Lloyd’s Underwriters v. Cominco Ltd.,


2007 BCCA 249

Date: 20070430

Dockets: CA034393, CA034394,

Docket: CA034393


Lombard General Insurance Company of Canada




Cominco Ltd., Teck Cominco Limited and
Teck Cominco Metals Ltd.



- and -

Docket: CA034394


Lloyd's Underwriters




Cominco Ltd., Teck Cominco Limited and
Teck Cominco Metals Ltd.




The Dominion of Canada General Insurance Company as Successor
to the Canadian Indemnity Company; ING Insurance Company of Canada f/k/a the Halifax Insurance Company; American Home
Assurance Company as Successor to New Hampshire Insurance
Company; and Seaton Insurance Company




The Honourable Madam Justice Newbury

The Honourable Mr. Justice Mackenzie

The Honourable Madam Justice Kirkpatrick


G.C. Weatherill
C. Ferris

Counsel for the Appellants,
 Cominco Ltd., Teck Cominco Limited and Teck Cominco Metals Ltd.

J.H. MacMaster
C.A. Rhone

Counsel for the Respondent,
Lombard General Insurance
Company of Canada

G. Mew
O.G. Jones

Counsel for the Respondent,
Lloyd's Underwriters

G.M. Nijman

Counsel for the Respondent,
Seaton Insurance Company

Place and Date of Hearing:

Vancouver, British Columbia

February 26 and 27, 2007

Place and Date of Judgment:

Vancouver, British Columbia

April 30, 2007


Written Reasons by:

The Honourable Madam Justice Newbury

Concurred in by:

The Honourable Mr. Justice Mackenzie

The Honourable Madam Justice Kirkpatrick

Reasons for Judgment of the Honourable Madam Justice Newbury:

[1]                Over the last 30 years, the Supreme Court of Canada has, like its counterparts in the United Kingdom and Australia, made wide-ranging changes to the rules of private international law governing choice of law, jurisdiction simpliciter and “forum non conveniens”, and the recognition and enforcement of foreign judgments.  The old principle of territoriality, so long interpreted and applied with particular rigour by English courts, has been set aside as reflecting a kind of outdated legal imperialism and more importantly, as failing to reflect the complexity and mobility of modern life and economic relationships.  Through judgments such as those in Moran v. Pyle National (Canada) Ltd. [1975] 1 S.C.R. 393, Morguard Investments Ltd. v. De Savoye [1990] 3 S.C.R. 1077, Amchem Products Inc. v. British Columbia (Workers’ Compensation Board) [1993] 1 S.C.R. 897, and most recently, Pro Swing Inc. v. Elta Golf Inc., 2006 SCC 52, the Supreme Court of Canada has adopted a more flexible approach which emphasizes the reasonable expectations of parties, the “connections” between them and the jurisdiction(s) in question, and the principle of “comity” — respect for foreign legal systems whose citizens also have an interest in the objectives of “justice, necessity and convenience” which are said to underlie comity.  (Morguard, supra, at 1098.)

[2]                Although it was concerned specifically with anti-suit injunctions (a remedy not sought in the case at bar), Amchem provides a useful starting point for any case where there are connections to more than one jurisdiction and the parties disagree as to the appropriate forum.  The Court summarized the overarching objective of the new approach to choice of forum as follows:

… The choice of the appropriate forum is still to be made on the basis of factors designed to ensure, if possible, that the action is tried in the jurisdiction that has the closest connection with the action and the parties and not to secure a juridical advantage to one of the litigants at the expense of others in a jurisdiction that is otherwise inappropriate.  [At 912.]

[3]                Of course, the task of determining the more appropriate forum is often difficult in practice.  In this case, it is complicated by the coming into force, during the proceedings below, of the Court Jurisdiction and Proceedings Transfer Act, S.B.C. 2003, c. 28.  According to the explanatory note with which it was introduced (as Bill 31) into the Legislature three years earlier, its purposes included ‘replacing’ the jurisdictional rules being used in the province with a “set of standards for determining court jurisdiction” as recognized by the Uniform Law Conference of Canada; and bringing Canadian jurisdictional rules into line with the principles laid down by the Supreme Court of Canada in the cases mentioned above.  Section 11 of the Act deals with jurisdiction in terms that are more compressed than those used by many courts in deciding choice of forum questions; and whether the Act is intended to effect, or does effect, any substantive change in the law is one of the many issues raised by this appeal.

[4]                The central issue, however, is the relative significance to be given by a Canadian court in deciding whether to decline jurisdiction when jurisdiction ‘simpliciter’ is already established, to the fact that concurrent proceedings are underway in another jurisdiction – in this instance, the United States District Court for the Eastern District of Washington.  All the parties have sought to rely on comity as they interpret it.  The appellant says that the assertion of jurisdiction by the Washington court is a factor of “overwhelming” significance in the determination of forum non conveniens, and that since that court has positively asserted jurisdiction, the Supreme Court of British Columbia was effectively bound to stay the concurrent actions in this province.  The respondents on the other hand submit that British Columbia is the “natural forum” for this insurance coverage dispute, and that the analysis contemplated by the case law and by the new statutory code requires a more nuanced approach under which comity is a significant factor, but not a conclusive one.


[5]                Mr. Justice Davies, the Chambers judge below, had before him applications brought by the defendant Teck Cominco Metals Ltd. (“TCML”), for orders staying two actions in the Supreme Court of British Columbia – one brought by Lloyd’s Underwriters in action number S‑056205, and one brought by Lombard General Insurance Company of Canada (“Lombard”) in action S‑056374.  Each of the plaintiff insurers sought a declaration that it had no obligation to defend or indemnify TCML in respect of environmental damage claims arising from the activities of a predecessor company, Cominco Ltd. (“Cominco”), at a total of four locations in British Columbia.  TCML had notified its insurers of the fact that such claims might be brought against it, and of its intention to look to them for defence and indemnity under the relevant policies.

[6]                It is perhaps unnecessary to state that Cominco (which became “Teck Cominco Metals Ltd.” in 2001) has carried on the business of mining and smelting operations in British Columbia since 1906 and has long played an important role in the economy of south-eastern British Columbia.  It was incorporated in Canada by letters patent and later continued under the Canada Business Corporations Act.  Its head office was in Vancouver.  In 2001, Cominco merged with Teck Corporation.  The merged company, now known as Teck Cominco Metals Ltd., is federally incorporated, and also has its head office in Vancouver.  It has various subsidiaries that carry on business in various countries.  Its subsidiary Cominco American Inc., for example, operates in six U.S. states, including Washington, where it is based.  However, none of TCML, Teck Cominco Ltd. nor Cominco itself has ever carried on business in Washington State.

[7]                The claims or potential claims of which TCML gave its insurers notice arose from Cominco’s operations in Port McNeill, Pinchi Lake, Vancouver, and Trail, British Columbia, respectively.  In Port McNeill, a subsidiary of Cominco allegedly used an ore storage and loading facility between 1961 and 1976, and may be a “responsible person” for purposes of the Waste Management Act, R.S.B.C. 1996, c. 482 and therefore liable with others for remediation costs.  At Pinchi Lake, Cominco operated a mercury mine between 1940 and 1944, and again between 1968 and 1975.  The Tl’azt’en First Nation has allegedly suffered health problems arising as a result of exposure to mercury which Cominco is alleged to have sluiced directly into the lake.  Between 1965 and 1987, Cominco was also a half-owner of a lead alloys production and fabrication facility in Vancouver.  Evidently, there is a potential claim by the City of Vancouver for contamination from that facility.

[8]                By far the largest claim, however, arises from Cominco’s discharge of waste material known as “slag” or “barren slag” into the Columbia River adjacent to its smelter in Trail from some time in the late 1920s until 1997.  This discharge was carried out in accordance with permits issued by the Province of British Columbia under the former Pollution Control Act, R.S.B.C. 1979, c. 332 and the newer Waste Management Act.  Allegedly, however, the effects of the discharge were not confined to British Columbia.  The Columbia River runs south from Trail into Washington State.  When the Grand Coulee Dam was constructed on the river in the 1930s, a lake known as Lake Roosevelt was created, entirely in Washington.  At some point, slag began to accumulate in the Columbia River and in Lake Roosevelt just south of the border.  U.S. environmental authorities became concerned about the discharge problem and entered into discussions with Cominco in the early 1990s regarding the renewal of effluent discharge permits.  A permit issued by the U.S. Environmental Protection Agency (“EPA”) in late 1992 required Cominco to cease discharging slag into the Columbia River by the end of 1996.  TCML says it met this obligation by the installation of a slag collection facility and the introduction of new smelting technology.

[9]                The Confederated Tribes of the Colville Reservation in Washington nevertheless asserted claims for environmental property damage in and around Lake Roosevelt resulting from the effects of the slag deposits.  Negotiations between the EPA and TCML with respect to the Upper Columbia River were unsuccessful, and in October 2003, the EPA issued a “Special Notice Letter” to TCML.  This was followed by a Unilateral Administrative Order (“UAO”) in December 2003 pursuant to the Comprehensive Environmental Response, Compensation and Liability Act, 42 U.S.C. §9601 (1997) (“CERCLA”), requiring TCML to carry out a “remedial investigation and feasibility study” of the Upper Columbia River in the vicinity of Colville, Washington.

[10]            In July 2004, the Colville Tribes commenced an action under CERCLA against TCML in the U.S. District Court for the Eastern District of Washington.  The Chambers judge below explained:

CERCLA is a strict liability statute that also allows environmentally affected parties to bring private civil actions for relief against environmental harm caused by polluters.  On July 21, 2004, the members of the Confederated Tribes of the Colville Reservation filed the U.S. Environmental Action against TCML in the United States District Court for the Eastern District of Washington pursuant to those private law remedial provisions.

In the U.S. Environmental Action which is formally styled Joseph A. Pakootas, an Individual and Enrolled Member of the Confederated Tribes of the Colville Reservation et al. v. Teck Cominco Metals Ltd., a Canadian corporation, the members of the Colville Tribes seek:

(1)        a declaration that TCML has violated and continues to violate CERCLA;

(2)        an order enforcing the UAO against TCML and compelling TCML to “correct its violations of the UAO”;

(3)        payment of civil penalties for each day of TCML’s alleged violation of the UAO; and

(4)        an award of costs and attorney’s fees incurred by the plaintiffs in the action.  [At paras. 47-8.]

A few months later, the State of Washington also joined as a plaintiff in the CERCLA action, asserting claims similar to those of the other plaintiffs.

[11]            One of TCML’s substantive defences to the action is that it “at all times believed and continues to believe that the barren slag was inert and not harmful to human health or the environment.”  TCML has also vigorously opposed the jurisdiction of the Washington courts over it in the CERCLA action.  As noted by the Chambers judge below (2006 BCSC 1276, para. 50), TCML applied in Washington to have the CERCLA action dismissed on the basis of lack of both personal and subject-matter jurisdiction, arguing that there was no sufficient connection between TCML and the State of Washington to enable the District Court to take jurisdiction.  The Court denied TCML’s motion and the United States Court of Appeals for the Ninth Circuit dismissed TCML’s appeal on July 3, 2006 (see Pakootas et al. v. Teck Cominco Metals Ltd. 452 F.3d 1066), shortly after the Chambers judge below had reserved judgment in the within proceeding in British Columbia.  The Court of Appeals concluded that the CERCLA action did not involve an extra-territorial application of CERCLA because liability “is triggered by an actual or threatened release of hazardous substances”, and the “leaching of hazardous substances from [the slag] … took place in the United States”.  A subsequent petition for the re-hearing of the appeal before the entire Court of Appeals for the Ninth Circuit was unsuccessful.  Mr. Weatherill advised us that TCML is seeking leave to appeal the jurisdictional ruling to the Supreme Court of the United States.

