Family Insurance Corp. v. Lombard Canada Ltd.,  2 S.C.R. 695, 2002 SCC 48
Family Insurance Corporation Appellant
Lombard Canada Ltd. Respondent
Canadian Universities Reciprocal Insurance Exchange Intervener
Indexed as: Family Insurance Corp. v. Lombard Canada Ltd.
Neutral citation: 2002 SCC 48.
File No.: 28093.
2002: February 19; 2002: May 23.
Present: McLachlin C.J. and Gonthier, Iacobucci, Major, Bastarache, Binnie and Arbour JJ.
on appeal from the court of appeal for british columbia
Insurance – Overlapping policies – Excess coverage – Two policies of insurance covering same risk – Both insurers relying on “other insurance” clauses to avoid primary liability – Whether clauses should be treated as mutually repugnant – Whether insurers should be required to bear loss equally.
A claim against the insured, Y, was settled with the quantum of damages payable by her two insurers set at $500,000. Y was insured by F under a homeowner/residential insurance policy with a maximum benefit of $1 million, and by L under a commercial general liability policy for up to $5 million. Both policies contained “other insurance” clauses that declared the policies to be “excess insurance” to any other insurance coverage. Each insurer relied on its “other insurance” clause to shield itself from primary liability. In a contest between the two insurers concerning the extent to which each of them was liable to pay the insured’s claim, the trial judge held that the two clauses were mutually repugnant, and that liability should be apportioned between the insurers equally. The Court of Appeal, however, took into consideration the surrounding circumstances and concluded that the underwriting considerations being different, the F policy provided the primary coverage to the insured, while the L policy provided excess coverage. F appealed.
Held: The appeal should be allowed and the decision of the trial judge restored.
The insurer may seek to limit its liability in the provisions of the policy, and thus the policy itself is the proper instrument to determine the liability of each insurer. Where the contest is only between the insurers, there is no basis for referring to surrounding circumstances or looking outside the policy, there being no privity of contract between the insurers. The policies here are clear and unambiguous, and the insurers’ intentions are unequivocal. The appellant, F, intended to provide primary coverage unless other valid insurance was available, in which case, it intended to provide only excess coverage. The respondent, L, intended to provide primary coverage unless other valid insurance was available, in which case, it intended to provide only excess coverage regardless of the type of coverage provided by the other insurer. In the face of these irreconcilable intentions, the court must determine the most equitable means of resolving the dispute, respecting the intentions of the parties and the right of the insured to recover fully. The Minnesota approach which includes the closeness to the risk approach and which entails a broad analysis of the policies and the surrounding circumstances to determine the overall policy insuring intent of the insurers, is rejected, as it does not accord with the principles of equitable contribution, nor does it respect the intentions of both insurers. The better approach is one that recognizes that the parties involved have not contracted with each other, so that their subjective intentions are irrelevant. Although the intentions of the insurers govern the interpretation exercise, the focus of the examination is to determine whether the insurers intended to limit their obligation to contribute, by what method, and in what circumstances vis‑à‑vis the insured. In the absence of such limiting intentions or where those intentions conflict, principles of equitable contribution demand that parties under a coordinate obligation to make good the loss must share that burden equally. This approach serves to respect the intentions of both the insurers while simultaneously respecting the insured’s contractual right to full indemnity. The “independent liability” approach to apportionment to the respective policy limits represents the fairest method of apportionment and accords with the principles on which the doctrine of equitable contribution is based.
On its face, the L policy provides the same primary coverage as that of F. To endorse the intentions of one insurer over another where both parties have sought to limit their liability to contribute and where the offending clauses, on their face, are irreconcilable, does violence to the intentions of the insurers and does not respect the obligation of both insurers to contribute. Were the court to give effect to each insurer’s intention, F would provide only excess coverage and L would provide only excess coverage to that, leaving the insured without primary coverage at all, an absurd result. Where the competing policies cannot be read in harmony, the conflicting clauses should be treated as mutually repugnant and inoperative, as this accords with the expectations of both the insured and the insurers. Thus both policies provide the insured with primary coverage, so that each insurer is independently liable to the insured for the full loss. Where liability is shared among insurers covering the same risk, the loss is borne equally by each insurer until the lower policy limit is exhausted, with the policy with the higher limit contributing any remaining amounts.
