Dornoch Ltd & Ors v Mauritius Union Assurance Company Ltd & Anor

This is a reinsurance jurisdiction dispute. The first appellant, The Mauritius Union Assurance Company Limited (MUA), insured the second appellant, The Mauritius Commercial Bank Limited (MCB), under three policies, each of which it reinsured 100% back to back in the London market. One of the reinsurances contained a Mauritius jurisdiction clause. This appeal concerns one of the others. Aikens J. held, [2005] EWHC 1887 (Comm), that MUA had failed to show a good arguable case that such a clause was incorporated into this contract. He held that English law was its proper law, as it was of the torts alleged against MUA and MCB, and that England was clearly the appropriate forum in which to decide the reinsurers’ claims for declaratory relief against MUA and in tort against MUA and MCB. Accordingly he refused to set aside his earlier order for service out of these proceedings. MUA and MCB appeal with the permission of the judge.

MCB is a commercial bank and MUA an insurance company in Mauritius. From June 1999 MUA provided bankers’ blanket insurance for MCB which it renewed for 12 months from 30 June 2002 to 30 June 2003. The renewal was not completed until MUA’s reinsurances were in place. MCB’s cover consisted of a primary bankers blanket policy on the broker’s BRS 98 form covering a variety of risks including employees infidelity, premises and transit and two excess policies. The excess policies provided increased cover for infidelity and premises and transit respectively. Each of these policies was expressly subject to Mauritius law and jurisdiction. Their other terms do not matter except that they covered losses occurring or discovered during the policy period.

MCB’s primary policy was reinsured in the bankers blanket market and led by Munich Re. This primary reinsurance covered most but not all the risks covered by the underlying policy. It was expressed to be on the “J (a) form plus Wordings as agreed Original Wording based on ….” BRS 98. The conditions on the slip included “Wording as per BRS 98” “Mauritius Jurisdiction Clause”, “Terrorism Exclusion Clause” and “90 day Premium Payment Warranty”. There is some uncertainty as to which original wording was attached to the slip but, like the judge, I shall assume that it showed that the underlying primary policy was subject to Mauritius law and jurisdiction.

The excess infidelity policy was also reinsured in the bankers blanket market and led by Munich Re. It was on a slip policy which contained the terrorism exclusion and premium payment warranty clauses contained in the primary reinsurance, but no jurisdiction clause.

The excess premises and transit policy was reinsured by the respondents (Reinsurers) in the specie (articles of high value) market. That is because premises cover of this kind is not for buildings and ordinary contents but for loss of or damage to high value contents held by banks and similar institutions. This is the reinsurance (the Reinsurance) the subject of this appeal. It was also written on a slip policy. The cover was excess 50m. Mauritian Rupees any one loss. The relevant conditions set out in the slip are:

On 14 February 2003 MCB announced that it had discovered a large scale fraud which had resulted in the misappropriation of over 600m. Rupees between 1991 and 2002. Three months later MCB started proceedings in Mauritius claiming the losses it had suffered from this fraud against thirty-eight defendants. The fraud is alleged to have been committed by one of the bank’s senior managers. He is alleged to have siphoned off and fraudulently misappropriated funds belonging to MCB over the eleven year period by means of various unauthorised advances and transfers, the proceeds of which were paid to a number of recipients by cheque.

The fraud action was soon followed by a claim by MCB in Mauritius against MUA under the three direct insurance policies (The Mauritius insurance action).

On 19 January 2005 Reinsurers notified MUA that they had avoided the Reinsurance for misrepresentation and non-disclosure. On 1 February 2005 Aikens J. gave permission to serve the present proceedings on MUA and MCB in Mauritius. Reinsurers claim declarations that they have validly avoided the Reinsurance alternatively that the loss falls outside its scope and damages for misrepresentation against MUA and damages for deceit, alternatively negligent misstatement, against MCB. They contend that there is no cover because the losses claimed are not attributable to premises or transit risks. The claim is for losses represented by cheques drawn on and paid from MCB accounts as a result of a long running fraud. None of the many losses occurred within the 72 hour discovery period and only one of them exceeds the 50m. Rupee excess point and that relates to a cheque drawn in December 1994. The allegations of misrepresentation made against MUA and MCB are based upon answers given in a Lloyds bankers policy proposal form signed by officers of MCB to the effect that it was a well run orthodox banking business when in fact it was nothing of the sort. Reliance is placed on a forensic accountants’ report which discloses impropriety by senior staff of the bank and serious irregularities in its management. The allegations of non-disclosure are based on this report and local press reports of wrongdoing at the bank.