[12]            In the meantime, TCML was able to reach an interim settlement with the EPA with respect to investigative costs relating to the Lake Roosevelt contamination.  Under this arrangement, TCML has caused a subsidiary, Cominco American, to provide certain ongoing funds, secured by the deposit of U.S. $20,000,000 in a U.S. bank and guaranteed by TCML, for the investigation.  As the Chambers judge noted at para. 61 of his reasons, TCML itself has not otherwise attorned to the jurisdiction of the District Court or accepted liability for the investigation costs.  The interim settlement, of course, does not deal with the damages or other amounts for which TCML may eventually be found liable pursuant to CERCLA.

The Insurers

[13]            The plaintiffs Lloyd’s Underwriters are underwriters at Lloyd’s of London and act for themselves and as representatives of various other insurance companies (listed in Schedule A to the Lloyd’s statement of claim) which issued contracts of insurance to Cominco between 1972 and 1985.  Although the first Lloyd’s policy issued to Cominco, in force between August 1972 and August 1975, was placed through the Seattle office of a U.S. broker headquartered in San Francisco, California, all the other Lloyd’s policies issued to Cominco were purchased through insurance brokers in Vancouver.  The Chambers judge noted that those brokers “corresponded with affiliated brokers in London who met with the various underwriters who eventually agreed to subscribe to the various Lloyd’s Policies.”  The premiums payable by Cominco for the policies, and the sums insured, were expressed in Canadian funds.

[14]            The plaintiff Lombard carries on business as an insurance company throughout Canada, including British Columbia.  It is headquartered in Toronto and has a branch office in Vancouver.  It has never carried on business in Washington.  In 1995, it acquired the business of The Continental Insurance Co. of Canada, a Canadian subsidiary of The Continental Corporation, a U.S. insurance company based in Chicago.  Lombard also acquired from another U.S. subsidiary of The Continental Corporation, the Continental Insurance Company (“CIC”), the policies that are at issue in this case.  CIC was registered to carry on business in Washington State.  The policies had been issued to Cominco through dealings with CIC’s Canadian division, which was headquartered in Toronto and had a branch office in Vancouver.  They were assigned to Lombard in accordance with s. 254 of the Insurance Companies Act, S.C. 1991, c. 47, and notice of the assignment was given to Cominco.  Accordingly, when TCML sued in Washington for coverage under these policies, it named Lombard as a defendant.  Later, it also joined CIC as a defendant.

[15]            Lloyd’s joined as defendants in its British Columbia action various other insurance companies that had entered into policies with Cominco, but all except one have settled those insurance claims.  The exception is Seaton Insurance Company (“Seaton”), a Rhode Island corporation that has its principal place of business in Massachusetts.  Seaton’s policies in favour of Cominco were issued when Seaton was known as Unigard Mutual Insurance Company.  It was domiciled in Washington State and had its principal place of business in Seattle, Washington, but evidently is no longer domiciled in that state.  Seaton appeared at the hearing of this appeal and adopted the positions taken by Lloyd’s and Lombard.

[16]            The insurance policies with which we are concerned were described in some detail by the Chambers judge at paras. 21-35 of his reasons.  In general, he noted:

Beginning in 1958 and ending in 1985, TCML purchased the Policies (which are general and excess liability insurance policies) from the Insurers.  It also purchased similar insurance from other insurers.

TCML submits that the intention of its global insurance program was to provide TCML with insurance coverage in any country where liability claims against it might arise, anywhere in the world.

With minor variations, the Policies provide similar coverage, requiring each insurer either to defend and indemnify TCML or to indemnify only (including the reimbursement of defence costs) in case of any alleged liability resulting from an “occurrence” of property damage happening or taking place during the period of coverage anywhere in the world.  Under each of the Policies, the various insurers agreed to cover “all sums” the insured becomes liable to pay as a result of such an “occurrence”.

Significantly, coverage under the Policies is triggered even when a claim is asserted against TCML after the policy period has ended, so long as the claim arose from an “occurrence” in progress during the policy period.  Thus, these coverage disputes relate to Policies that were all purchased by TCML at least twenty and up to thirty five years ago relating to alleged occurrences that allegedly occurred during that same time frame.  [At paras. 21-4; emphasis added.]

He summarized the financial exposure of Lloyd’s, Lombard and Seaton (the “Insurers”) under their policies as follows:

In summary … at issue in these proceedings and in the U.S. Coverage Action are the following worldwide “occurrence” based property damage liability insurance Policies issued to TCML:

(1)        CIC/Lombard Policies: primary coverage of $1 million from August 29, 1969 to September 30, 1976;

(2)        Unigard/Seaton Policies: excess insurance of $5 million over the CIC/Lombard Policies’ primary coverage policies from August 29, 1972 to August 29, 1975, as well as an additional $15 million excess liability coverage from December 12, 1972 to December 12, 1975; and

(3)        Lloyd’s Policies: excess insurance with combined per occurrence limits totalling over $758 million between August 29, 1972 and June 30, 1985.  [At para. 35.]

The policies do not contain choice of law or forum selection clauses.

[17]            TCML takes the position that the alleged contamination of the Upper Columbia River site at issue in the CERCLA action constitutes “property damage” within the meaning of the various policies and that such property damage “occurred” during policy periods covered by those policies.  The full amount of the damages faced by TCML is unknown, but the amounts claimed are expected to exceed the limits under the policies.  The Chambers judge described the Insurers’ position on the coverage issues that have been asserted thus far:

The Insurers deny that they are obligated to compensate TCML under the Policies.  The position advanced by Lloyd’s in its argument is reflective of the position that is taken by all of the Insurers and informs some of the disputes that will ultimately have to be resolved concerning whether Lombard is obligated to provide coverage under the CIC/Lombard Policies for the defence of the U.S. Environmental Action and also whether the Policies obligate any or all of Lloyd’s, Lombard or Seaton to pay any damages or costs for which TCML may ultimately be held liable in that litigation.

My understanding of counsel for Lloyd’s submissions is that for the following reasons (and such other reasons as may become relevant as these issues are further explored), Lloyds will be denying that there is coverage under the Lloyd’s Policies for the environmental claims made by TCML because:

(1)        The pollution and other harm alleged to have been caused by TCML is not unexpected or unintended.  Rather, it is the natural and certain result of mining and smelting operations and was entirely expected by TCML and is therefore not an accident, an occurrence or an insurable risk.

(2)        The Policies contain pollution exclusions.

(3)        Any costs claimed by TCML to remediate or address other injury resulting from operations at Trail (or any other location) are costs associated with the normal business of mining and smelting and are not the proper subject of liability insurance.

(4)        The environmental claims now being advanced in the U.S. Environmental Action do not constitute covered “occurrences” as that term is defined in the Policies.

(5)        Sufficient underlying insurance, deductibles and/or self-insured retentions exist to cover such losses as TCML may incur.

(6)        TCML failed to provide its insurers with timely notice as required by the Policies.

(7)        To the extent TCML is not an insured under the Policies, there has been no valid assignment of the Policies from Cominco Ltd. to TCML.

(8)        Applicable statutes of limitation, the equitable doctrine of laches, and/or the doctrines of waiver and/or estoppel preclude the enforcement of any coverage which might otherwise have been available to TCML under the Policies.

(9)        Some or all of any claimed expenses or costs are not related to “claims” and/or “suits”.

(10)      To the extent that TCML seeks compensation and/or indemnification for losses which have not yet arisen or for amounts TCML has not yet paid or become legally obligated to pay as the result of a final judgment or settlement, such claims are not covered.

(11)      To the extent the damages and/or claims related to the underlying matters were a known loss or loss in progress, there is no coverage.

(12)      Some or all of the claims for which TCML seeks coverage are for equitable and injunctive relief for which no coverage is available under the meaning and intent of the Policies.

(13)      To the extent TCML seeks coverage for restitutionary damages or the costs of undertaking prophylactic measures necessary to prevent further damages associated with the incidents giving rise to environmental claims, there is no coverage under the meaning and intent of the Policies.

(14)      The sums that may be sought by TCML do not constitute sums TCML is obligated to pay as “damages” within the meaning and intent of the Policies.

(15)      The acts or omissions giving rise to the environmental claims were in violation of law, governmental regulations or public policy for which no coverage is available under the meaning and intent of Policies, and/or to the extent TCML seeks indemnification for fines and/or penalties assessed or which may be assessed against TCML, no coverage is available.

(16)      The expenses for which TCML may seek indemnity were incurred to clean up property owned by TCML or in its care, custody or control for which no coverage is available.

(17)      To the extent any of the property damage or bodily injury claimed occurred prior to the inception of, or after the expiration of, the Policies, no coverage is available.

(18)      To the extent TCML seeks indemnity in excess of the limits of liability set forth in the Policies, no coverage is available.

(19)      TCML has voluntarily made payments or assumed obligations or incurred expenses for the environmental claims for which no coverage is available.

(20)      During the process of applying for, negotiating, or renewing the Policies: TCML failed to disclose relevant and material facts which they were obligated to disclose and/or misrepresented relevant and material facts.  [At paras. 64-5.]

Chronology of the Insurance Disputes

[18]            I have already mentioned that TCML gave notice to its insurers of the four claims or potential claims which are the subject of this action, on various dates in 2002 and 2003.  On April 13, 2006, however, TCML informed Lombard that any liability issues relating to the Port McNeill claim had “resolved to TCML’s satisfaction” and that it did not intend to continue with, and irrevocably withdrew, its claims for defence coverage and indemnity in respect that matter.  As well, since no actual claim for indemnity or defence coverage has ever been made in respect of the Pinchi Lake and Vancouver matters, TCML has notified Lloyd’s and Lombard that it:

… has not made and is not now making any claim or request for coverage under any contracts of insurance in its favour issued by [Lloyd’s and Lombard] with respect to any of the ‘Environmental Claims’ referred to in the Statement of Claim, with the exception of claims arising out of the alleged contamination of the Upper Columbia River and Lake Roosevelt in Washington State in respect of which TCML is currently being sued in Washington State.

At the same time, TCML has stated that it is not “abandoning” the remaining claims.

[19]            The only claim referred to in TCML’s action against the Insurers in the State of Washington is the Lake Roosevelt claim – thus the insurance coverage actions in the state and in British Columbia are concurrent but not “parallel”.  The Chambers judge below commented on this point:

… I am satisfied that the Lloyd’s Action and the Lombard Action [in B.C.] raise broader issues of insurance coverage than does the U.S. Coverage Action.  While it is obvious that the Insurers have sought to maximize the import of the claims and potential claims concerning the British Columbia sites and equally obvious that TCML has sought to distance itself from them, the fact remains that issues concerning the British Columbia sites have been and still are the subject of pleadings in these proceedings that have not been fully resolved.  [At para. 76.]

