”, comprising a primary liability policy, together with a number of excess layers. Each of the excess policies was in the form of the “Market Reform Contract”, which is now the standard form of agreement used in the London market. The Market Reform Contract consists of a series of sections, the first of which sets out the “Risk Details”. These include such matters as the unique market reference, the type of policy, the interest insured, the monetary limits and (importantly in this case) the parties’ choice of law and jurisdiction. A later section of the policies sets out what appear to be standard terms and conditions.

In each of the excess policies in issue there were two relevant, and potentially conflicting, clauses. The choice of law and jurisdiction clause in the Risk Details provided for disputes to be subject to “the same law and the same jurisdiction as the primary policy”, although the primary policy itself did not contain any choice of law or jurisdiction clause. The standard terms and conditions in the later section of the excess policies contained a clause which provided for English law to apply and provided also (or in one case, allegedly provided) for the exclusive jurisdiction of the English court.

The claimants found their application for an anti-suit injunction on the exclusive jurisdiction clauses in the later section of the policies. The defendants, however, say that the applicable clause is that set out in the Risk Details, a clause referred to in these proceedings as the “primary policy jurisdiction clause” or “PPJC”, which does not provide for exclusive English jurisdiction, with the consequence that their pursuit of proceedings in Alberta is not a breach of any obligation owed to the claimants.

The judge held in each case that the policies did provide for English exclusive jurisdiction and therefore granted the injunctions sought. The defendants appeal. Accordingly we are concerned with questions of construction of the various policies. If the judge was right to conclude that, on their true construction, the policies provide for exclusive English jurisdiction, an injunction should be granted. If not, not.

In opening the appeal, Mr Roger Stewart QC for the defendants submitted (or came close to submitting) that the English court should not embark on this question of construction, but should leave it to be decided by the court in Alberta on an application by the claimants (the defendants in the Alberta case) for a stay of the proceedings against them there. This was said to be a matter of judicial comity. It was not a submission made to the judge below and did not feature in the defendants’ appellants’ notice. Accordingly it is not open to them in this court.

Nevertheless it may be helpful to say that when the parties have agreed an exclusive jurisdiction clause, the view taken in this jurisdiction is that judicial comity is best served by giving effect to their agreement and that the grant of an anti-suit injunction in an appropriate case is a proper means of doing so. This appears clearly from the decision of this court in OT Africa Line Ltd v Magic Sportswear Corporation [2005] EWCA Civ 710, [2005] 2 Lloyd’s Rep 170, which followed and built on the decision of the Supreme Court of Canada in Amchem Products Inc v Workers Compensation Board [1993] 1 SCR 897. Lord Justice Longmore said:

This approach was developed further in Ecobank Transnational Inc v Tanoh [2015] EWCA Civ 1309, [2016] 1 WLR 2231. Lord Justice Christopher Clarke explained that comity may have a greater role to play, even in an exclusive jurisdiction clause case, if the anti-suit injunction applicant is guilty of delay, thereby causing time, effort and expense in the foreign proceedings to be wasted. However, no such considerations arise in the present case.

Ultimately Mr Stewart did not press the submission that this court should not determine the true construction of the policies. Instead he submitted that, before granting an anti-suit injunction, the English court should be satisfied to a high standard (“a high degree of probability”) that pursuit of the Alberta proceedings would be a breach of an English exclusive jurisdiction clause. This was the test which the judge applied. In fact, the principles applicable to the grant of anti-suit injunctions founded on breach of an arbitration or exclusive jurisdiction clause were not in dispute before the judge. He set them out at [58] of his judgment. For present purposes it is sufficient to set out principles (c), (f) and (g):

I shall adopt this approach.

Finally by way of introduction, I should record that before the judge there was an argument that no injunction should be granted because the claimants had failed in their duty to make a fair presentation of the matter when first seeking an injunction without notice. The judge rejected that argument and the defendants were refused permission to appeal on that issue.

The first defendant, John Wood Group Plc, is the parent company of a multinational engineering business. It is incorporated and has its principal office in Aberdeen, Scotland. The second defendant, Wood Group Canada Inc, is an indirect wholly-owned subsidiary incorporated in Alberta, Canada. It is a defendant to proceedings issued in Alberta in June 2017 arising out of the rupture of a pipeline in July 2015. The claimant in those proceedings is Nexen Energy ULC, which alleges that the rupture was caused by the negligence and breach of contract of Sunstone Projects Ltd, a predecessor in interest of the second defendant. Nexen seeks damages of approximately CAD $450 million. Liability has yet to be determined.