On 21 February 2005 the Supreme Court of Mauritius gave permission to join Reinsurers to the Mauritius insurance action as “Defendants – in – Guarantee”, the equivalent of a Part 20 claim against a third party in this jurisidiciton.

Before considering the issues which arise on this appeal I should note that the hearing before the judge took 4 days. He records that he was provided with 8 large and 1 small ring-binder files of evidence, 90 authorities and over 100 pages of skeleton arguments. His polite protest is recorded at [10] of his judgment.

I should like to add my own rather stronger protest. We were provided with the same volume of material. Counsel very skilfully managed to finish their submissions in the two days which we had available, but we, like the judge, were provided with an unnecessarily large volume of material for an application/appeal of this kind where the general legal principles to be applied are well settled and the judge is not required to reach final conclusions on any issue other than which is the appropriate forum for trial of the proceedings. This may not have been a case to which Lord Templeman’s strictures in Spiliada [1987] 1 AC 460 at 465 needed to be followed to the letter but it is possible that memories of what he said have faded. They echo what was said earlier by Lord Radcliffe (Vitkovice v Korner [1951] AC 869 at 884) and are worth repeating:

It seems to me that the solution of disputes about the relative merits of trial in England and trial abroad is pre-eminently a matter for the trial judge. Commercial court judges are very experienced in these matters. In nearly every case evidence is on affidavit by witnesses of acknowledged probity. I hope that in future the judge will be allowed to study the evidence and refresh his memory of the speech of Lord Goff in this case in the quiet of his room without expense to the parties; that he will not be referred to other decisions on other facts; and that submissions will be measured in hours and not days. An appeal should be rare and the appellate court should be slow to interfere.

This court should, and I will, follow this advice also.

With what I have just said in mind it is clear that the issue at the heart of this appeal is whether the Reinsurance was subject to a Mauritius jurisdiction clause. This is essentially a short question of construction of the slip policy in the context of the other insurance and reinsurance contracts to which I have referred. If the Reinsurance was not subject to such a clause, the judge’s conclusion that England was clearly the appropriate forum for the trial of the English action is difficult to assail. On the other hand, if the Reinsurance was subject to a Mauritius jurisdiction clause the whole basis for the judge’s decision is undermined. Unless there is a strong reason for doing so, the parties should be held to their bargain and their present dispute should be tried in Mauritius. The Spiliada question then only arises in relation to the tort claims. But, if Reinsurers’ contract claim against MUA is to be tried in Mauritius, it is difficult to justify continuing the tort claims here. For these reasons it makes sense to consider this central issue first.

Two preliminary points arise: which law should be used to answer the question and in the context of a jurisdiction dispute of this kind how precisely should it be framed.

Both Mr Malek Q.C. for MUA and Mr Swainston Q.C. for Reinsurers agreed that English law should be used. Mr Kealey Q.C. for MCB however argued and continues to argue that the judge should have used Mauritius law as the putative proper law.

The judge disagreed. He said that he should approach the question as part of the exercise of discovering the proper law of the Reinsurance. Before the proper law had been determined, the only law he could use was the law of the forum: English law.

Mr Kealey argues that this was not the right approach where the question was whether the contract contained a jurisdiction clause. Here the court was required to apply the putative proper law of the contract as if that clause had been incorporated (Mauritius law) applying the principle established in The Parouth [1982] 2 Lloyds Rep. 351 and applied in The Atlantic Emperor [1989] 1 Lloyds Rep. 548.

I do not accept this submission because the principle established in those two cases does not apply to the situation here. In The Parouth the court had to decide whether there was an agreement at all. In The Atlantic Emperor the question was whether there was an arbitration agreement. In both cases, however, if there was an agreement, it was unarguably subject to English law. Our case is not concerned with the existence of an agreement but with one of its terms. Where there is more than one possible putative law (as here) it makes no sense to decide which one to choose by any putative law. At this stage the court has no choice but to apply the law of the forum. (See The Heidberg [1994] 2 Lloyds Rep. 287 at p. 306-308 where Judge Diamond Q.C. adopted this approach).