[20]            The circumstances giving rise to the existence of the concurrent proceedings in the two jurisdictions were described by the Chambers judge at paras. 77-91 of his reasons under the heading “The Race to the Courthouse”.  The “race” was triggered by TCML’s termination of “standstill agreements” it had entered into with Lloyd’s and Lombard in 2002.  The termination became effective November 23, 2005.  On that date, TCML commenced an action against the Insurers in the Superior Court of the State of Washington seeking declaratory relief regarding its rights to insurance coverage under the policies.  For its part, Lloyd’s began its coverage action in the Supreme Court of British Columbia, also seeking declaratory relief as to its obligations (or lack thereof) to defend and indemnify TCML in respect of the environmental claims or potential claims, including the CERCLA claim.  Although both coverage actions were begun on the same date, TCML was able to deliver originating process to a judge of the State Superior Court at his home at one second past midnight, whereas Lloyd’s had to wait until the Supreme Court Registry opened approximately nine hours later in Vancouver to begin its action.  Although TCML now disclaims any intention to rely on a “first to file” rule, it did, as the Chambers judge noted at para. 81 of his reasons, seek to rely on such a rule in the Superior Court of Washington.  Lombard began its action in British Columbia, paralleling that of Lloyd’s, on December 1, 2005.

[21]            On November 28, 2005, TCML sought a Temporary Restraining Order and Request of Preliminary Injunction (“TRO motion”) requesting the Washington court to prohibit its insurers from proceeding in any other forum “until this Court can hear the arguments of the parties and determine that Washington is the proper forum to resolve this dispute”.  TCML contended in its legal argument filed with the Washington court that in accordance with the “first to file” rule, the Washington Superior Court had gained jurisdiction and control of all subsequent proceedings when TCML’s complaint was filed; and that therefore “no court of coordinate authority [was] at liberty to interfere with its action”.  (Citing State ex rel. Greenberger v. Superior Court of King County et al., 235 P. 957 (Wash. Sup. Ct. 1925).)

[22]            TCML set down the TRO motion to be heard at 3:00 p.m. on November 30, 2005 in Ephrata, Washington, apparently ex parte.  In support, it filed an argument that suggested the law of British Columbia was "uncertain" on some issues relevant to the case (including the allocation of liability among insurers) and that for that reason and because of the "strong interest" of Washington State in the action "because it involves the protection of Washington's natural environment", the law of Washington "should and will be applied in this case."  However, this argument was never considered by the Washington court, since Lloyd's became aware of the TRO motion and quickly attended before the Supreme Court judge in Chambers in Vancouver.  He granted an order temporarily enjoining TCML from proceeding with the TRO motion in Washington – effectively, an ‘anti‑anti‑suit injunction’.  TCML and Lloyd’s then reached an agreement that, in the words of Davies J. at paras. 84-5, jurisdictional issues would be addressed only on notice to potentially affected parties, and that both TCML and Lloyd’s would file motions in the Washington and British Columbia courts no later than January 31, 2006 “seeking orders both dismissing and/or staying those proceedings by reason of their respective jurisdictional challenges.”

[23]            It was during that interim period that one of the Lloyd’s underwriters was successful in having the U.S. coverage action moved from the Superior Court of Washington to the federal District Court for the Eastern District of Washington.  The Chambers judge said he was satisfied this step “simply constituted the legitimate refusal of a foreign corporation to submit to the jurisdiction of an American state court rather than a United States’ federal court under United States’ law so that there is no merit to TCML’s ‘attornment’ or ‘seeking out’ submissions.”  (At para. 88.)

The Washington Court’s Judgment

[24]            Having been the “first to file” its action, TCML was also successful in being the “first to judgment”: on May 1, 2006, Judge Suko of the U.S. District Court denied the applications of Lloyd’s and Lombard to dismiss TCML’s claims against them both for want of personal jurisdiction and on the basis of forum non conveniens.  I will not review Judge Suko’s reasons in detail, but some description is in order in light of the significance attached by Davies J. to some aspects of them.  Judge Suko began by finding that Washington’s long-arm statute applied to give “specific jurisdiction” over the non-resident defendants because TCML’s cause of action against them arose “from a dispute over insurance contracts which constitute the insurers’ contact with Washington”.  Seven factors relevant to the “reasonableness” of exercising personal jurisdiction were examined, namely:

1.         the extent of defendants’ “purposeful interjection”;

2.         the burden on defendant in defending in the forum;

3.         the extent of conflict with the sovereignty of the defendants’ state;

4.         the forum state’s interest in adjudicating the dispute;

5.         the most efficient judicial resolution to the controversy;

6.         the importance of the forum to plaintiff’s interest in convenient and effective relief; and

7.         the existence of an alternative forum.

The Insurers drew our attention in particular to Judge Suko’s comments concerning items 4 and 6.  With respect to “Washington’s Interest in the Adjudication”, he said:

Washington has a strong and well-established interest in protecting the health and safety of its citizens; to that end, it has a compelling interest in the insurers’ obligation to indemnify, especially, as in this case, at the remediation stage of a polluted site within its borders.  Canron, Inc. v. Federal Ins. Co., 82 Wn. App. 480, 494. 918 P.2d 937 (1996).  For example, Washington State has promulgated regulations that codify this public interest:

There are many insurance coverage disputes involving Washington insureds who face potential liability for their roles at polluted sites in this state.  State and federal mandates exist for cleaning up the environment in order to address the adverse effects of hazardous substances on human health and safety and the environment in general.  It is in the public interest to reduce the costs incurred in connection with the environmental claims and to expedite the resolution of such claims.  The state of Washington has a substantial public interest in the timely, efficient and appropriate resolution of environmental claims involving the liability of insureds at polluted sites in this state.  This interest is based on practices favoring good faith and fair dealing in insurance matters and on the state’s broader health and safety interest in a clean environment.

Washington Administrative Code (WAC) § 284-30-900 (2005).

Because indemnification for clean-up costs at Lake Roosevelt is at issue, Washington’s interest in this adjudication is compelling.  Canron, 82 Wn.App. at 494.  Further, the Ninth Circuit, in assessing the reasonableness of personal jurisdiction over a foreign insurer in an insurance contract dispute case, has found a third party beneficiary of an insurance policy has a great interest in the outcome of insurance coverage litigation.  As a potential third party beneficiary of TCML’s insurance coverage in the CERCLA Action, Washington has an obvious interest in the outcome; this interest weighs heavily in favor of personal jurisdiction.  [Emphasis added.]

On the topic of the plaintiff’s interest in obtaining relief, the judge also noted:

TCML is not a citizen of the U.S. and is not domiciled in Washington.  However, TCML states it prefers the Washington forum where insurance coverage law is more developed and trying the case would not result in piecemeal litigation of the CERCLA Action and insurance coverage.  TCML asserts the case should be tried in the same forum for consistency of factual and legal bases.  [Emphasis added.]

(The reference to "more developed coverage law" would appear to be in accord with the affidavit of Mr. Zariski, referred to at para. 46 below.)

[25]            Based on his consideration of all seven factors, Judge Suko concluded that Lloyd’s and Lombard had not “controverted” the alleged facts sufficiently to warrant dismissal of the action and that “[t]he ‘reasonableness’ considerations of judicial efficiency in resolution of the dispute, Washington’s strong interest in the insurance coverage litigation, and TCML’s interest in avoiding duplicate adjudication and inconsistent findings of fact [weighed] strongly in [favour] of [the Washington] court’s exercise of specific jurisdiction.”

[26]            Judge Suko then moved to consider the Insurers’ motions to dismiss the Washington coverage action on the basis of forum non conveniens.  To prevail on this point, he said, the applicants were required to show that (1) an adequate alternative forum exists; (2) the balance of public and private factors favours dismissal; and (3) these showings outweigh the deference shown to plaintiff’s choice of forum.  (Citing Lockman Found. v. Evangelical Alliance Mission 930 F.2d 764 at 767 (9th Cir. 1991).)  A plaintiff’s choice of forum would not be disturbed “unless the private and public interest factors [favour] dismissal”, he said, citing Lueck v. Sundstrand Corp. 236 F.3d 1137 at 1145 (9th Cir. 2001).  The relevant “private interest factors” were the residence of the parties and witnesses; the convenience of the forum to the parties; access to witnesses, documents and physical evidence; whether unwilling witnesses could be compelled to testify; the cost of bringing witnesses to trial; the enforceability of any judgment; and all other practical factors – i.e., considerations similar to those traditionally considered by Canadian courts.  (See e.g., Stern v. Dove Audio, Inc. [1994] B.C.J. No. 863, at para. 62, per Low J. (as he then was).)

[27]            Judge Suko concluded that the “alternative forum” (i.e., British Columbia) did not offer a more convenient location in terms of parties, witnesses, and documents, and that Washington was not “overly burdensome” for any of the parties.  With respect to compelling the attendance of witnesses, since the Insurers could not compel or facilitate the presence of U.S. regulatory personnel in British Columbia, that factor weighed against dismissal.  The Court declined to give significant weight to the factor of enforceability in light of the parties’ “differing opinions” regarding the enforceability of a foreign judgment in British Columbia.

[28]            The Court then turned to the applicable “public interest factors”, namely, local interest in the lawsuit; the Court’s familiarity with governing law; the burden on local courts and juries; congestion in the courts; and the cost of resolving disputes unrelated to the forum, again citing Lueck, supra, at 1147.  Judge Suko again noted Washington State’s “very compelling interest in the coverage dispute” as evidenced by policy statements in administrative regulations, the previous decision of the Court of Appeals of Washington in Canron, Inc. v. Federal Ins. Co., 918 P.2d 937 (1996), and the amount of local media coverage given to TCML’s alleged role in the contamination of Lake Roosevelt.  In contrast, the foreign Insurers had not presented any evidence indicating a similar degree of interest in British Columbia.  On the issue of the Court’s familiarity with governing law, Judge Suko continued:

… TCML contends a choice of law determination should be made.  However, the requirement of a choice of law analysis applies only to cases involving statutes that require venue in the United States, such as the Jones Act, or the Federal Employees’ Liability Act.  Lueck, 236 F.3d at 1148.  Where no such statute applies, “the choice of law determination is given much less deference on a forum non conveniens inquiry.”  Id.  Washington courts have also held a choice of law analysis is not necessary in a forum non conveniens determination.  Hill v. Jawanda Transport Ltd., 96 Wn.App 537, 546 (1999).  It should be emphasized that the law to be applied related to insurance contract interpretation, not environmental law or CERCLA.  As this case does not involve a U.S. statute requiring venue within the U.S., the court declines to engage in a lengthy choice of law analysis at this time.  [Emphasis added.]

He did not reach a conclusion, tentative or otherwise, as to what law the Court would apply to the policies.

[29]            Finally, with respect to the “deference” factor (i.e., deference to the plaintiff’s choice of forum), Judge Suko noted that the presumption in favour of the plaintiff’s choice applies with less force where the plaintiff is a foreign citizen, but that “less deference” did not mean “no deference”, citing Ravelo Monegro v. Rosa 211 F.3d 509 at 514 (9th Cir. 2000).

[30]            Together, the public and private interest factors led the Court to refuse the Insurers’ motion to decline jurisdiction on the basis of forum non conveniens.  In Judge Suko’s words:

… The distance between Eastern Washington and Vancouver, B.C. is negligible, and the Trail facility and Lake Roosevelt lie almost equidistant from the competing forums.  The burden on [Lloyd’s], Lombard and Seaton to litigate this matter in Eastern Washington does not rise to the level of oppressive or vexatious.  The court finds no logistical, administrative or legal problems compelling enough to upset TCML’s choice of forum.

[31]            Three days after the Washington court’s reasons were released, Mr. Justice Davies began hearing TCML’s stay applications in the Supreme Court of British Columbia.  For reasons dated August 21, 2006 and indexed as 2006 BCSC 1276, he declined to stay the coverage actions brought by Lloyd’s and Lombard in British Columbia.  It is from that ruling that TCML appeals to this Court.