The second defendant commenced proceedings in Alberta on 5^th February 2021 against (among others) the claimants, seeking defence cover for the claim by Nexen and an indemnity in respect of any damages that it is ordered to pay to Nexen. That led the claimants to seek without notice anti-suit injunctions from the Commercial Court in London, which were granted to the claimants in claim CL-2021-454 (“the AIG claimants”) on 2^nd August 2021 and to the claimants in claim CL-2021-484 (“Chubb”) on 18th August 2021, in both cases until a return date. The return date hearing took place on 15^th September 2021 before Mr Justice Jacobs, who handed down his judgment on 24th September 2021.

The policies issued by the various insurers comprise a programme of excess liability insurance placed in the London market. There was what is described as an “insurance tower”, which (for my part) is most easily understood by reference to the diagram at Annex A to this judgment, which should be read together with the summary which follows. Although the claim made in the Alberta proceedings refers to both the 2014/15 and 2015/16 policy years, there is for present purposes no material distinction between those years. The judge described the coverage for the 2015/16 year as follows.

The claimants allege that, at the bottom of the tower, there was a self-insured retention of US $2 million. There is a dispute as to the existence of this retention, but it is not relevant to the current issues.

Between the (alleged) self-insured retention, and the excess programme subscribed by the insurers (comprising, in sequence, the Global Umbrella Policy and then the First, Second and Third Excess Policies – referred to collectively by the judge as “the four excess policies”), were a very large number of “underlying” policies. Each of the four excess policies contained an identical schedule of these underlying policies headed: “List of Underlying Policies”. This was followed by the following text:

These underlying policies, collectively, were referenced in the wording of each of the four excess policies. Thus, in the Global Umbrella Policy, the “Limit of Liability” was expressed to be US $15 million each and every occurrence and US $45 million in the aggregate “In Excess of various Underlying limits as follows”. The schedule, comprising approximately 70 policies, then followed. The Global Umbrella Policy then contained a further reference to this schedule in the following terms:

The First Excess Policy contained similar provisions, but modified to reflect the fact that the First Excess Policy sat above the Global Umbrella.

Thus, the Limit of Liability provision was in the following terms:

Similarly, the “Underlying Insurer(s)” were described in the following terms:

The schedule of approximately 70 policies then followed.

The Second Excess Policy contained materially identical provisions, save that it reflected the fact that it sat above the First Excess. Thus, the Underlying Insurer(s) referred to were:

Similarly, the Third Excess Policy added the Second Excess layer to the list of policies identified against “Underlying Insurer(s)”.

Thus each of the excess policies identified one particular policy, against the words “Underlying Insurer(s)”, under the heading “Primary”. This was policy CSZ G27175314 003 issued by ACE INA Insurance. This policy was referred to by the parties and the judge as the “Global CGL”. It provided, among other things, “Commercial General Liability Coverage” (hence the abbreviation “CGL”), as well as employee benefits coverage and contingent auto liability coverage. It was not truly “Global”, in that its coverage excluded the United States of America. However, it was “Global” in the sense that it covered liabilities arising in the rest of the world. The Global CGL was signed by two officers of Ace American Insurance Company, described in the policy as “Ace USA”. The Named Insured under the Global CGL was John Wood Group PLC.

We are concerned in this appeal with the Global Umbrella Policy, written by Chubb, and with the First and Third (but not the Second) Excess Policies, written by various insurers including AIG Europe.

The Global CGL Policy issued by Ace USA contained no express clause concerning either applicable law or jurisdiction. The four excess policies did contain such clauses. I take them in turn.

The Global Umbrella Policy contained the two clauses to which I have already referred. The first, the clause referred to in these proceedings as the “Primary Policy Jurisdiction Clause” or the “PPJC”, had the side heading “Choice of Law and Jurisdiction”. It provided:

An identical clause was also contained in each of the other three excess policies. Perhaps mindful of Lord Hoffmann’s rational businessman and the sweeping away of fine linguistic distinctions (see Fiona Trust & Holding Corporation v Privalov [2007] UKHL 40, [2007] 4 All ER 951 at [7], [12] and [13]), no party sought to attach any significance to the fact that the disputes referred to in this clause were only those concerning the “interpretation” of the terms of the policy.