The question answered by the judge was whether MUA had a good arguable case that the Reinsurance contained a Mauritius jurisdiction clause. There is no issue about the standard of proof required: a good arguable case. But Mr Swainston submits that the correct question was whether Reinsurers had a good arguable case that the Reinsurance did not contain a Mauritius jurisdiction clause. He makes good this submission by reference to Islamic Arab Insurance Company v Saudi Egyptian American Reinsurance Company [1987] 1 Lloyds Rep. 315, another reinsurance jurisdiction dispute where an order for service out on a Saudi reinsurer was upheld in this court. One of the insurer’s grounds for founding jurisdiction was that English law was the proper law of the contract. Reinsurers contended for Saudi law. In the course of his judgment Parker L.J., with whom Bingham L.J. agreed, said ( 317):

The question is whether the plaintiffs have a good arguable case that English law is the proper law. If they have, then there is jurisdiction to give leave. It may well be that there is also a good arguable case for some other law being the proper law and, if the action goes forward that case will prevail at the trial. That is not the point, at all events unless it is clear that the question of proper law cannot be further illuminated at the trial.

In this case the argument about the jurisdiction clause formed an important part of the broader question as to whether English law was the proper law of the Reinsurance so as to found jurisdiction under CPR 6.20 (5) (c), even though a basis for jurisdiction was conceded under 6.20 (5) (a) and (b) because the contract was made here through an agent. So I think Mr Swainston is right to formulate the question as he does and the fact that MUA may have a good arguable case also is not directly to the point. As it was, the judge’s question favoured MUA, so no harm was done.

Finally I must explain how the words Jurisdiction Clause came to be in the Reinsurance slip.

On 29 May 2002 a junior broker at BRS approached the leading underwriter of the first respondent (syndicate 1209 at Lloyds) to obtain a quote. He produced a typed quotation sheet which set out the terms of the cover sought by MUA. It is common ground that before giving her quote the underwriter asked the broker three questions. The first two questions related to terrorism exclusion and premium warranty clauses. She noted the answers to these questions “NMA 2921” and “LSW 3000 – (90 days)” on the quotation sheet. Her third question related to a jurisdiction clause. There is a conflict of evidence about how the broker answered this question.

In his written statement made in April 2005 the broker said:

I told her that there was a local jurisdiction clause in the primary and she noted jurisdiction clause. I recall specifically explaining to her that these conditions would be included in the excess layer as the terms and conditions followed those of the primary policy

After she had seen the broker’s statement and the quotation sheet in June 2005 the underwriter made a statement in which she said she did not recall any discussions about a Mauritius jurisdiction clause either at quotation or placement. She said:

My reference to jurisdiction clause [on] the quotation sheet was, I am confident, the result of my raising this important issue with [the broker]. Unlike the other two notes I made it is obviously incomplete. It has a colon after it. I simply do not accept what [the broker] says... it is obviously the case that the question of jurisdiction was not resolved. I am sure, having now seen the quotation sheet, that my note, with the colon, reflects the fact that I must have asked [the broker] what jurisdiction applied and he did not know and had to go back … for instructions.

No contract was made until the underwriter scratched the slip on 19 and 20 June 2002. It is common ground that there was no further discussion about a jurisdiction clause on these occasions. She says that:

On the slip that he brought back to me, the colon had been removed and it simply read "jurisdiction clause”. It seems to me now that I must have missed this.

She refers to her underwriting notes which contain no reference to a jurisdiction clause although they do refer to the terrorism exclusion and premium warranty clauses.

Before the judge MUA contended that the words “to follow all terms and conditions of the primary policy together with riders and amendments thereto covering the identical subject matter and risk …” incorporated the Mauritius jurisdiction clause from the primary reinsurance. The words “jurisdiction clause” appearing on the slip confirmed this or at least did so taken with the broker’s evidence.

The judge disagreed [53]. He said that the general words of incorporation did not embrace a jurisdiction clause. They only related to risk. This was confirmed by the fact that the general words were separated from the words Jurisdiction Clause which showed that the parties contemplated that there would be a separate clause setting out the applicable jurisdiction agreement. In the event there was nothing in or attached to the slip policy which showed what the words meant. They were therefore meaningless. However he thought that ultimately the issue could be decided easily if the court resolved the dispute between the broker and the underwriter about what happened on 29 May 2002. But he could not resolve it on paper and expressed no view as to either party’s prospect of success.