The British Columbia Judgment

[32]            As the Chambers judge noted at para. 92 of his reasons, TCML did not contest that the Supreme Court of British Columbia had jurisdiction simpliciter over it and the subject insurance policies.  Its primary position was that as a matter of comity, the Court should decline jurisdiction over the coverage action in favour of the Washington court.  The Insurers responded that the more appropriate forum was clearly British Columbia, and that TCML was engaging in “forum shopping” in an attempt to avoid the interpretation of the policies in accordance with British Columbia law.

[33]            Davies J. began his analysis by referring to the Court Jurisdiction and Proceedings Transfer Act (“CJPTA”) which came into force on May 4, 2006.  Section 11 of the Act provides:

11(1)    After considering the interests of the parties to a proceeding and the ends of justice, a court may decline to exercise its territorial competence in the proceeding on the ground that a court of another state is a more appropriate forum in which to hear the proceeding.

(2)        A court, in deciding the question of whether it or a court outside British Columbia is the more appropriate forum in which to hear a proceeding, must consider the circumstances relevant to the proceeding, including

(a)        the comparative convenience and expense for the parties to the proceeding and for their witnesses, in litigating in the court or in any alternative forum,

(b)        the law to be applied to issues in the proceeding,

(c)        the desirability of avoiding multiplicity of legal proceedings,

(d)        the desirability of avoiding conflicting decisions in different courts,

(e)        the enforcement of an eventual judgment, and

(f)         the fair and efficient working of the Canadian legal system as a whole.

[34]            The Chambers judge noted that the model Court Jurisdiction and Proceedings Transfer Act proposed by the Uniform Law Conference of Canada in 1994 had not been adopted by the Province of British Columbia until May 2003 and did not come into force until May 2006.  Thus, he said, during the interim of more than ten years, the courts of British Columbia and other jurisdictions had “considered and applied the principles enunciated in Morguard and Amchem to a myriad of factual circumstances.  Following the usual course of the common law, those principles evolved and developed over time.”  (Para. 99.)  As well, he noted s. 2(2) of the Act, which states that a court’s territorial competence (a term he had already equated to jurisdiction simpliciter) is to be determined “solely” by reference to Part 2 of the Act.  Section 11 is included in that Part.  Davies J. then continued:

I am accordingly satisfied that the provisions of s. 11 of the CJPTA should be considered to be part of a comprehensive remedial statutory scheme that is intended to codify the determination of jurisdictional issues in British Columbia.  As such the provisions of s. 11 should be interpreted as being informed by but neither dictated nor constrained by the existing case law.  Expression must be given to the statutory provisions by means of a fair, large and liberal construction and interpretation that best ensures that the objects of the legislation are attained.  [At para. 102; emphasis added.]

[35]            The Chambers judge quoted an excerpt from the comments of the Attorney General when introducing the Act in the Legislative Assembly in 2003, who stated that its purpose was “… to establish clear and harmonized statutory rules to accord with the principle enunciated by the Supreme Court of Canada respecting the basis upon which a court in a province … may properly hear and determine a matter upon which its decision is sought.”  Davies J. said he intended to consider the parties’ arguments regarding forum shopping and comity that were “at the heart of this application” by reference to the decided case law prior to the enactment of the Act, determine whether and to what extent s. 11 affected those submissions or any other aspects of the common law doctrine of forum non conveniens, and apply the provisions of s. 11 “as informed by that analysis” in deciding whether the Court should decline to exercise its “acknowledged territorial competence” in favour of the Washington court as a more appropriate forum.  (Para. 105.)  He noted the similarity between the factors considered at common law in a forum non conveniens analysis, as set out in cases such as Stern, supra, and the factors listed in s. 11(2) of the CJPTA.  In his words, “While the statutory considerations listed in s. 11(2) are cast in different language and enumerate fewer categories than those suggested by Stern, the content is in many respects the same.”

[36]            The Chambers judge then embarked on a consideration of the factors listed in s. 11(2) and their respective counterparts normally considered at common law.

Section 11(2)(a): The comparative convenience and expense for the parties to the proceeding and for their witnesses, in litigating in the court or in any alternative forum:

[37]            The Chambers judge found that apart from the existence of the CERCLA action “as a proceeding requiring the attendance of TCML in the District Court in Washington State”, all the circumstances to be considered under this rubric pointed in favour of British Columbia as a more convenient forum for the resolution of the insurance issues.  (Para. 111.)  He noted TCML’s argument that various scientists and regulatory officials in Washington might be required to testify as to the (alleged) property damage in Washington State, when it occurred, whether it was “expected or intended”, whether it was a “known loss or loss in progress”, whether the acts or omissions alleged against TCML were in violation of any law of Washington State, etc.  As well, TCML argued that the witnesses likely to be called by Lloyd’s concerning the decision to accept risk under Lloyd’s policies were more likely to reside in England than in British Columbia, and that witnesses for Lombard and Seaton were also more likely to live outside the Province.  Further, while TCML had committed to make all necessary witnesses subject to its control available in Washington State, it had no control over potential Washington State witnesses (including the CERCLA plaintiffs and Washington officials), who could not be compelled to appear in British Columbia. 

[38]            The Chambers judge agreed with the Insurers’ argument, however, that the issues that would involve Washington witnesses would be peripheral to the central questions of disclosure, risk assessment and policy interpretation, all of which tended to emphasize the importance of witnesses in British Columbia or elsewhere.  (Para. 125.)  He found that the overall cost of litigating the insurance issues in Washington would be greater than in British Columbia.  In summary, he said:

… I am satisfied that the convenience and expense of the parties and potential witnesses are factors which favour litigation in British Columbia rather than Washington State.  I reach that conclusion primarily because of: the residence of the parties; where each carries on business; and the fact that, in my view, these proceedings and the U.S. Coverage Action will ultimately be far more concerned with pre-contractual disclosure, risk assessment and interpretation issues related to coverage and exclusions (all of which have little, if any, connection to Washington State or its residents) than with the scientific issues identified by TCML which will largely, if not entirely, be resolved in the U.S. Environmental Action.

In addition, these proceedings are also concerned with coverage issues related to the Pinchi Lake, Port McNeill and Granville Street claims or potential claims in British Columbia as pleaded by the insurers all of which have no connection to Washington State or Washington State witnesses.  [At paras. 128-9.]

Section 11(2)(b): The law to be applied to issues in the proceeding:

[39]            The Chambers judge was of the view that among the common law factors referred to in Stern, the place where the cause of action arose, the place where the loss or damage occurred, the applicable substantive law, and considerations related to the difficulty and cost of proving foreign law if necessary, could be conveniently considered under s. 11(2)(b), although they might also have been appropriately addressed under s. 11(2)(c) or (d).  (Para. 133.)  He noted that the Washington court had not embarked on a “detailed analysis” of the law that would likely be applied in its determination of the coverage issues.  Although Davies J. agreed that in many cases the question of choice of law cannot be finally resolved at the outset of the litigation, “there are also some situations in which the question of the law that will likely be applied is so central to the issues … that a detailed ‘proper law’ analysis is not only appropriate but necessary.”  (Para. 137.)  In this instance, the question was significant, he said, for two reasons:

(1)        TCML admittedly seeks to secure what it considers will be an interpretation of the Policies in Washington State that is more favourable to TCML’s interests than it would likely obtain in British Columbia.

(2)        The Insurers submit that not only must the laws of British Columbia govern the interpretation of the Policies but that the application of British Columbia law will ensure that they are availed of the juridical advantages which they contemplated when agreeing to insure TCML for losses covered by the Policies.  [At para. 139.]

[40]            The Chambers judge noted TCML’s argument, which Judge Suko had adopted, that some or all of the policies included terms that contemplated that the Insurers could be “haled into the courts of the United States” on coverage claims.  After quoting (at para. 142) the relevant terms from the policies, Davies J. did not agree that they should be as broadly construed as TCML asserted.  Although the clauses in question indicated an acceptance of the jurisdiction of U.S. courts (among others) for purposes of the policies, that acceptance did not extend to the choice of the law to be applied in construing the policies.  Further, he disagreed with TCML’s argument that one of the Lloyd’s policies, in existence for two years in the mid-seventies, had contained a clause under which Lloyd’s submitted to the jurisdiction of any competent court in the United States and agreed that all matters arising under the policy would be determined “in accordance with the law and practice of such Court”.  Davies J. found that that particular policy had been issued in error and had been cancelled “back to its inception” and replaced with another certificate of insurance, which contained no similar ‘service of suit’ clause.  (Para. 150.)  Thus he said:

For the purposes of this application, I am satisfied that to the extent TCML seeks to rely on the United States service of suit clause contained in the certificate No. LO 88954 that was cancelled as of its inception date, such reliance is wholly unfounded.

It follows that, to the extent that Judge Suko may have relied upon any United States service of suit clause or clauses alleged by TCML in dismissing Lloyd’s forum non conveniens application, his decision was based upon a factual foundation that is entirely at odds with the facts established by the evidence adduced before me.

I also find it noteworthy that while there are no express choices of law clauses in any of the Lloyd’s Policies issued to TCML, many (but not all) do include service of suit clauses which provide:

It is agreed that in the event of the failure of the Underwriters hereon to pay any amount claimed to be due hereunder, Underwriters hereon, at the request of the Assured, will submit to the jurisdiction of any Court of competent jurisdiction within Canada and will comply with all requirements necessary to give such Court jurisdiction and all matters arising hereunder shall be determined in accordance with the law and practice of such Court.  [At paras. 151-3; emphasis added.]

[41]            The Chambers judge described various arguments asserted by TCML in favour of the selection of the U.S. District Court as the “arbiter of coverage issues under the Policies” and the Insurers’ arguments in response.  At para. 156, he set out his reasons for concluding that “factors related to the appropriate law to be applied in these proceedings and in the U.S. Coverage Action strongly [favour] British Columbia as a more appropriate forum than Washington State.”  These reasons are significant enough for purposes of this appeal to reproduce in full:

(1)        My consideration of the totality of the evidence on this application and the submissions of counsel leads me to conclude that it is highly probable that the court ultimately determining the issues amongst the parties will be required to apply the laws of British Columbia to the interpretation of the Policies.  I reach that conclusion primarily by reference to the provisions of the Insurance Act.  Although those provisions are not absolutely determinative, there is little, if any, evidence to suggest that they should not be applied in this case especially in regard to the Lloyd’s Policies which are monetarily by far the most significant of the Policies at issue.  My consideration of the principles and factors enunciated in [Imperial Life Assurance Co. v. Colmenares [1967] S.C.R. 443] leads me to the same conclusion.

(2)        Given that under the usual choice of law principles and the provisions of the Insurance Act, it is highly likely that the proper law of the Policies will be British Columbia law and there is little likelihood that it will be necessary to prove foreign law in this Court.

(3)        Perhaps more importantly for the purposes of this application, I am satisfied that it would be unreasonable to apply the laws of Washington State to the interpretation of the Policies.  I say that because the only connection of Washington State to the Policies is the fact that it and some of its residents are potentially the victims of TCML’s alleged wrongful actions all of which occurred solely in Canada.  Those Washington residents are not, however, beneficiaries of the Policies.  They have no privity of contract with any of the Insurers whose sole obligations under the Policies are to TCML.  Also, while Seaton’s head office is in Seattle, I am not satisfied that that single factor overrides all other considerations which establish a far more real and substantial connection of the Unigard/Seaton Policies to British Columbia.