In each case the PPJC was included in the Risk Details section of the policy. Standard practice, described in Merkin, Colinvaux’s Law of Insurance, 12^th Ed, paras 1-082 to 1-094, is that the risk presented by the broker to the market will consist of an introductory section, the Risk Details, setting out the most important details of the risk (including the choice of law and jurisdiction), together with (among other things) standard terms and conditions.

In the case of the Global Umbrella Policy, the standard terms and conditions included clause 11, which provided as follows:

There was, therefore, at least a potential conflict between the PPJC, which provided for “the same law and the same jurisdiction as the primary policy”, and clause 11, which (except for policies issued in Scotland) provided for English law and jurisdiction.

In addition to the PPJC, the First Excess Policy contained a standard clause 12 as follows:

Again, therefore, there was a potential conflict between the PPJC and clause 12. In clause 12 of this policy, in contrast with clause 11 in the Global Umbrella Policy, the jurisdiction of the English court was expressly stated to be “exclusive” and there was no exception in the case of Scottish policies.

In the Second Excess Policy, the relevant clause did not address jurisdiction, but only applicable law. Clause 4.10 provided:

The judge held, not surprisingly, that this clause did not provide for English jurisdiction and, accordingly, that in the case of this policy there was no basis for an anti-suit injunction. That decision is not challenged. Accordingly we are not concerned with the Second Excess Policy in this appeal.

The Third Excess Policy was in materially identical terms to the First Excess Policy.

One of the insurers subscribing to the Second and Third Excess Policies was Allied World Assurance Co (Europe) DAC (“AWAC”). Unlike the other insurers subscribing, for whom the terms agreed were as described above, AWAC included an arbitration clause in its policies, providing for English arbitration, and annotated the PPJC to make clear that it was not agreed. The judge held, again not surprisingly, that these arbitration clauses entitled AWAC to an anti-suit injunction. There is no appeal from that decision.

It follows that, regardless of the outcome of these appeals, there may be proceedings relating to the insurance coverage in both England and Canada. In England there will be arbitrations under the Second and Third Excess policies subscribed by AWAC, while in Canada there will be court proceedings under the Global CGL Policy and under the Second Excess policies subscribed by the remaining insurers.

The claimants relied on clause 11 of the Global Umbrella Policy and clause 12 of the First and Third Excess Policies as providing for the exclusive jurisdiction of the English court.

The judge dealt first with the Global Umbrella Policy. The defendants submitted that clause 11 was not a proper basis for the grant of an anti-suit injunction, or at any rate that this had not been demonstrated to a high degree of probability, for three reasons (I mention these in the order in which the judge referred to them, although Mr Stewart told us that his primary case was always based on the PPJC):

First, the Global Umbrella Policy was “issued in Scotland”, so that the provision in clause 11 for the Commercial Court in London to have jurisdiction did not apply.

Second, clause 11 should be construed as providing for the non-exclusive jurisdiction of the Commercial Court in London, thus leaving the parties free to sue in any court of competent jurisdiction, including the Alberta court.

Third, the applicable jurisdiction clause was the PPJC, which provided for the same law and the same jurisdiction as the primary policy, that is to say the Global CGL Policy. As that policy did not contain any jurisdiction clause, it was open to the defendants to sue on that policy in any court of competent jurisdiction; and the same court should have jurisdiction under the Global Umbrella Policy. Alternatively, reading the PPJC together with clause 11 was a further reason to construe clause 11 as providing for non-exclusive English jurisdiction.

The judge rejected all three submissions. In summary he held that:

What mattered for the purpose of clause 11 was “the location of the office which is the origin of the policy document provided to the insured”. The fact that John Wood Group Plc was a company registered in Scotland with a “Principal Address” in Aberdeen identified in the Risk Details section of the policy was irrelevant. In any event other insured companies within the group were located in multiple jurisdictions. Here the policy was negotiated in London between brokers acting for the insureds and underwriters who were based in London. The policy was issued when the contract was concluded as a result of the policy documentation being “scratched” by the underwriters, which had happened in London.

On its true construction, clause 11 provided (except in the case of Scottish policies) for the exclusive jurisdiction of the Commercial Court in London. Many of the considerations which had led the Court of Appeal in Compania Sud Americana de Vapores SA v Hin-Pro International Logistics Ltd [2015] EWCA Civ 401, [2015] 2 Lloyd’s Rep 220 to conclude that the clause in that case was an exclusive jurisdiction clause (even though it did not use the word “exclusive”) were equally applicable in the present case.