There are many cases in which the courts have had to decide whether terms from one contract have been incorporated into another. A number of these cases concern the incorporation of terms from a direct insurance into a reinsurance. But no hard and fast rules emerge from these cases as one would expect. The question in each case is one of construction: did the parties to the contract in which the general words of incorporation appear intend that their contract should include the particular term from the other contract referred to? It may be, as Mr Kealey submits, that the courts will answer this question in favour of incorporation more readily in some categories of cases than in others, but that is no more than saying that the contractual context and the words used are all important. As choice of law and jurisdiction clauses are important, clear words of incorporation are required. In the insurance context where the contracts concerned are back to back and cover the same subject matter and interest incorporation is more likely to have been intended than where the contracts are not so closely connected.

In this case I accept that the contracts are closely connected. But they are not a complete match. The Reinsurance followed the primary reinsurance in the sense that it provided an extra layer of cover above that provided by the primary. But the primary did not cover all the risks covered by the underlying primary insurance and the Reinsurance only covered premises and transit risks. Moreover the primary reinsurance does not contain the 72 hour infidelity extension clause contained in the Reinsurance. It is also worth noting that the other excess reinsurance slip policy makes no express reference to jurisdiction.

Like the judge, I do not think the general words in the slip incorporate the Mauritius jurisdiction clause from the primary reinsurance. They relate, as they say, to the “subject matter” and “risk” and I do not think they were intended to include a jurisdiction clause. This conclusion is borne out by the fact that the slip clearly intended to deal with jurisdiction separately. The primary reinsurance slip also dealt with jurisdiction separately and expressly, even though it refers to “wordings as agreed original wording” and “wordings as per BRS 98” and such wordings were attached to the slip.

So what do the words Jurisdiction Clause mean and what is the relevance of the conflict of evidence between the broker and the underwriter about this? I will deal with the latter question first.

Because Mr Malek argued that the meaning of the words on the slip were clear, his primary submission was that this evidence was inadmissible because it was part of the pre-contract negotiations between the parties. However he suggested that it might be admissible if it could be said to amount to some freestanding oral agreement or to resolve an ambiguity. Mr Kealey agreed and suggested that it also might be admissible to show that the parties had agreed on the meaning of some contract term.

On analysis I think this evidence was inadmissible. It is not suggested that any agreement was reached on 29 May 2002. The underwriter was asked to give a quote but was not bound until she first scratched the slip on 19 June. The broker does not say that they agreed what the words jurisdiction clause should mean and it would have been difficult for him to do so in the face of the underwriter’s colon. His evidence goes no further than saying that he told her what the Reinsurance would contain. These were therefore pre-contract negotiations which could not be used to shed light upon the meaning of the words on the slip.

Mr Malek and Mr Kealey forcefully submitted that these words were clearly intended by the parties to pick up the jurisdiction clause in the primary reinsurance slip and should be so construed so as to give them meaning. Mr Kealey invited us to compare the two slips which each contained terrorism exclusion and 90 day premium warranty clauses and submitted that when one did so it was obvious that the words in the reinsurance slip referred to the words Mauritius Jurisdiction Clause in the primary reinsurance slip.

I do not agree. The terrorism exclusion and premium warranty clauses are not identically worded in the two slips. More importantly the primary slip shows that where the brokers intend a Mauritius jurisdiction clause they say so. The words Jurisdiction Clause on their own, not followed by a full stop, simply cannot be taken to refer to the words in the primary.

A further argument might be that the words indicate that the parties intended there to be some sort of jurisdiction clause and as the only candidate was Mauritius that is what they must be taken to have agreed.

I do not accept this argument. The words certainly indicate that the parties intended to have a jurisdiction clause, but I do not accept that Mauritius was the only candidate. The matter was left at large. An English jurisdiction clause was just as much a candidate as Mauritius and one might have expected an English jurisdiction clause to appear as a condition in this slip.

Taken on their own the words Jurisdiction Clause are meaningless. Taken in context I think they are still meaningless. The most one can say is that the parties intended to agree a jurisdiction clause but in the event never did so.