(4)        It is neither possible nor appropriate to reach a final conclusion concerning the issues raised by both Lombard and TCML concerning the effect of the transfer of the CIC/Lombard Policies; however, the recent filings by CIC in the U.S. Coverage Action and the written submissions made by Lombard in these proceedings relating to s. 254 of the Insurance Companies Act, S.C. 1991, c. 47, when read in conjunction with s. 5 of the British Columbia Insurance Act, satisfy me for the purposes of this application that the fact that CIC is not a party in these proceedings does not weigh in favour of this Court’s declining jurisdiction over Lombard and the CIC/Lombard Policies.  It must be recalled that TCML itself initially gave notice of the claims that are the subject of the U.S. Environmental Action to Lombard rather than CIC.

(5)        While Seaton is not yet a plaintiff in a British Columbia proceeding, as a defendant in the Lloyd’s action it has advanced defences against TCML raising the same issues as those affecting it in the U.S. Coverage Action.

(6)        While TCML focuses on the existence of the U.S. Environmental Action as the basis upon which to avail itself of the hearing of insurance coverage issues in the state of Washington, these proceedings also involve declarations in relation to potentially environmentally affected sites in British Columbia which have no connection whatsoever to Washington State.

(7)        There is, in my view, substantial merit to the submissions of Lloyd’s and Lombard that their obligations under worldwide policies of insurance should, in the absence of any indication of an intention to split the proper law of their respective policies, be subject to only one proper law of the contract which should be determined on the basis of the usual choice of law principles applicable to contracts of insurance.  To determine otherwise could result in the interpretation of policies of insurance by various foreign courts with no connection to an alleged tortfeasor other than a perceived interest in the fruits of its insurance, with each court applying local law to obtain redress for its citizens, a situation never contemplated by either of the parties to the insurance contract.  [At para. 156; emphasis added.]

(I note parenthetically that s. 5 of the Insurance Act, R.S.B.C. 1996, c. 226, referred to by the Chambers judge at para. (4) above, provides that a contract of insurance is "deemed to have been made in British Columbia and must be construed accordingly" if it insures a person resident in the Province at the date of the policy or "has as its subject matter property … located in British Columbia.")

Section 11(2)(c): The desirability of avoiding multiplicity of legal proceedings; and Section 11(2)(d): The desirability of avoiding conflicting decisions in different courts:

[42]            The Chambers judge observed that these two factors “broadly encompass concerns relating to judicial comity as expressed in Morguard and Amchem”, and that they were also considered at common law in connection with the existence of parallel proceedings, the legitimate juridical advantages or disadvantages to the parties in the competing jurisdictions and the discouragement of forum shopping.  (Para. 157.)  It was under this rubric that TCML contended that the Washington court’s assertion of jurisdiction over the coverage action was “virtually conclusive” of the issues before the Chambers judge, citing 472900 B.C. Ltd. v. Thrifty Canada Ltd. (1998) 57 B.C.L.R. (3d) 332, Westec Aerospace Inc. v. Raytheon Aircraft Co. (1999) 67 B.C.L.R. (3d) 278, 1999 BCCA 243,  (app. dism’d without a hearing on the merits [2001] 1 S.C.R. iv, 2001 SCC 26), and Ingenium Technologies Corp. v. McGraw-Hill Companies, Inc. (2005) 49 B.C.L.R. (4th) 120, 2005 BCCA 358, all decisions of this court.  TCML had also advised Judge Suko in oral submissions that “… it is highly likely … that the British Columbia court, as a matter of how they handle such situations, will stand down and allow the action to go forward here.” 

[43]            The Chambers judge found that these arguments both ‘over-simplified and distorted’ the inquiry that a British Columbia court must undertake before determining issues of international judicial comity.  (Para. 163.)  In his view, none of Thrifty, Westec or Ingenium established “the single bright line rule” asserted by TCML, and in another case, Western Union Insurance Co. v. Re-Con Building Products Inc. (2001) 95 B.C.L.R. (3d) 253, 2001 BCCA 513 (lve. to app. refused, (2002) N.R. 398 n.), the Court had expressly rejected the assertion of jurisdiction by a court in California as conclusive of a forum non conveniens argument.  Giving the principle of comity the level of importance suggested by TCML would, the Chambers judge stated, “erroneously diminish to the point of virtual extinction” the many other factors relevant to any forum non conveniens analysis, whether as described by the Court in Amchem or as “now statutorily modified and mandated by s. 11 of the CJPTA.” (Para. 181.)

[44]            Proceeding on the assumption that the Washington court’s assertion of jurisdiction was a “very significant” factor, but not conclusive or virtually conclusive, the Chambers judge noted that in applying forum non conveniens principles, Judge Suko had considered deference to the plaintiff’s choice of forum and “nationalistic concerns that were largely rejected in Amchem (and also by the CJPTA)” in favour of international comity.  (Para. 186.)  This fact was significant, Davies J. said, not because British Columbia law should be considered more enlightened than the law of Washington, but because of TCML’s contention that “the assertion of jurisdiction by the District Court precludes a finding that Washington State is not an appropriate forum or that TCML has engaged in forum shopping.”  (Para. 187; my emphasis.)  He quoted s. 11(1) of the CJPTA, emphasizing that a British Columbia court may decline jurisdiction where another forum is more appropriate.  In his words:

To give effect to that provision in circumstances where this Court has admitted territorial jurisdiction over both TCML and the subject matter of the Policies, TCML must establish, based upon a consideration of all of the factors listed in s. 11 (and any other factors that may be relevant and significant), that the [U.S.] District Court is not only an appropriate forum but also the more appropriate forum for the determination of the issues in this litigation.  [At para. 189.]

[45]            The Chambers judge found it significant that he had had before him a different evidentiary record than that placed before the Washington court; that he had undertaken an analysis of what law was likely to be properly applied to the insurance coverage dispute, while the Washington court had declined to do so; that in addition to the Lake Roosevelt claim, the British Columbia proceedings involved three other potential claims by TCML against the Insurers, all involving conduct and alleged damage in British Columbia; that the dispute was likely to involve not only the issues of contract interpretation, dates of occurrence and remediation costs that were emphasized by TCML, but factual issues regarding Cominco’s knowledge of possible downstream contamination and the times at which it purchased the subject  insurance policies; and that although the existence of the CERCLA action provided a “connection” between TCML and the State of Washington, the same connection could exist with any other jurisdiction in which environmental damage was alleged to have occurred as a result of TCML’s conduct.  As well, it was clear that the Washington court had been concerned with the State’s “strong interest in the insurance coverage litigation and TCML’s interest in avoiding duplicate adjudication and inconsistent findings of fact.”  (Para. 191.)  This approach could be contrasted with the “balancing” of the legitimate expectations of all parties which the principle of comity, as described in Amchem, involved.

[46]            The Chambers judge then turned to the issue of “legitimate expectations” and judicial advantage specifically.  Since TCML had always asserted, and continued to assert, that the District Court had no jurisdiction over it in the CERCLA action, he was satisfied that the only reason it sought to have the same court adjudicate the insurance coverage issues was that “… it perceives that it will obtain a more generous interpretation of the Policies in Washington State than would be the case in British Columbia.  It is to that end that TCML has advanced its submission alleging that Washington State law should apply because the law of British Columbia on environmental insurance issues is ‘unsettled or non-existent’”.  (Para. 194.)  The phrase “unsettled or non-existent” was taken from an affidavit of Mr. Zariski, a Washington attorney representing TCML.  He described four significant features of the law of Washington relating to environmental insurance coverage.  If the affidavit is correct, these features are significantly advantageous to an insured and disadvantageous to an insurer, as opposed to what one might expect under British Columbia law.  Perhaps most notably, where several insurers are “on risk”, Washington courts do not allocate liability among them, but impose joint and several liability for the entire amount of damages up to their respective policy limits.  Mr. Zariski also opined that if the Washington court determined either that British Columbia law on the issues was “unsettled or non-existent, or that the law or interests of the Province of British Columbia do not materially differ from the law or interests of the State of Washington …, it will apply Washington law as the presumptive law of the forum.”  In support, Mr. Zariski cited several U.S. cases in which Washington courts applied the lex fori where a party had failed to establish "with reasonable certainty" the law of the foreign jurisdiction that would otherwise have applied.

[47]            In accordance with the Court’s comments at paras. 32 and 55 of Amchem regarding ‘forum shopping’, the Chambers judge reasoned that TCML’s pursuit of the potential advantages posed by Washington law could have some validity if it could have reasonably contemplated that “coverage issues related to the potential applicability of United States environmental legislation to policies of insurance were entered into in British Columbia by a Canadian company would fall to be interpreted by application of the laws of Washington State.”  (At para. 195.)  In his opinion, any such finding was precluded by TCML’s denial of the extra-territorial application of CERCLA to its Trail operations, a position that TCML had taken consistently (and as mentioned earlier, hopes to take to the Supreme Court of the United States).

[48]            Finally under this rubric, the Chambers judge at paras. 196-7 characterized TCML’s submissions in favour of the Washington court as an appropriate forum, as an attempt to manipulate the existence of the CERCLA litigation “by conflating potential liability issues under CERCLA with coverage issues under the Policies.”  He continued:

I consider it to be a transparent attempt to defeat the Insurers’ legitimate expectations that British Columbia law would govern their obligations under the Policies and would coincidentally serve the purpose of garnering to TCML benefits to which it would not otherwise be entitled. Those benefits would include an “all sums” allocation of insurance proceeds under Washington State law as opposed to “pro rata recovery” from individual Insurers under British Columbia law if coverage under the Policies for alleged environmental losses is found to exist.  [At para. 196.]

He concluded that TCML had engaged in “inappropriate forum shopping” in an attempt to preclude litigation in British Columbia, the jurisdiction with which TCML itself and the policies were most closely connected.  (Para. 197.)

Section 11(2)(e): The enforcement of an eventual judgment:

[49]            The Chambers judge had no concerns about the enforceability in British Columbia of any judgment of the Washington District Court awarding damages to TCML against Lombard.  (Damages were not sought in Washington against Lloyd’s.)  The enforcement of any declaratory judgment could be more problematic, given the historic position of Canadian courts not to enforce foreign judgments other than those for a sum of money.  That rule had been weakened in recent years, Davies J. noted, and had been recently raised  for reconsideration by the Ontario Court of Appeal in Pro Swing Inc. v. ELTA Golf Inc. (2004) 71 O.R. (3d) 566.  (The judgment of the Supreme Court of Canada in that case had not been issued by the time of the Chambers judge’s reasons; see now 2006 SCC 52.)

[50]            Issues relating to registration of judgments under Part 2 of the Court Order Enforcement Act, R.S.B.C. 1996, c. 78 were more unclear.  As a federal court, the Washington District Court was likely not a “court in a reciprocating state” that could issue orders for registration under the statute: see Murphy v. Wynne 2002 BCSC 1835.  Thus it was arguable at least in theory that the potential unenforceability of a judgment of the Washington court in British Columbia would weigh against the British Columbia court’s granting a stay in these proceedings.  As a practical matter, however, the Chambers judge said it was unlikely TCML would have to resort to execution proceedings to obtain satisfaction from its Insurers in the event of a favourable coverage ruling, either in the British Columbia court or the U.S. District Court.  Overall, he was satisfied that issues regarding enforceability should have no bearing on the exercise of his discretion in this case.  (I note that this part of Davies J.’s discussion must have assumed the non-existence of a contrary, or different, judgment from a British Columbia court.  The existence of inconsistent judgments in the two jurisdictions could well pose difficulties or at least uncertainty regarding enforceability: see para. 80 below.)