Followed: Godin v. London Assurance Co. (1758), 1 Burr. 489, 97 E.R. 419; Wawanesa Mutual Insurance Co. v. Commercial Union Assurance Co.,  10 W.W.R. 701; McGeough v. Stay ’N Save Motor Inns Inc. (1994), 116 D.L.R. (4th) 137; Wawanesa Mutual Insurance Co. v. Co‑operative Fire & Casualty Co. (1980), 119 D.L.R. (3d) 188; Marchand v. Dominion of Canada General Insurance Co.,  O.J. No. 329 (QL), leave to appeal refused,  1 S.C.R. x; Guardian Insurance of Canada v. New Hampshire Insurance Co. (1997), 193 N.B.R. (2d) 111; Simcoe & Erie General Insurance Co. v. Kansa General Insurance Co. (1994), 93 B.C.L.R. (2d) 1; Dominion of Canada General Insurance Co. v. Wawanesa Mutual Insurance Co. (1985), 64 B.C.L.R. 122; Weddell v. Road Transport and General Insurance Co.,  2 K.B. 563; considered: Seagate Hotel Ltd. v. Simcoe & Erie General Insurance Co. (1980), 22 B.C.L.R. 374, aff’d (1981), 27 B.C.L.R. 89; Carriers Insurance Co. v. American Policyholders’ Insurance Co., 404 A.2d 216 (1979); Brown v. Travelers Insurance Co., 610 A.2d 127 (1992); Hoffmaster v. Harleysville Insurance Co., 657 A.2d 1274 (1995), leave to appeal refused, 668 A.2d 1133 (1995); Lamb‑Weston, Inc. v. Oregon Automobile Insurance Co., 341 P.2d 110 (1959); Commercial Union Assurance Co. v. Hayden,  1 Q.B. 804; referred to: Hartford Steam Boiler Inspection & Ins. Co. v. Cochran Oil Mill & Ginnery Co., 105 S.E. 856 (1921); Auto Owners Insurance Co. v. Northstar Mutual Insurance Co., 281 N.W.2d 700 (1979); Interstate Fire & Casualty Co. v. Auto‑Owners Insurance Co., 433 N.W.2d 82 (1988); Garrick v. Northland Insurance Co., 469 N.W.2d 709 (1991); American Family Insurance v. National Casualty Co., 515 N.W.2d 741 (1994); Eli Lilly & Co. v. Novopharm Ltd.,  2 S.C.R. 129; North Star Mutual Insurance Co. v. Midwest Family Mutual Insurance Co., 634 N.W.2d 216 (2001).
Brown, Craig. Insurance Law in Canada, vol. 1. Scarborough, Ont.: Carswell, 1999 (loose‑leaf updated 2001, release 2).
Couch Cyclopedia of Insurance Law, vol. 16, 2nd ed. By Ronald A. Anderson. Revised volume by Mark S. Rhodes. Rochester, N.Y.: Lawyers Co-operative, 1983.
Grimes, Samuel G. “Insurance __ The ‘Other Insurance’ Clause Conflict” (1967-1968), 46 N.C. L. Rev. 433.
Guthrie, Mark C. “‘Other Insurance’ Conflicts: A Common‑Sense Proposal” (1984), 36 Baylor L. Rev. 689.
Hasse, Linda Kogel. “Is There a Solution to the Circular Riddle? The Effect of ‘Other Insurance’ Clauses on the Public, the Courts, and the Insurance Industry” (1980), 25 S.D. L. Rev. 37.