For these reasons I think the judge answered the question he posed for himself correctly. If the question is formulated in the negative, as I think it should be, Reinsurers clearly have a good arguable case that the Reinsurance was not subject to a Mauritius jurisdiction clause.

I propose to deal with these issues quite shortly in view of what I have said in [12].

Having resolved the central issue the judge went on to consider the proper law of the contract [57]-[71], noting that everyone was agreed that he could not finally decide that question at this stage. The applicable provisions of the Rome Convention are:

Article 3 (1)

A contract shall be governed by the law chosen by the parties. The choice must be express or demonstrated with reasonable certainty by the terms of the contract or the circumstances of the case.

Article 4 (1)

To the extent that the law applicable to the contract has not been chosen in accordance with Article 3, the contract shall be governed by the law of the country with which it is most closely connected.

Article 4 (2)

… it shall be presumed that the contract is most closely connected with the country where the party who is to effect the performance which is characteristic of the contract has, at the time of the conclusion of the contract, … in the case of a body corporate or unincorporate, its central administration. …

The judge decided that he could not say that either party had demonstrated with reasonable certainty by the terms of the contract or the circumstances of the case what the proper law was. There was therefore a stalemate so far as Article 3 was concerned. However under Article 4 he decided that there was a good arguable case that the contract was most closely connected to England because the characteristic performance of a reinsurance contract was payment in the event of a claim and that would be performed here.

MUA and MCB contend that the judge was wrong to conclude that there was a stalemate under Article 3. He could and should have expressed a provisional view that the proper law was Mauritius law because the choice of law clause in the primary reinsurance was incorporated into the Reinsurance. Choice of law clauses are more easily incorporated than jurisdiction clauses because there is a general presumption that contracts which follow one another are subject to the same law. Mr Kealey submits that because the judge did not reach a decision under Article 3 it was not open to him to consider Article 4.

I do not accept these submissions. One must remember that the judge was only concerned to decide whether Reinsurers had a good arguable case that English law was the proper law of the contract and that he was not making any final decision on the point. The argument for incorporation went both ways: the general presumption relied on by MUA and MCB, as against the fact that the Reinsurance was broked on the basis of a Lloyds proposal form by English brokers and written on a Lloyds slip in the London market on London market terms where it would be “surprising and improbable” (Hobhouse J. in Vesta v Butcher) that the contract was governed by anything other than English law. In these circumstances I do not think the judge’s decision that there was a stalemate can be faulted. If he could not determine the matter by reference to Article 3, I do not see how he had any other option but to go to Article 4. Applying this Article he obviously reached the right conclusion.

Here the judge had to apply sections 11 and 12 of the Private International Law (Miscellaneous Provisions) Act 1995 which provide:

11 (1) The general rule is that the applicable law is the law of the country in which the events constituting the tort or delict in question occur.

11 (2) Where elements of those events occur in different countries the applicable law under the general rule is to be taken as being … (c)… the law of the country in which the most significant elements of those events occurred.

12 (1) If it appears, in all the circumstances, from a comparison of

(a)

the significance of the factors which connect a tort or delict with a country whose law would be the applicable law under the general rule; and

(b)

the significance of any factors connecting the tort or delict with another country,

that it is substantially more appropriate for the applicable law for determining the issues arising in the case, or any of those issues, to be the law of the other country, the general rule is displaced and the applicable law for determining those issues or that issue (as the case may be) is the law of that other country.

(2)

The factors that may be taken into account as connecting a tort or delict with a country for the purposes of this section include, in particular, factors relating to the parties, to any of the events which constitute the tort or delict in question or any of the circumstances or consequences of those events.

The judge identified six significant elements which made up the torts alleged [105]. He concluded that the most significant elements were making the untrue statements in the proposal form, presenting the untrue statements through the broker chain to Reinsurers and reliance by them on those untrue statements to their loss. The true state of affairs in MCB was important, but it was what was done with those facts which really mattered for the torts alleged. That started in Mauritius, but was continued in England and relied on here. Reliance was the most significant element [106] and [107].

Mr Kealey criticises the judge’s assessment of the significance of the elements of the tort which he identified. He says that the most significant element was whether the statements made by MCB were in fact inaccurate and whether MCB acted negligently or dishonestly in making them. Those elements had nothing to do with England.