Section 11(2)(f): The fair and efficient working of the Canadian legal system as a whole:

[51]            The Chambers judge regarded this factor as less important in an international (as opposed to an inter-provincial) jurisdictional dispute, but again suggested that “it would not be efficient to have contracts of insurance that may have application in more than one jurisdiction interpreted in accordance with more than one system of law unless the particular policy under consideration expressly provides otherwise.”  (Para. 215.)

The Chambers Judge’s Conclusions

[52]            At the end of his long and detailed reasons, Davies J. ruled that he would not grant the stays sought by TCML - i.e., that he would not decline to exercise the jurisdiction of the Supreme Court of British Columbia over the coverage action on the ground the U.S. District Court was the more appropriate forum.  This was, he said, based on his detailed consideration of all the factors referred to above, and more importantly, on the finding of “inappropriate forum shopping” by TCML in attempting to avoid the forum most closely connected with TCML and the policies, in favour of a jurisdiction with “at best a tenuous connection” to them.  Finally, he described TCML’s assertion (before Judge Suko) that the law of Washington State was “more developed”, as a “thinly disguised ruse” to have the insurance policies interpreted under a system of law that “none of the parties to the Policies ever intended or ever expected would govern that interpretation” – a conclusion reinforced by TCML’s denial that it should be subject to the jurisdiction of the Washington court in the CERCLA action.  (Para. 218.)


[53]            In this court, TCML advanced four inter-related grounds of appeal, namely that the Chambers judge erred in:

(a)        determining that the interpretation of s. 11 of the CJPTA was “neither dictated nor constrained by the existing case law”;

(b)        failing to respect the principle of comity as he was required to do in accordance with Amchem, Thrifty, Westec and Ingenium;

(c)        finding that TCML had engaged in “inappropriate forum shopping”; and

(d)        using the CJPTA to adopt a “parochial approach” to issues of jurisdiction that has been consistently rejected by the Supreme Court of Canada and this Court for several years.

Overall, TCML also submits that the Chambers judge showed no respect for the U.S. District Court and its jurisdictional laws, and that for “no good reason” he placed the parties in the “unenviable position” of having parallel legal proceedings in two jurisdictions over the same subject-matter.

CJPTA and Existing Case Law

[54]            The new CJPTA obviously has potential relevance to all the grounds of appeal.  As mentioned earlier, the Act has its genesis in the Uniform Law Conference of Canada’s recommended statute, published in 1994.  (See Proceedings of the 75th Annual Meeting of the Uniform Law Conference of Canada, August 1993).  The Commentary of the Conference stated that the Uniform Act had four main purposes:

(i)         to replace the widely different jurisdictional rules currently used in Canadian courts with a uniform set of standards for determining jurisdiction;

(ii)        to bring Canadian jurisdictional rules into line with the principles laid down by the Supreme Court of Canada in Morguard and Amchem, supra;

(iii)       by providing uniform jurisdictional standards, to provide an essential complement to the rule of nation-wide enforceability of judgments in the Uniform Enforcement of Canadian Judgments Act; and

(iv)       to provide, for the first time, a mechanism by which the Superior courts of Canada can transfer litigation to a more appropriate forum in or outside Canada, if the receiving court accepts such a transfer.

The Commentary continued:

To achieve the first three purposes, this Act would, for the first time in common law Canada, give the substantive rules of jurisdiction an express statutory forum instead of leaving them implicit in each province’s rules for service of process.  In the vast majority of cases this Act would give the same result as existing law, but the principles are expressed in different terms.  Jurisdiction is not established by the availability of service of process, but by the existence of defined connections between the territory or legal system of the enacting jurisdiction, and a party to the proceeding or the facts on which the proceeding is based.  The term “territorial competence” has been chosen to refer to this aspect of jurisdiction … and distinguish it from other jurisdictional rules relating to the subject-matter or other factors.  [Emphasis added.]

[55]            It is apparent, then, that the CJPTA is not intended simply as a restatement or codification of existing law but does effect some substantive (albeit not dramatic) changes in the previous law.  Perhaps most notably, the Rules of Court no longer determine questions of service and jurisdiction simpliciter, or “territorial competence”.  A comparison of  Supreme Court Rule 13 with s. 10 of the Act shows that some of the longstanding criteria (e.g., defendant’s domicile in the province) have given way to the more flexible and practical test of “real and substantial connection”.  The Act also introduces the concept of transferring a proceeding to or from a foreign court.  With respect to forum conveniens, on the other hand, the Act seems intended to codify, rather than effect substantive changes to, the previous law.  The court is still required to consider “the circumstances relevant to the proceeding” and the list of six factors in s. 11(2) is obviously not exhaustive.  The factors are broadly phrased and, as the Chambers judge has shown in this instance, can be generally equated to various factors considered at common law, as described in Stern, supra.

[56]            Section 11(1) requires the court to consider whether the court of another state is “a more appropriate forum” for the proceedings and s. 11(2) refers to whether the domestic or foreign court is “the more appropriate forum”.  This conforms broadly to the common law of forum non conveniens as adopted (from Scottish law) by the House of Lords in the seminal 1986 decision of Spiliada Maritime Corp. v. Cansulex Ltd. [1986] 3 All E.R. 843.  There, their Lordships reviewed the Scottish authorities and summarized the law with respect to the granting of a stay on that ground as follows:

(a)        The basic principle 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 interests of all the parties and the ends of justice.

(b)        As Lord Kinnear’s formulation of the principle indicates, in general the burden of proof rests on the defendant to persuade the court to exercise its discretion to grant a stay ….  It is, however, of importance to remember that each party will seek to establish the existence of certain matters which will assist him in persuading the court to exercise its discretion in his favour, and that in respect of any such matter the evidential burden will rest on the party who asserts its existence.  Furthermore, if the court is satisfied that there is another available forum which is prima facie the appropriate forum for the trial of the action, the burden will then shift to the plaintiff to show that there are special circumstances by reason of which justice requires that the trial should nevertheless take place in this country (see para. (f) below).

(c)        The question being whether there is some other forum which is the appropriate forum for the trial of the action, it is pertinent to ask whether the fact that the plaintiff has, ex hypothesi, founded jurisdiction as of right in accordance with the law of this country, of itself gives the plaintiff an advantage in the sense that the English court will not lightly disturb jurisdiction so established …

… In my opinion, the burden resting on the defendant is not just to show that England is not the natural or appropriate forum for the trial, but to establish that there is another available forum which is clearly or distinctly more appropriate than the English forum.  In this way, proper regard is paid to the fact that that jurisdiction has been founded in England as of right …; and there is the further advantage that on a subject where comity is of importance it appears that there will be a broad consensus among major common law jurisdictions. …

(d)        Since the question is whether there exists some other forum which is clearly more appropriate for the trial of the action, the court will look first to see what factors there are which point in the direction of another forum.  These are the factors which Lord Diplock described, in MacShannon’s case [1978] 1 All E.R. 625 at 630 …, as indicating that justice can be done in the other forum at ‘substantially less inconvenience or expense’.  Having regard to the anxiety expressed in your Lordships’ House in the Société du Gaz case (1926) S.C. (H.L.) 13 concerning the use of the word “convenience” in this context, I respectfully consider that it may be more desirable, now that the English and Scottish principles are regarded as being the same, to adopt the expression used by Lord Keith in The Abidin Daver [1984] 1 All E.R. 470 at 479 … when he referred to the ‘natural forum’ as being that “with which the action has the most real and substantial connection”.  So it is for connecting factors in this sense that the court must first look; and these will include not only factors affecting convenience or expense (such as availability of witnesses), but also other factors such as the law governing the relevant transaction … and the places where the parties respectively reside or carry on business.

(e)        If the court concludes at that stage that there is no other available forum which is clearly more appropriate for the trial of the action, it will ordinarily refuse a stay. …

(f)         If, however, the court concludes at that stage that there is some other available forum which prima facie is clearly more appropriate for the trial of the action, it will ordinarily grant a stay unless there are circumstances by reason of which justice requires that a stay should nevertheless not be granted.  In this inquiry, the court will consider all the circumstances of the case, including circumstances which go beyond those taken into account when considering connecting factors with other jurisdictions.  One such factor can be the fact, if established objectively by cogent evidence, that the plaintiff will not obtain justice in the foreign jurisdiction … How far other advantages to the plaintiff in proceedings in this country may be relevant in this connection, I shall have to consider at a later stage.  [At 854-6, per Lord Goff; emphasis added.]

As to the question of such “advantages”, Lord Goff also stated:

[T]he mere fact that the plaintiff has a legitimate personal or juridical advantage in proceedings in England cannot be decisive …

The key to the solution of the problem lies, in my judgment, in the underlying fundamental principle.  We have to consider where the case may be tried ‘suitably for the interests of all the parties and for the ends of justice’.  [At 859.]

[57]            Spiliada was adopted, with some modification, by the Supreme Court of Canada in Amchem, although Amchem was about the granting of anti-suit injunctions rather than stays of proceedings.  At p. 912 of his reasons for the Court, Sopinka J. described the anti-suit injunction as “a more aggressive remedy” which, although it operates in personam on the plaintiff in the foreign proceeding and not on the foreign court itself, has “the latter effect and therefore raises serious issues of comity.”  He continued:

Although both the remedy of a stay and an injunction have as their main objectives the selection of an appropriate forum for the trial of the action, there is a fundamental difference between them which is crucial to the development of the principles which should govern each.  In the case of the stay the domestic court determines for itself whether in the circumstances it should take jurisdiction whereas, in the case of the injunction, it in effect determines the matter for the foreign court. …  In assessing the role of comity in the formulation of the principles which should inform the exercise of this power, I adopt the definition of comity approved by La Forest J. in Morguard Investments Ltd. v. De Savoye, [1990] 3 S.C.R. 1077, at p. 1096:

"Comity" in the legal sense, is neither a matter of absolute obligation, on the one hand, nor of mere courtesy and good will, upon the other.  But it is the recognition which one nation allows within its territory to the legislative, executive or judicial acts of another nation, having due regard both to international duty and convenience, and to the rights of its own citizens or of other persons who are under the protection of its laws ...  [At 913-4; emphasis added.]

[58]            One of the Court’s modifications to Spiliada concerned the role of juridical advantage.  Sopinka J. said he saw no reason in principle why it should be treated as “a separate and distinct condition rather than being weighed with the other factors which are considered in identifying the appropriate forum.”  (At 919.)  The weight to be given to this factor was linked to the reasonable expectations of the parties, given their connections with the jurisdiction in question.  In the absence of a real and substantial connection, a party could have no reasonable expectation of invoking the law of the jurisdiction in question, and his or her seeking out that jurisdiction would be regarded as “forum shopping”.  Where on the other hand the connections to the jurisdiction in question are such that the party could reasonably expect to avail himself or herself of the forum, he or she has a “legitimate claim to the advantages that that forum provides.”  (At 920.)

[59]            Returning to the first ground of appeal, TCML submits that Davies J. failed to recognize the presumption against legislation changing the common law and the rule that unless they clearly require it, “statutes should not be construed so as to make any alteration in the common law or to change any established principle of law.”  (See 36 Halsbury’s Laws of England, (3rd ed.) vol. 36, p. 412 at para. 625, quoted in Sullivan and Driedger on the Construction of Statutes (4th ed., 2002) at 341.)  In particular, TCML says that in finding that s. 11 should be informed by “but neither dictated nor constrained by the existing case law”, the Chambers judge effectively determined that case law decided prior to the enactment of the CJPTA was no longer binding and that he was free to fashion “new jurisprudence based on purported statutory modifications”.