Ivamy, Edward Richard Hardy. General Principles of Insurance Law, 3rd ed. London: Butterworths, 1975.
Ivamy, Edward Richard Hardy. General Principles of Insurance Law, 6th ed. London: Butterworths, 1993.
Note. “Concurrent Coverage in Automobile Liability Insurance” (1965), 65 Colum. L. Rev. 319.
APPEAL from a judgment of the British Columbia Court of Appeal (2000), 187 D.L.R. (4th) 605, 139 B.C.A.C. 181, 227 W.A.C. 181, 75 B.C.L.R. (3d) 263, 18 C.C.L.I. (3d) 165,  6 W.W.R. 111,  I.L.R. I‑3861,  B.C.J. No. 1076 (QL), 2000 BCCA 330, setting aside a decision of the British Columbia Supreme Court (1999), 10 C.C.L.I. (3d) 58,  I.L.R. I‑3692,  B.C.J. No. 515 (QL). Appeal allowed.
Neo J. Tuytel and Jonathan L. S. Hodes, for the appellant.
James H. MacMaster and Donald B. Lebans, for the respondent.
Patrick J. Monaghan, for the intervener.
The judgment of the Court was delivered by
1 Bastarache J. -- This appeal concerns the issue of overlapping insurance coverage and the proper approach to the reconciliation of disputes among two insurers, each of whom has sought to limit its liability where other insurance covers the same loss. The issue has engendered a significant amount of jurisprudence and is far from novel. Nevertheless, the reconciliation of competing and apparently irreconcilable insurance policy provisions has plagued the courts and given rise to much academic comment in Canada as well as in American jurisdictions. This is a good opportunity for this Court to clarify the law in this area.
I. Factual Background
2 In March of 1996, Michelle Patterson was injured in a fall from a horse. Patterson sued Lesley Young, the owner of the stable, and Douglas Blanchard, the owner of the horse. Ms. Patterson’s claim was eventually settled, with the quantum of damages payable by Ms. Young's insurers set at $500,000.
3 At the time of the accident, Lesley Young was insured by the appellant, Family Insurance Corporation (“Family”), under a homeowner/residential insurance policy with a maximum benefit of $1 million. Young was also insured by the respondent, Lombard Canada Ltd. (“Lombard”), under a Commercial General Liability Policy for up to $5 million, a policy Young obtained when she became a member of the Horse Council of British Columbia (“HCBC”). Although Lombard denied liability initially, both insurers concede that Young’s liability claim falls within the scope of coverage of both of the policies.
4 Although the wording of each is different, both insurance policies contain “other insurance” clauses that declare the policies to be “excess coverage” to any other insurance coverage held by the insured. The Family policy contains the following “other insurance” clause:
If other insurance exists which applies to a loss or claim or would have applied if this policy did not exist, this policy will be considered excess insurance and the Insurer is not liable for any loss or claim until the amount of such other insurance is used up.
The Lombard policy phrased its “other insurance” clause as follows:
If other valid and collectible insurance is available to the Insured for a loss we cover under Coverages A, B or D of this form, our obligations are limited as follows:
. . .
This insurance is excess over other existing insurance if any, whether such other insurance be primary, excess, contingent or on any other basis, that is liability insurance such as, but not limited to comprehensive personal liability, comprehensive general liability coverages or similar coverage for liability arising out of the activities of any insured.
The contest, then, is between the two insurers and concerns the extent to which each of them is liable to pay the insured’s claim. Each insurer relies on its “other insurance” clause as a means of shielding itself from primary liability.
5 At trial, the parties agreed that the only issue before the court was the extent, if any, of the liability of each insurer. They further agreed that, if the court held both insurers liable under the respective policies, each would share equally in the liability to their respective policy limits. McEwan J. of the British Columbia Supreme Court found each insurer to be primary and, accordingly, that each should contribute equally to the loss. Lombard appealed, arguing that the trial judge erred in finding Lombard to be a primary insurer. Similarly, Family cross-appealed, claiming the trial judge erred in finding that Family was liable as a primary insurer. At the hearing, however, Family abandoned its cross-appeal and agreed with the Court of Appeal that its position was untenable.