The “primary policy” referred to in the PPJC was the Global CGL policy (which was common ground between Chubb and the defendants) but, as this policy contained no law or jurisdiction clause, it was inapplicable and could not displace the clear provisions of clause 11. The PPJC was designed to be used in conjunction with primary policies which contain an express applicable law and jurisdiction clause, in which case there would be no difficulty in importing the relevant law and jurisdiction of the primary policy into the excess layer. This construction had the benefit of certainty, providing a single applicable law and a single jurisdiction to govern disputes under the Global Umbrella Policy. It avoided the unacceptable consequences of the defendants’ submissions, namely that there would be a floating proper law dependent upon where the underlying claims were made and a large range of potential courts having jurisdiction; indeed, if an insurer under the primary policy were to pay a claim without the need for legal proceedings anywhere, the applicable law and jurisdiction under the PPJC would be unknown and unknowable. There was, moreover, nothing in the PPJC to require the agreement for exclusive jurisdiction in clause 11 to be “watered down” by being read as providing for non-exclusive jurisdiction.

Clause 12 of the First and Third Excess Policies provided expressly for the exclusive jurisdiction of the Commercial Court in London. Accordingly the arguments as to the true construction of clause 11 of the Global Umbrella Policy did not arise. The only issue was as to the relationship between clause 12 and the PPJC. For the same reasons as in the case of the Global Umbrella Policy, the judge held that clause 12 of the First and Third Excess Policies was the applicable jurisdiction clause. In the course of doing so, he rejected an alternative submission by the AIG claimants that the primary policy referred to in the PPJC in their policies was the Global Umbrella Policy and not the CGL Policy.

It is now common ground between all parties that the primary policy referred to in the Global Umbrella Policy and both the First and Third Excess Policies is the CGL Policy.

Mr Stewart for the defendant appellants relied principally on the PPJC. He submitted that the judge had been wrong to prefer clause 11 of the Global Umbrella Policy and clause 12 of the First and Third Excess Policies to the PPJC, and that he had adopted the wrong approach. His approach had been to start with clauses 11 and 12 respectively and to ask whether the PPJC “displaced” them. This was an error, effectively adopting a presumption in favour of clauses 11 and 12 which the PPJC was then required to displace.

Mr Stewart submitted, in summary, that there was a clear conflict between the PPJC and the later clauses. That was unquestionably so in the case of the First and Third Excess policies, and equally so in the case of the Global Umbrella Policy if clause 11 is to be construed as providing for exclusive as distinct from non-exclusive jurisdiction. The conflict should be resolved in favour of the PPJC. Choice of law and jurisdiction was a mandatory heading in the Risk Details section of the Market Reform Contract. The rational reader of the policy, familiar with the market, would know that the Market Reform Contract is required to include a clause in the Risk Details dealing with applicable law and jurisdiction, and would therefore know where to look to find out what law and jurisdiction applied. Having done so, he would not scour the policy to see if what was said there was contradicted or qualified elsewhere; but even if he did, he would regard the clause contained in the Risk Details as the governing clause.

Mr Stewart relied on various canons of construction as supporting this approach. The PPJC was upfront in the Risk Details (or main terms) section of the policies rather than being contained in the printed standard terms towards the back of a policy which extended to 30 pages (cf. Lord Hoffmann’s approach to a bill of lading contract in The Starsin [2003] UKHL 12, [2004] 1 AC 715 at [82], observing that in some cases the reasonable reader who has found the information he is looking for on the front of the document does not trouble with what is said on the back). The PPJC was a specifically negotiated term of the agreement (because the parties had to decide what to say under that mandatory heading) as distinct from the standard terms and conditions, and should therefore prevail (Generali Italia SpA v Pelagic Fisheries Corpn [2020] EWHC 1228 (Comm), [2020] 1 WLR 4211 at [85]). It served the useful commercial purpose of ensuring that, even though the applicable law and jurisdiction might not be determined at the outset, the same law and jurisdiction would apply to disputes throughout the insurance tower. This was in accordance with the policy of one-stop adjudication and was a valid and sensible choice for the parties to make (e.g. Fiona Trust; Deutsche Bank AG v Sebastian Holdings Inc (No 2) [2010] EWCA Civ 998, [2011] 2 All ER (Comm) 245).

As for the PPJC itself, Mr Stewart submitted that it was no obstacle to its operation that the primary policy (i.e. the CGL Policy) does not contain any clause dealing with the applicable law or jurisdiction. This is not a case of incorporation of terms from the CGL Policy; rather, the operative provision of the PPJC is that the applicable law and jurisdiction will be the same as for the CGL Policy. The CGL Policy must have a governing law from the outset which is capable of being ascertained (although Mr Stewart made no submission as to what law does apply) and, although there may be many courts which might accept jurisdiction, what mattered to the parties was that the same law and jurisdiction would apply consistently to the primary policy and the excess layers.