[60]            I agree with counsel for TCML that the CJPTA provisions regarding forum non conveniens were meant to be assimilated into the existing body of common law and that the Chambers judge was required to consider and apply the case law already in existence at the time the statute came into force, as well as cases decided subsequently.  However, I do not agree that he effectively ignored or set aside cases such as Amchem, Thrifty, Westec and Ingenium as having been somehow superseded or made obsolete by the new Act.  To the contrary, the Chambers judge went to some length to “harmonize” the common law with the factors listed in s. 11(2) and reviewed and discussed the cases referred to by counsel in that connection.  Taken in context, his description of the case law as “now statutorily modified” and of the correct analysis “as being informed by but neither dictated nor constrained by the existing case law” was in my opinion simply a recognition at the outset of his analysis that the Act had the potential to change the existing law.

[61]            Nor do I agree that Davies J. fashioned “new jurisprudence based on purported statutory modifications”.  Instead, he applied the case law in existence at the time CJPTA was drafted, as well as later authorities, relating them to the factors listed in s. 11(2).  That provision does not change the weighing or ‘balancing’ exercise mandated by the cases.  The process remains a discretionary one which takes into account the facts of each particular case:  see J. Walker, Canadian Conflict of Laws (6th ed., looseleaf, 2006) at s. 13.5.  As that author also observes, the standard of appellate review must also reflect “appropriate deference for the exercise of fact-specific discretion.”

[62]            Here, the Chambers judge carefully considered each factor and described at length the reasons for his conclusions.  I am not persuaded that he misapprehended the effect of s. 11(2) or the ‘weighing’ exercise he was required to carry out.  At the end of it, he found that British Columbia was the most appropriate forum for the trial of the coverage action.  I see no error in that conclusion, although I might not have emphasized the matter of “forum shopping” to the extent he did.  (On this point, I agree with the comment of Donald J.A. in Marchand (Guardian ad Litem of) v. Alberta Motor Ass’n. Insurance Co. [1994] 5 W.W.R. 764 (B.C.C.A.), that:

… forum shopping is not so much a basis of objection to jurisdiction as a condemnatory label applied after the real analysis has taken place.  The labelling does not add anything to the decision-making process.  The task is to determine which jurisdiction has the closest connection to the case.  The factors involved in that analysis should not be affected by the motive of the party in choosing the disputed jurisdiction because it is only rational to sue in the most advantageous place.”  [At para. 10.])

[63]            To restate briefly the connections in this case, TCML is a Canadian corporation having its head office in British Columbia; the site of the smelter (and thus the conduct giving rise to the CERCLA action) is in British Columbia; the Insurers have obvious connections with British Columbia; the insurance policies were purchased largely in British Columbia or through British Columbia brokers; and the B.C. coverage actions include three claims or potential claims that TCML has not sued on, and could not sue on, in Washington State.  With respect to the “proper law” applicable to the insurance policies, I agree with Davies J. that the parties would reasonably have expected that since the policies cover global risks, they would be interpreted according to British Columbia law, rather than the law of the site of the particular damage in question in each case.  If, for example, effluent had been discharged into a river that flowed from British Columbia to California, three U.S. states could have asserted jurisdiction and each could have had an “interest” (assuming that is a legitimate consideration in this context) in applying its own law to the insurance coverage questions.  In such a case, it seems reasonable to expect that one law should be applied to the policies – i.e., the law of British Columbia. (On this point, see also Gerling Global General Insurance Co. v. Canadian Occidental Petroleum Ltd. (1998) 230 A.R. 39, 1998 ABQB 714, at para. 68; Commonwealth Insurance Co. v. Canadian Imperial Bank of Commerce (2005) 21 C.C.L.I (4th) 226 (Ont. S.C.J.) at para. 68; and Travelers Casualty and Surety Co. of Canada v. Sun Life Assurance Co. of Canada (U.K.) Ltd. [2006] EWHC 2716 (Q.B.) at para. 22.)  Like the Court in Western Union, then, I find there was ample evidence to support the Chambers judge’s conclusion that British Columbia was the more appropriate forum and that the connections of the parties and the litigation to Washington were weak in comparison.

The Role of Comity

[64]            TCML’s real complaint, with respect, lies in its second ground of appeal – that the Chambers judge did not interpret Amchem, Thrifty, Westec and Ingenium to mean that the assertion of jurisdiction by the Washington court was “virtually conclusive” of the question of forum non conveniens.  TCML again relies heavily on those cases for the proposition that “Comity is the overriding governing principle of forum non conveniens in Canada” and that, as stated at para. 38 of its factum:

This jurisprudence establishes that where there has been a prior assertion of jurisdiction by a foreign court, a British Columbia court should inquire to determine whether that decision was made based on principles that generally conform to our rule of forum non conveniens.  If so, comity demands that the decision be respected.  [Emphasis added.]

[65]            In support, TCML notes the observation of Esson J.A. for the Court in Thrifty that:

The Abidin Daver and The Spiliada laid to rest the insular English rule in both its original and modified forms and mandated the radically different approach of determining which is the more appropriate jurisdiction. Comity, which played no part in the old rule, is now a major consideration.  Parallel actions dealing with the same subject matter must now be avoided unless the party resisting the application to stay can demonstrate possible loss of a juridical advantage.  The right of the plaintiff to sue in the court of his choice is not now a significant factor.  A primary purpose of the present rule is to avoid having two actions proceeding in different jurisdictions with the attendant risk of conflicting decisions.  [At para. 32; emphasis added.]

and of Rowles J.A. for the Court in Westec that:

Refusing to grant a stay based on the fact that Raytheon seeks only declaratory relief would also risk offending the principle of comity.  Having verified that Kansas is an appropriate forum to adjudicate the dispute, a stay cannot then be refused simply on the basis that the action as framed in Kansas may not be recognized if brought under the laws of this Province.  If comity is to be the rule, procedural differences cannot be allowed to govern.  [At para. 47.]

and of Lowry J.A. for the majority in Ingenium that:

It may be that, as a general rule on a stay application, comity does not become a determining factor to be considered by the domestic court where parallel proceedings exist until a jurisdictional decision has been made by the foreign court. But the statement made in Loewen upon which Wedge J. relied to the effect that comity only becomes relevant in the context of parallel proceedings when one is considering an anti-suit injunction cannot be right. This Court's decision in Thrifty, and the English authority upon which this Court relied in Thrifty as being the law in this province on the point, is decidedly to the contrary.

Given that the positive assertion of jurisdiction by the Ontario court was said to be conclusive in Thrifty, I see no reason that it should not be conclusive in this case, although there are distinguishing features.  [At paras. 21-2.]


… Given this Court's decision in Westec, and the significance that comity requires must be given to a positive assertion of jurisdiction as held in Thrifty, it is difficult to see on what basis the stay in this action can be refused.  [At para. 25.]

[66]            However, as I have already noted, the Court in Amchem expressly adopted an approach to forum non conveniens similar to that formulated by Lord Goff in Spiliada (see para. 56 above).  The Court was also careful to distinguish the law relating to stays as opposed to that relating to anti-suit injunctions, as was the Privy Council in SNI Aérospatiale v. Lee Kui Jak, [1987] 3 All E.R. 510 (an opinion also adopted by the Amchem court at 930).  As I read Amchem, where an anti‑suit injunction is sought, the court must undertake a more complex analysis than the “weighing” process for a stay based on forum non conveniens.  The court must first try to determine the “natural” or most appropriate forum for the case, on the basis of the usual “connections” already discussed.  (At 931.)  If there is no “natural” forum, or if there is a “genuine disagreement” with a foreign court, the court must go on to determine whether the foreign court assumed jurisdiction on a basis inconsistent with forum non conveniens principles.  If it finds the foreign court did not act consistently with forum non conveniens principles and could not have reasonably have reached the conclusion it did had it applied those principles, the court must go to the ‘second step’ of the SNI analysis, but asking whether injustice would result if the plaintiff were deprived of advantages in the foreign  forum.  (At 932.)

[67]            At the very least, Sopinka J.’s complex analysis precludes a simplistic approach that defers to the first court to assert jurisdiction.  Such an approach would be very close to, and perhaps equally objectionable as, the old rule of deference to the plaintiff’s choice – a rule described as obsolete by this court at para. 32 of Thrifty.  The Amchem analysis also carries with it the possibility that an anti-suit injunction will be refused – that a Canadian court will “not purport to make the decision for the foreign court.”  (At 934.)  But the Court made no suggestion that on reaching such a conclusion, the domestic court should automatically grant a stay of the domestic proceedings.  Indeed, at para. 24, Sopinka J. suggested in another context that the existence of parallel proceedings “would not be disastrous” and that since it is unlikely they could be tried concurrently, “the judgment of the first court to resolve the matter would no doubt be accepted as binding by the other jurisdiction in most cases” – at least “in a world where comity was universally respected.”  (At 914.)

[68]            As for the decisions of this court relied on by TCML, the relative weights of the forum non conveniens factors in those cases were different from those in the case at bar.  In Thrifty, the parties to the disputed contract had expressly agreed that it would be interpreted in accordance with the laws of Ontario and had attorned to the “non-exclusive jurisdiction” of the courts of that province. In the course of his review of the English and Canadian law, Esson J.A. noted an earlier decision of the Supreme Court of Canada, Antares Shipping Corp. v. “Capricorn” (The) [1977] 2 S.C.R. 422, which he viewed as something of a ‘lost gem’.  There Ritchie J. for the majority had summarized the principle of forum non conveniens as follows:

… The factors affecting the application of this doctrine have been differently described in various cases … and they include the balance of convenience to all the parties concerned, including the plaintiff, the undesirability of trespassing on the jurisdiction of a foreign state, the impropriety and inconvenience of trying a case in one country when the cause of action arose in another where the laws are different, and the cost of assembling foreign witnesses.

In my view the overriding consideration which must guide the Court in exercising its discretion by refusing to grant such an application as this must, however, be the existence of some other forum more convenient and appropriate for the pursuit of the action and for securing the ends of justice.  [At 448; emphasis added.]

At para. 42 of Thrifty, Esson J.A. lamented that the confusion in Canadian courts prior to Amchem was largely attributable to their having paid so little attention to Antares.  Ultimately in Thrifty, the various connections of the litigation and the parties to the Province of Ontario, including the fact an Ontario court had declined to grant a stay (the “matter of comity between provinces”) were found to warrant the granting of a stay of the British Columbia proceedings.  (I emphasize the reference to interprovincial comity because recent decisions of the Supreme Court of Canada have indicated that comity between provinces is considerably more powerful than that between nations: see Spar Aerospace Ltd. v. American Mobile Satellite Corp. 2002 SCC 78, [2002] 4 S.C.R. 205, at paras. 51-4 and Antwerp Bulkcarriers, N.V. (Re) 2001 SCC 91, [2001] 3 S.C.R. 951, at para. 51.)