II. Judicial History
A. British Columbia Supreme Court (1999), 10 C.C.L.I. (3d) 58
6 McEwan J. first noted that the wording of the two clauses was not identical. Thus the first step was to determine whether the two clauses could be reconciled by determining the intent of the two insurers as revealed by the content of their respective policies: Simcoe & Erie General Insurance Co. v. Kansa General Insurance Co. (1994), 93 B.C.L.R. (2d) 1 (C.A.). McEwan J. noted that the appellant’s policy was worded so that the test for liability was whether the other policy would function as the primary insurance if the appellant’s policy did not exist. The respondent argued that its policy was excess even to excess policies, so that even if the appellant met its own definition as an excess insurer, then the respondent would be excess to it. McEwan J. rejected this view, noting that such an interpretation leads to circular reasoning that does not resolve the problem of which policy provides primary coverage: McGeough v. Stay ’N Save Motor Inns Inc. (1994), 116 D.L.R. (4th) 137 (B.C.C.A.). He stated, at para. 8: “Beyond a certain point I do not think it makes much sense to arbitrate what amounts to a kind of drafting ‘tag’. The clauses are irreconcilable. Applying the principle set out in the cases the excess clauses are inoperative.”
7 McEwan J. then addressed the issue of how to apportion liability as between the two insurers. He noted that the Lombard policy provided a formula for sharing liability whereas the Family policy was silent on the issue. He held that without an agreement between the parties, Lombard’s formula could not bind Family. Applying the “independent liability approach” from Dominion of Canada General Insurance Co. v. Wawanesa Mutual Insurance Co. (1985), 64 B.C.L.R. 122 (S.C.), McEwan J. held that the “other insurance” clauses cancel each other out. As both “other insurance” clauses were inoperative, each policy is regarded as providing primary coverage, so that each insurer is required to contribute equally to the applicable limits of each policy or until none of the loss remains: McGeough, supra, at p. 146; Dominion of Canada v. Wawanesa, supra, at pp. 130 and 132-33.
B. British Columbia Court of Appeal (2000), 75 B.C.L.R. (3d) 263, 2000 BCCA 330
8 Hall J.A. reviewed the competing policies and noted that the HCBC advised its members that liability coverage under the Lombard policy was excess to personal liability insurance that its members might possess. Although he did not believe that the HCBC’s advertisement governed the interpretation of the policies, he reasoned as follows, at para. 4:
One must have regard to the language of the policies although a consideration of policy language cannot be completely divorced from surrounding circumstances and commercial reality.
In reaching his decision, he adopted the approach of Legg J.A. in McGeough, supra, that reconciliation of overlapping coverage policies is governed by the intent of the insurers as manifested by the terms of the policies issued. Hall J.A. also relied upon E. R. H. Ivamy, General Principles of Insurance Law (3rd ed. 1975), at pp. 312‑13, where the author states:
The cardinal rule of construction is that the intention of the parties must prevail. But the intention is to be looked for on the face of the policy, including any documents incorporated therewith, in the words which the parties have themselves chosen to express their meaning. The Court must not speculate as to their intention, apart from their words, but may, if necessary, interpret the words by reference to the surrounding circumstances.
9 Hall J.A. agreed with the trial judge that, on a consideration of the clauses alone, the “other insurance” clauses were irreconcilable. He further noted that the “traditional way” to apportion liability where the wording of the policies cannot be reconciled accords with what McEwan J. did in this case. Finally, he agreed with the trial judge that Simcoe & Erie v. Kansa, supra, stipulated that the liability of insurers with overlapping policies was governed by the intent of the insurers as manifested by the terms of the policies.