Alternatively, Mr Stewart submitted that clause 11 in the Global Umbrella Policy should be construed as providing only for non-exclusive jurisdiction, either on its own terms or when considered together with the PPJC. That was one way in which the otherwise apparently conflicting terms could be read consistently with each other.

Finally, Mr Stewart submitted that the Global Umbrella Policy was issued in Scotland within the meaning of clause 11, so that the provision for English jurisdiction did not apply. The exception for policies issued in Scotland must have been intended to have a real effect in a contract where the principal insured was a Scottish company with an address in Aberdeen.

It is unnecessary to reiterate the principles for construing contracts, set out in a series of Supreme Court cases culminating in Wood v Capita Insurance Services Ltd [2017] UKSC 24, [2017] AC 1173. They are well known. I propose to begin with the meaning of the PPJC. I would be inclined to accept that if there is a conflict between the PPJC and the later clauses, Mr Stewart’s submissions that the former should prevail would have considerable force. In particular, where clauses conflict with each other, I would accept that the location of the clauses within the policy may indicate that one clause “is intended to have a higher contractual status” than another (cf. Generali Italia SpA v Pelagic Fisheries Corpn at [90]). If, however, as the judge held, the PPJC only applies when the primary policy contains a clause dealing with law and jurisdiction, there is no conflict between the PPJC and the later clauses which needs to be resolved and no reason why the later clauses should not be given effect.

In my judgment, and in agreement with the judge, it is clear from the language and context of the PPJC that it does only apply when the primary policy contains such a clause. In the absence of such a clause, the PPJC does not apply – or, in colloquial terms, has nothing to bite on.

First, the words “the same law and same jurisdiction as the primary policy” clearly contemplate a single law and jurisdiction. They refer the reader to the primary policy in order to discover what law and jurisdiction will be applicable to disputes under the excess policy and contemplate that an answer will be found there. But that is not so on Mr Stewart’s construction. Although he submitted that the primary policy must have an applicable law at the time of making the contract (as distinct from a “floating proper law”: cf. The Armar [1981] 1 WLR 207, 215E-216A), it is difficult to know what law would be applied without also knowing in which jurisdiction that question will arise. Courts in different countries apply different rules of private international law to the ascertainment of the applicable law. Moreover, Mr Stewart submitted that there might be a multitude of potential jurisdictions which could apply to disputes under the CGL Policy in the absence of any jurisdiction clause. Either party was at liberty to issue proceedings in any court of competent jurisdiction (there was, as he described it, a “permissive approach”). Where proceedings would take place on that approach would therefore depend on which party was the claimant, where it chose to issue proceedings, and on the jurisdictional rules of the court concerned (for example, whether a stay would be available on forum non conveniens grounds).

Accordingly I would hold that the natural meaning of the clause to an ordinary policyholder would be that the excess policy is to be subject to the same law as is prescribed by the primary policy, not whatever law is applied to the policy by a court where proceedings can be brought by one of the parties to the primary policy; and that the excess policy is to be subject to the same jurisdiction as is prescribed by the primary policy, not to the jurisdiction of any court which a party to the primary policy can persuade to accept jurisdiction. The use of the terms “said territory” and “such court” reinforces this view, as it suggests a single jurisdiction which has already been identified in the primary policy, not one which depends on the happenstance of future litigation under the primary policy.

Second, this conforms in my view with what I would expect to be the approach of the reasonable policyholder. Seeing that clause, the reasonable policyholder could be expected to enquire of the broker what was the law and jurisdiction agreed under the primary policy. On being told that there was no such agreement, the reasonable policyholder would naturally conclude that the PPJC would have no application and that he need not worry about it further.

Third, if the primary policy contains no law or jurisdiction clause, the PPJC is incapable of being applied to a wide range of potential disputes under the excess policies. The primary insurers (whose limit of liability is relatively low) may pay a claim without resort to litigation being necessary, in which case there will be no way of knowing, on Mr Stewart’s approach, what jurisdiction would have applied if there had been litigation. Some disputes under the excess policies will not arise under the primary policy, for example a dispute about the quantum of a claim which on any view exceeds the limit under the primary policy. Other disputes may arise which have nothing to do with the primary policy, for example a dispute about premium under the excess policies. In such a case, if there is no clause in the primary policy, it will be impossible to say what law or jurisdiction is intended to apply: in effect, despite the heading of the PPJC which demonstrates that the parties are making a choice of law and jurisdiction, they will have made no choice at all.