But an assessment of this kind is quintessentially a question for the commercial judge. I can see no basis for saying that the judge made such an error in his assessment as to require this court to make its own assessment.

Then it is said that the judge misunderstood and failed to apply section 12. This section gives the judge a wider remit, with its reference to connecting factors, than section 11 which focuses on the constituent elements of the tort. The judge did not conduct a separate exercise under section 12 because he said this involved, in this case at least, considering the same elements all over again [109].

Having regard to the way in which the judge considered the various elements of the tort I think he was right to say that section 12 added nothing to MCB’s argument. He was not saying that section 12 would never require the court to look at factors which had not been considered under section 11, but simply that it did not add anything to do so in this case.

The judge said that his conclusion that English law was the proper law of the contract was of very great importance. Based on expert evidence of Mauritius law it appeared that the courts of Mauritius would not apply the Rome Convention, would conclude that the Reinsurance was governed by Mauritius law which would apply significantly different principles to the issue of construction and to the issues of non-disclosure and misrepresentation which would mean that MUA’s defences to MCB’s claims in the Mauritius insurance action were unlikely to succeed. Reinsurers would be likely to have to follow MUA’s fortunes [72]-[79].

So far as the tort claims were concerned the judge said that there was a danger that the courts of Mauritius would apply the wrong proper law and he had no evidence as to how Mauritius law would consider the claim against MCB [110].

Both Mr Malek and Mr Kealey say that the judge was wrong to have attached any real weight to these considerations. The assumption that the courts in Mauritius will make the “wrong” decision according to English law principles is impermissible. There is no justification for treating those principles as being superior to those of any other country, particularly where it is our conflict of law rules which are being invoked to justify jurisdiction over parties domiciled in that country.

The judge rightly rejected any suggestion that Reinsurers would not get a fair trial in Mauritius [96] and [97]. Nor was he being critical of Mauritius law. But I think he was entitled to take into account the juridical disadvantages to Reinsurers if the trial took place in Mauritius, notably the absence of the relief which English insurance law affords to an insurer (or reinsurer) where there has been misrepresentation or non-disclosure.

A further point is taken relying on the decision in Vesta v Butcher [1989] AC 852. In that case a contract of insurance and a facultative reinsurance, under which part of the original risk was reinsured, contained warranties in identical terms. The House decided that the warranty in the Reinsurance which was governed by English law, should be construed so that it had the same effect as the warranty in the insurance which was governed by Norwegian law which required the breach to be causative of loss.

Here it is said that the same applies to the 72 hour infidelity extension clause and, so it is argued, Mauritius law will govern the construction of this clause anyway. This reduces the importance attached by the judge to the proper law. The judge rejected this argument because the situation in this case is not the same as it was in Vesta v Butcher. Here the wording incorporated is that of the primary reinsurance policy which does not contain the 72 hour infidelity extension clause and there was no reason to conclude that the parties to the Reinsurance intended that the clause should be interpreted in the same way as the similar but not identical clause in the underlying excess policy [85].

MUA and MCB argue that the judge was wrong about this. I do not agree, but even if he was the point has not been finally decided and if it was decided the other way I do not think that this would significantly undermine the judge’s assessment of the importance of the proper law being English law.

At the end of his long and careful judgment the judge made a full assessment of the factors for and against the English forum in relation to the claim against MUA and then the claim against MCB [117] – [123]. Overall he concluded that England was the forum in which the claims could be most suitably tried for the interests of all the parties and the ends of justice. In doing so he recognised that the general policy of English law to avoid multiplicity of proceedings could not be upheld [125].

Both MUA and MCB submit that the judge attached too little weight to the latter point. I do not think he did. His assessment included a careful analysis of where there would be duplication and where there would not. By this stage of his judgment the judge was deep into discretion territory. I can find no fault in the way he exercised it.

As I suggested in [12], I think the outcome of this appeal depends upon whether Reinsurers have a good arguable case that the Reinsurance was not subject to a Mauritius jurisdiction clause. I think they have. This means that the judge was right to proceed to answer the Spiliada question. Despite the attack which is made on various aspects of his consideration of this question I do not think it has been shown that there are any grounds for interfering with his conclusion.

For these reasons I would dismiss this appeal.