10 Hall J.A. distinguished Simcoe & Erie v. Kansa, supra, where it was held that the clauses cancelled each other out and liability was apportioned equally, on the basis that the competing policies in that case were alike in type and coverage. He held that had the present policies been similar, he would have agreed with the result reached by McEwan J. at trial. Turning to the surrounding circumstances of this case, Hall J.A. reasoned as follows, at para. 15:
[I]t seems to me that the court can and should take account of the circumstance that these two policies do appear rather different from each other in type and purpose. I believe we must consider the particular circumstances reflected by the description and wording of the respective policies. It is also, in my judgment, appropriate to take account of the limited surrounding facts we have in this case in an effort to resolve this dispute between the competing insurers.
11 Hall J.A. then noted that the Family policy was a standard homeowner policy covering inter alia dwelling premises, personal property and personal liability. The Lombard policy covered Young through her membership in the HCBC. Hall J.A. found that, because the Lombard policy was offered through a club, it was “the intent of the insurer to offer a type of excess coverage generally to a class of insureds with a proviso that, absent any other coverage, the policy could provide primary insurance coverage” (para. 16). He also considered that the HCBC believed Lombard’s policy provided excess coverage for its members. Although he did not give this factor much weight, he took it into account as a surrounding circumstance which gave some indication that such an interpretation of the Lombard policy would accord with commercial reality.
12 Hall J.A. took note of the higher coverage provided by the Lombard policy, the broader net of coverage and its availability to a group of insureds. He also observed that the underwriting considerations were different for this policy than for the Family policy. He stated at para. 19:
Concerning the Family policy, rates will be based on past or anticipated loss experience related to all homeowners likely to purchase an individual homeowners policy. A limit of $1 million is provided for to limit maximum exposure to that sum in a personal liability claim. The Lombard policy is underwritten based on past or anticipated loss experience related to the covered group, bearing in mind on the one hand the strong probability of other existing coverage for most members, and thus the avoidance of coverage or responsibility on most claims and on the other hand providing a higher limit to afford, in the rare case, excess coverage to an insured having a large claim. Excess coverage may very occasionally be required for a catastrophic claim over the amount covered by the limits of the anticipated existing insurance coverage and in most cases, primary coverage will not be given under the Lombard policy, because of the existence of existing liability insurance coverage of the insured.
13 Hall J.A. noted again that, in most cases, the analysis undertaken by the trial judge would have properly disposed of the matter. In this particular case, however, a departure from the normal course was necessary due to the unusual circumstances presented by the policies under consideration, which were designed to be different as to their purpose and anticipated scope of coverage. He concluded that based on the policies and the surrounding circumstances, the Family policy provided the primary coverage to the insured, while the Lombard policy should be viewed as excess coverage. As a result he allowed the appeal.
14 It is a well-established principle of insurance law that where an insured holds more than one policy of insurance that covers the same risk, the insured may never recover more than the amount of the full loss but is entitled to select the policy under which to claim indemnity, subject to any conditions to the contrary. The selected insurer, in turn, is entitled to contribution from all other insurers who have covered the same risk. This doctrine of equitable contribution among insurers is founded on the general principle that parties under a coordinate liability to make good a loss must share that burden pro rata. It finds its historic articulation in the words of Lord Mansfield C.J. in Godin v. London Assurance Co. (1758), 1 Burr. 489, 97 E.R. 419 (K.B.), at p. 420:
If the insured is to receive but one satisfaction, natural justice says that the several insurers shall all of them contribute pro rata, to satisfy that loss against which they have all insured.
15 More recently, Ivamy’s General Principles of Insurance Law (6th ed. 1993) set out at p. 518 the general principles concerning the right of contribution among insurers as follows:
1 All the policies concerned must comprise the same subject‑matter.
2 All the policies must be effected against the same peril.
3 All the policies must be effected by or on behalf of the same assured.
4 All the policies must be in force at the time of the loss.
5 All the policies must be legal contracts of insurance.
6 No policy must contain any stipulation by which it is excluded from contribution.
16 It is in the context of this doctrine of equitable contribution and the opportunity for insurers to contract out of that obligation that the problem of reconciling “other insurance” clauses occurs. Although the insurer’s obligation to contribute does not arise out of the insurance contract per se, the insurer may express its intention to limit its liability in the provisions of the policy. Thus the proper instrument to determine the liability of each insurer is the policy itself.