[69]            In Westec, the defendant was a Kansas company that carried on business of manufacturing aircraft in that state.  It had contracted with the plaintiff, a British Columbia company, for certain software and hardware which were delivered to the defendant in Kansas for use in Kansas.  Evidently, neither jurisdiction was seen as more appropriate than the other:  Rowles J.A. stated at para. 32 that both Kansas and British Columbia appeared to be appropriate fora for the resolution of the dispute.  (See also paras. 46 and 48.)  The plaintiff failed to show any “objective, cogent evidence” of a juridical advantage that would be lost if the proceedings in British Columbia were stayed.  (Para. 55.)  Finally in Ingenium, the court below was found to have erred in attaching no significance to the fact that proceedings were already underway in New York.  This court found the facts before it to be indistinguishable from those in Thrifty.  At the same time, it stated that “the existence of parallel proceedings does not trump all other factors” and that ‘there will necessarily be cases where parallel proceedings will not carry the day.”  (Para. 9.)

[70]            Western Union was an insurance coverage dispute with obvious similarities to the case at bar.  The insurers were Canadian companies carrying on business in Vancouver.  They did not conduct business in California or have agents or offices in that state.  The defendant was a British Columbia company that manufactured and exported roofing products regularly to California, where it was not authorized to carry on business.  It was being sued in a class action in that state and commenced an action in the California state court for a declaration that the insurers were obliged to defend the class action.  The insurers then sued in British Columbia for a declaration that the policies were void ab initio.  The court below considered the “connections” between the parties, the subject matter of the dispute, and the two jurisdictions, and concluded that British Columbia was “the appropriate jurisdiction for litigation of the insurance issues and … that where an action was first commenced was ‘a factor for consideration’, but not ‘the sole determining factor’.”

[71]            Prior to the hearing of the appeal, the California court dismissed an application by the insurers for an order quashing service of the summons and complaint on them, and staying the California action.  That fact was adduced as “fresh evidence” in this court.  Nevertheless, it was found that the judge below had not erred either in refusing a declaration that the British Columbia courts were without jurisdiction or in declining to exercise that jurisdiction.  Finch J.A. (as he then was) for the Court noted counsel’s arguments based on Thrifty and Westec, but did not regard them as establishing a new test for the determination of forum non conveniens in cases where there are parallel proceedings in two appropriate jurisdictions.  Rather, he said, the two cases were particular applications of the forum non conveniens test and the principle of comity.  He continued:

Both cases are distinguishable from this case, because the learned chambers judge found that, on a consideration of all relevant factors, B.C. had the closest real and substantial connection to the litigation and the parties.  There was ample evidence to support that finding, and I can see no basis on which this court could interfere with it. In my view, the learned chambers judge applied the forum non conveniens test as it was intended to be applied.  The chambers judge properly considered the fact that the California action was first in time, but rejected that factor as being conclusive.  In my respectful opinion, that decision was correct.  If the time at which actions are commenced were conclusive of the issue it would override all of the other factors properly to be considered in a forum non conveniens analysis.  [At para. 29; emphasis added.]

Again it was implicit that a simple rule of deference to another court asserting jurisdiction was not to be applied.

A “Parochial” Approach?

[72]            This brings us to the fourth ground of appeal advanced by TCML – that the Chambers judge “used” the CJPTA to adopt the “parochial” approach to jurisdiction which was discarded two decades ago in Canada; and that he showed no respect for the U.S. District Court and its jurisdictional laws.

[73]            There is strong support for the position that Canadian courts should seek to avoid situations in which two jurisdictions may be dealing with the same subject matter.  The Chambers judge acknowledged this at paras. 170 and 171 of his reasons, and quoted from Thrifty, where Esson J.A. had stated:

The right of the plaintiff to sue in the court of his choice is not now a significant factor.  A primary purpose of the present rule is to avoid having two actions proceeding in different jurisdictions with the attendant risk of conflicting decisions.  [At para. 32.]

[74]            Taken to its logical conclusion, however, this would mean that (assuming some connection between the foreign court and the parties or the subject matter of the litigation) the first court to assert jurisdiction would always be deferred to by the “later” domestic court.  The old rule which deferred to the plaintiff’s choice of forum would be replaced by a new rule deferring to the first court to assert jurisdiction.  (This is essentially the position which TCML advised Judge Suko would be taken by a British Columbia court.)  But in contrast to the case at bar, British Columbia was not the ‘natural’ or most appropriate jurisdiction in any of the cases relied upon by TCML for this proposition.  No “single bright line rule” was formulated under which the existence of parallel proceedings in a foreign court would trump a determination, such as that made in Western Union and in the case at bar, that British Columbia has the closest and most substantial connection with the proceeding.  I do not regard such a conclusion as ‘disrespectful’ to the foreign court, or as negating the principle of comity – a principle based on mutual respect and on “justice, necessity and convenience”. 

[75]            TMCL's assertion that the Chambers judge effectively conducted himself like an appellate court reviewing Judge Suko's decision is also, in my opinion, misconceived.  I read Davies J.'s reasons as carrying out at least part of the complex Amchem analysis, which requires the court in a case of "genuine disagreement", to determine whether the foreign court applied principles of international law generally similar to those applied in British Columbia.  Davies J. pointed out that there were differences in the two courts' approaches (see para. 45 above) – most notably in the Washington court's failure to deal with the "proper law" issue, and its concern with the "interests" of the State of Washington in the litigation.  As well, the Washington court had been given to understand that if it asserted jurisdiction, the courts of British Columbia would 'stand down'.  This analysis was not necessary to the stay application before the Chambers judge, but these differences would have been validly considered in connection with an anti‑suit injunction.  In any event, I do not regard Davies J.’s reasons as 'disrespectful' of the Washington court or as inappropriately critical.  At bottom, it appears he was merely trying to understand how his counterpart had come to a different conclusion than he had regarding forum conveniens.

Forum Shopping and Juridical Advantage

[76]            TCML’s remaining ground of appeal is that the Chambers judge below erred in finding that TCML had engaged in “inappropriate forum shopping”.  In its factum, TCML made several wide-ranging arguments, many of which I have already dealt with in these reasons.  Again, TCML asserts a very simple and blunt argument to the effect that Amchem established that “once a real and substantial connection is found, a prior assertion of jurisdiction must be respected and the party is entitled to any juridical advantage available to it in the foreign jurisdiction.”  Again, this overlooks the fact that at common law and in accordance with s. 11 of the CJPTA, the court must try to determine which jurisdiction is the more appropriate in terms of “connections” with the parties and the subject matter of the litigation.  Where the court finds there is “no other available forum which is clearly more appropriate for the trial of the action”, and where the remedy sought is a stay of proceedings, the court will, in the words of Lord Goff in Spiliada, “ordinarily refuse a stay …. It is difficult to imagine circumstances when, in such a case, a stay may be granted.”  (At 856.)

[77]            Considerations of juridical advantage generally arise where an anti-suit injunction is sought and the court is required to carry out the more complex analysis described at paras. 53-6 of Amchem.  This is because an application for injunctive relief engages equitable considerations of the “ends of justice”.  (Amchem, at 922, 929 and 932; SNI at 522.)  Sopinka J. elaborated at 933-4:

When will it be unjust to deprive the plaintiff in the foreign proceeding of some personal or juridical advantage that is available in that forum?  I have already stated that the importance of the loss of advantage cannot be assessed in isolation.  The loss of juridical or other advantage must be considered in the context of the other factors.  The appropriate inquiry is whether it is unjust to deprive the party seeking to litigate in the foreign jurisdiction of a judicial or other advantage, having regard to the extent that the party and the facts are connected to that forum based on the factors which I have already discussed.  A party can have no reasonable expectation of advantages available in a jurisdiction with which the party and the subject matter of the litigation has little or no connection.  Any loss of advantage to the foreign plaintiff must be weighed as against the loss of advantage, if any, to the defendant in the foreign jurisdiction if the action is tried there rather than in the domestic forum.  I pointed out in my discussion of the test for determining the forum non conveniens that loss of juridical advantage is one of the factors and it will have been considered in step one.  It will also be considered in the second step to determine whether, apart from its influence on the choice of the most appropriate forum, an injustice would result if the plaintiff is allowed to proceed in the foreign jurisdiction.  The loss of a personal or juridical advantage is not necessarily the only potential cause of injustice in this context but it will be, by far, the most frequent.  Indeed most of the authorities involve loss of juridical advantage rather than personal advantage.  Nonetheless, loss of personal advantage might amount to an injustice if, for example, an individual party is required to litigate in a distant forum with which he or she has no connection.  I prefer to leave other possible sources of injustice to be dealt with as they arise.  [Emphasis added.]

[78]            In the case at bar, Davies J. found that by commencing its proceedings in Washington State, TCML was hoping to avail itself of the advantages of Washington law for insured parties – the Washington courts’ approach to “sudden and accidental” events in insurance policies, its treatment of “response costs” as damages, its “continuous trigger” approach to property damage, and most importantly, the fact the Washington courts have rejected the allocation of liability between insurers and insureds, and (in the words of Mr. Zariski in his affidavit) have “indicated that any insurer that was on the risk when the triggering event occurred will be jointly and severally liable for the entire amount of damages up to its policy limit”.  One of the lessons of Amchem, however, is not that parties are not entitled to pursue their own self-interest when it comes to choice of forum, but that they are not entitled to invoke the laws of a jurisdiction with which they have little or no connection, in order to avoid the laws of the jurisdiction which is the most appropriate forum for trying the dispute.  This is what the Chambers judge found had occurred here, and in my view, the evidence supported his conclusion.  I would dismiss this ground of appeal.

Parallel Proceedings

[79]            Finally, I should comment briefly on TCML’s contention that the Chambers judge “placed the parties in the unenviable position of having parallel legal proceedings in two separate jurisdictions over the same subject matter”.  Of course it is a matter of common sense that the existence of parallel actions in two jurisdictions is something to be avoided if possible: see Esson J.A. in Thrifty at para. 32, quoted above at para. 73.  However, such statements should not be taken out of context and in particular should not obscure the overarching objective of the new conflicts rules which is “to ensure, if possible, that the action is tried in the jurisdiction that has the closest connection with the action and the parties and not to secure juridical advantage to one of the litigants at the expense of others in a jurisdiction that is otherwise inappropriate.”  (Amchem, at 912.)

[80]            The fact of concurrent (not parallel) proceedings in the case at bar is not, with respect, due to error on the part of the Chambers judge, but is the result of the parties’ divergent views as to their different interests.  The situation is unfortunate, and it is unclear what the result will be if and when one court or the other decides the case and the successful party attempts to enforce its judgment against the other party or parties.  If for example the Washington court were to rule (on an application of whatever law it found should be applied), that Lloyd’s and Lombard were liable to indemnify TCML, it might become necessary for TCML to enforce that order against the Insurers in British Columbia or other jurisdictions where they carry on business or have significant assets.  If the British Columbia court were then to rule, or had already ruled, that the Insurers were not liable to indemnify TCML, or were liable for lesser amounts than specified in the Washington judgment, one would expect that TCML would at least encounter difficulty in registering or otherwise enforcing its judgment.  (See e.g., s. 29(6)(g) of the Court Order Enforcement Act.)  Another possibility is that the Washington court would apply British Columbia law to the insurance coverage issues.  It would make sense for it then to defer to the British Columbia court's determination of those same issues; otherwise, the Washington proceeding would be duplicative and a waste of time and money.

[81]            Counsel before us did not venture down these paths, however, and since I need not do so to dispose of this appeal, I am content to leave them for counsel to ponder.  I would dismiss the appeal, with thanks to counsel for their able arguments.

“The Honourable Madam Justice Newbury”

I Agree:

“The Honourable Mr. Justice Mackenzie”

I Agree:

“The Honourable Madam Justice Kirkpatrick”