Thus if the primary policy contains an express choice of law and jurisdiction, there is no difficulty in ascertaining what law and jurisdiction apply to any dispute under the excess policy. But if the PPJC is construed as applying where there is no such express choice of law and jurisdiction in the primary policy, its application becomes hopelessly uncertain and, in some cases, impossible. This in my view would be thoroughly uncommercial.

Accordingly there is no conflict between the PPJC and the later clauses which needs to be resolved. The PPJC has no application because the primary policy contains no law or jurisdiction clause and there is no reason why the later clauses should not apply. That is sufficient to dismiss the appeal in the case of the First and Third Excess Policies. In the case of the Global Umbrella Policy, however, it remains to consider the further points which are specific to clause 11 in that policy.

It will be recalled that clause 11 of the Global Umbrella Policy provided as follows:

In my judgment, although the clause does not use the word “exclusive”, the jurisdiction which the Commercial Court in London is intended to have is exclusive rather than non-exclusive. Although each clause turns on its own wording, this conclusion is in accordance with the consistent approach of English law in such cases. In Compania Sud Americana de Vapores SA v Hin-Pro International Logistics Ltd the clause in question, contained in a bill of lading, provided as follows:

Lord Justice Christopher Clarke (in a judgment with which Lord Justice Elias and Lord Justice Beatson agreed and which summarised the effect of previous authority) identified seven reasons for concluding that the clause provided for exclusive English jurisdiction. Obviously there are material differences between the language of the clause in Hin-Pro and clause 11 of the Global Umbrella Policy. In particular, the latter contains nothing corresponding to the second and third sentences of the former. Moreover, the cases are concerned with different kinds of contract. Three of Lord Justice Christopher Clarke’s reasons have no application to the present case. Nevertheless, four of them are relevant and important:

This reasoning has been applied in later cases. For example Mr Justice Foxton in Generali Italia SpA v Pelagic Fisheries Corpn at [92] said that “the choice of English law in conjunction with the reference to English jurisdiction is a powerful factor in favour of construing the choice of English jurisdiction as exclusive”, citing Global Maritime Investments Cyprus Ltd v OW Supply & Trading A/S [2015] EWHC 2690 (Comm) at [50]. It is equally applicable to clause 11 in the present case. The words “shall be governed by and construed in accordance with” provide for the mandatory application of English law. The same mandatory language (“shall have jurisdiction”) is used in relation to jurisdiction. The natural commercial purpose of the clause is to stipulate that English law will govern any dispute and that the Commercial Court will be the court having jurisdiction over any dispute. That makes obvious commercial sense, while the fact that the English cases have generally taken this approach provides some confirmation of what view reasonable business people would take. Moreover, there are no countervailing indications to suggest that, while the application of English law was mandatory, the clause was intended to provide for the non-exclusive jurisdiction of the Commercial Court. In particular, there is nothing in the PPJC to call this conclusion into question, for the reasons already given.

Mr Stewart recognised the imperative language of “shall have jurisdiction”, but submitted that the imperative or direction was directed at the Commercial Court and not the parties. I cannot accept this. It is not the function of a jurisdiction clause to give instructions to a court.

Finally, there is no doubt, and it is common ground, that the Global Umbrella Policy was in fact issued in England. I see no reason to construe clause 11 as if it referred to a policy issued to a (principal) policyholder with an address in Scotland regardless of where in fact the policy was issued. Mr Stewart submitted that it is necessary to construe the clause in this way in order to give effect to the exception for policies issued in Scotland which, he said, must have been intended to have some effect.

In my judgment this begs the question. Clause 11 is plainly a standard term intended for inclusion in policies which may be issued in any one of a number of countries. If the policy in question is issued in England (or for that matter any country other than Scotland) the exception for policies issued in Scotland does not apply. There is no justification for manipulating the plain language of the clause merely because there is a Scottish policyholder.

For these reasons, which are substantially the same as the reasons given by the judge, I am satisfied to a high degree of probability that each of the three excess policies in issue on this appeal provided for the exclusive jurisdiction of the English court, so that the pursuit of proceedings in Canada is a breach of contract by the defendants. I would therefore dismiss the appeal.

I agree.

I also agree.