17 In accordance with the general principles of contract interpretation, the exercise is properly one of determining the parties’ intentions. The proper approach is best described in Couch Cyclopedia of Insurance Law (2nd ed. 1983), vol. 16, at p. 498:
Intent of insurers as controlling
There is authority that the liability of insurers under overlapping coverage policies is to be governed by the intent of the insurers as manifested by the terms of the policies which they have issued. Thus, it has been said that where two or more liability policies overlap and cover the same risk and the same accident, the respective liabilities of the insurers must rest upon a construction of the language employed by the respective insurers, and not upon the so-called “primary tortfeasor doctrine” or upon any other arbitrary rule or circumstance.
18 The intention which the court seeks to determine is found by looking at the means by and extent to which each insurer has sought to limit its liability to the insured when the insured has purchased other policies covering the same risk. Thus, the interpretation exercise is concerned with determining the intentions of the insurers vis-à-vis the insured. In Seagate Hotel Ltd. v. Simcoe & Erie General Insurance Co. (1981), 27 B.C.L.R. 89 (C.A.), McFarlane J.A. emphasized this point and noted, at p. 90, as follows:
The second matter, which in my opinion is a very important consideration to keep in mind here, is that the court is not asked to interpret a contract made between the appellant and the respondent companies. The issue depends upon the interpretation and the application of two contracts, both insurance policies, the first being a contract between the appellant and the insured and the second the contract between the respondent and the insured.
19 Thus, while it remains true that the intentions of the insurers prevail, the inquiry is of necessity limited to the insurers’ intentions vis-à-vis the insured. In the case of an insurance contract, the entire agreement between the insurer and the insured is contained within the policy itself and evidence of the parties’ intentions must be sought in the words they chose. Were the dispute between the insurer and the insured, reference to surrounding circumstances may be appropriate if provisions of the policy are ambiguous. Once the interest of the insured is no longer at stake, that is, where the contest is only between the insurers, there is simply no basis for looking outside the policy. In the absence of privity of contract between the parties, the unilateral and subjective intentions of the insurers, unaware of one another at the time the contracts were made, are simply irrelevant.
20 Looking to the policies themselves, there is no concern here that the words or the intentions that they reveal are ambiguous. Both the trial judge and the Court of Appeal found the insurers’ intentions clear and unequivocal. In the case of the appellant Family, the policy reveals an intention to provide primary coverage unless other valid insurance is available, in which case it intended to provide only excess coverage. In the case of the respondent Lombard, the policy reveals that the insurer intended to provide primary coverage unless other valid insurance is available, in which case it intended to limit its liability to only excess coverage, even where the other insurance also provides excess coverage, thus rendering the Lombard policy “excess to excess”. Accordingly, the insurers’ intentions, expressed unambiguously in the policies, give rise to a “circular reasoning that does not resolve the problem of which policy provides primary coverage” (McGeough, supra, at p. 144).
21 In the face of these irreconcilable intentions, the court must determine the most equitable means of resolving the dispute, one which respects the intentions of the parties as well as the right of the insured to recover fully. In certain jurisdictions of the United States, this has led courts to examine the overall purpose of the policies in question in order to discern which of the insurers intended more clearly to insure the particular risk. This endeavour, known as the “Minnesota approach”, derives from the interpretive “specific versus general” rule. In 1921, the Georgia Court of Appeals, in Hartford Steam Boiler Inspection & Ins. Co. v. Cochran Oil Mill & Ginnery Co., 105 S.E. 856 (1921), stated the issue and the principle to be applied as follows, at p. 858:
In a case such as is now before us (that is, where the several policies are of a different nature and character, and where the risk assumed is only partly coextensive, and where each policy expressly provides that the risk assumed, at the point of mutual contact, is “excess” insurance only, and therefore secondary to the other policy) the decisive test to be applied in determining which of these two limitations is to be given effect (that is, which is really primary or basic insurance, and which is excess insurance only) lies in the answer to the question as to which insurance is general and which is specific in its nature.
22 In its more modern manifestation, this approach has evolved into an analysis of the competing policies in an effort to determine each insurer’s intention to cover the particular loss. In assessing each of the insurers’ “closeness to the risk”, courts look to the following non-exclusive factors:
(1) Which policy specifically described the accident-causing instrumentality?
(2) Which premium is reflective of the greater contemplated exposure?
(3) Does one policy contemplate the risk and use of the accident-causing instrumentality with greater specificity than the other policy – that is, is coverage of the risk primary in one policy and incidental to the other?
(Auto Owners Insurance Co. v. Northstar Mutual Insurance Co., 281 N.W.2d 700 (Minn. 1979), at p. 704)
23 More recently, Minnesota courts have tended toward a broad analysis of the policies and the surrounding circumstances with a view to determining the “total policy insuring intent” of the insurers. Courts look to the primary risks on which each insurer’s premiums are based and the primary function of each policy: Interstate Fire & Casualty Co. v. Auto-Owners Insurance Co., 433 N.W.2d 82 (Minn. 1988); Garrick v. Northland Insurance Co., 469 N.W.2d 709 (Minn. 1991); American Family Insurance v. National Casualty Co., 515 N.W.2d 741 (Minn. Ct. App. 1994); North Star Mutual Insurance Co. v. Midwest Family Mutual Insurance Co., 634 N.W.2d 216 (Minn. Ct. App. 2001).
24 The Minnesota approach has been the subject of significant criticism. Several American courts have criticized it as an exercise in “hair-splitting” and “semantic microscopy”: Carriers Insurance Co. v. American Policyholders’ Insurance Co., 404 A.2d 216 (Me. 1979), at p. 219; Brown v. Travelers Insurance Co., 610 A.2d 127 (R.I. 1992); Hoffmaster v. Harleysville Insurance Co., 657 A.2d 1274 (Pa. Super. Ct. 1995), leave to appeal refused, 668 A.2d 1133 (1995). Moreover, academic commentators have denounced it as promoting uncertainty and increased litigation. One commentator stated it most emphatically as follows:
The mental gymnastics engaged by the courts in their attempt to untangle the web of confusion has served only to maintain the perpetual merry-go-round of drafting and re-drafting by the insurers and continued litigation by the courts, resulting only in increased premiums for the insurance buying public.
L. K. Hasse, “Is There a Solution to the Circular Riddle? The Effect of ‘Other Insurance’ Clauses on the Public, the Courts, and the Insurance Industry” (1980), 25 S.D. L. Rev. 37, at p. 37. See also: Note, “Concurrent Coverage in Automobile Liability Insurance” (1965), 65 Colum. L. Rev. 319; S. G. Grimes, “Insurance – The ‘Other Insurance’ Clause Conflict” (1967-1968), 46 N.C. L. Rev. 433, at p. 439; M. C. Guthrie, “‘Other Insurance’ Conflicts: A Common-Sense Proposal” (1984), 36 Baylor L. Rev. 689, at pp. 692-93.
25 The Minnesota approach has also been criticized in Canada. In Wawanesa Mutual Insurance Co. v. Co-operative Fire & Casualty Co. (1980), 119 D.L.R. (3d) 188 (Sask. Q.B.), MacLeod J. explained, at p. 192:
Such an approach would be an exception to the ordinary rule, and pose new problems. If, for example, the insured sought to recover from the “general” insurer in advance of the “specific” insurer, would the “general” insurer seek to recover the loss in full from the “specific” insurer before having resort to the “general” (thus creating an in seriatum approach to coverage)? If allowed, this would be indemnity, not contribution. It would permit the “general” insurer, in effect, to rely upon the excess coverage clause, while at the same time denying that right to the “specific” insurer. Such an interpretation would be another absurdity.
26 Moreover, the Ontario Court of Appeal recently rejected the “closeness to the risk” approach and noted briefly as follows in Marchand v. Dominion of Canada General Insurance Co.,  O.J. No. 329 (QL), at para. 3, leave to appeal refused,  1 S.C.R. x:
There is no real suggestion in Canadian jurisprudence that a “closest to the risk” policy should be embraced, and no consensus in the American authorities. Commercial efficacy and the avoidance of litigation between insurance companies supports a split responsibility and the avoidance of litigation as to which policy is closer to the risk or the coverage.
27 I find the reasoning of MacLeod J. in Wawanesa Mutual v. Co-operative Fire & Casualty, supra, and the Court of Appeal in Marchand, supra, compelling. In effect, the Minnesota approach to resolution of overlapping coverage disputes results in the preference and endorsement of the intentions of one insurer over another. This result does not accord with the principles of equitable contribution nor does it respect the intentions of both insurers.
28 The better approach is one that recognizes that the principles of contract interpretation must be applied here in light of the fact that the parties involved have not contracted with one another. Although the intentions of the insurers govern the interpretation exercise, the focus of the examination is to determine whether the insurers intended to limit their obligation to contribute, by what method, and in what circumstances vis-à-vis the insured. In the absence of such limiting intentions or where those intentions cannot be reconciled, principles of equitable contribution demand that parties under a coordinate obligation to make good the loss must share that burden equally.
29 Turning to the American jurisprudence, it is interesting to note that in Lamb-Weston, Inc. v. Oregon Automobile Insurance Co., 341 P.2d 110 (1959), the Oregon Supreme Court appreciated the problematic nature of a strict contract interpretation approach to a dispute that was really about the insurer’s liability to contribute. The court in Lamb-Weston examined extensively the various approaches to the resolution of overlapping coverage disputes. Referring to the line of cases wherein courts found the competing clauses mutually repugnant and each insurer liable for its proportionate share, the court reasoned as follows, at pp. 118-19:
We are of the opinion that in these later cases the courts have placed this problem in its true perspective by recognizing the absurdity of attempting to assume that where conflicting “other insurance” provisions exist by reason of overlapping coverages of the same occurrence the provisions of one policy must yield to the provisions of the other.
. . .
The “other insurance” clauses of all policies are but methods used by insurers to limit their liability, whether using language that relieves them from all liability (usually referred to as an “escape clause”) or that used by St. Paul (usually referred to as an “excess clause”) or that used by Oregon (usually referred to as a “prorata clause”). In our opinion, whether one policy uses one clause or another, when any come in conflict with the “other insurance” clause of another insurer, regardless of the nature of the clause, they are in fact repugnant and each should be rejected in toto.
30 Canadian courts have adopted a different approach in dispensing with the insurer’s intentions. They have generally followed the approach set out by Rowlatt J. in Weddell v. Road Transport and General Insurance Co.,  2 K.B. 563; Dominion of Canada v. Wawanesa, supra; McGeough, supra; Simcoe & Erie v. Kansa, supra; Wawanesa Mutual Insurance Co. v. Commercial Union Assurance Co.,  10 W.W.R. 701 (Man. Q.B.).
31 In Weddell, supra, Rowlatt J. was keenly aware of the absurdity that would result if the court gave effect to the intentions of all insurers and reasoned at p. 567:
In my judgment it is unreasonable to suppose that it was intended that clauses such as these should cancel each other (by neglecting in each case the proviso in the other policy) with the result that, on the ground in each case that the loss is covered elsewhere, it is covered nowhere. On the contrary the reasonable construction is to exclude from the category of co-existing cover any cover which is expressed to be itself cancelled by such co-existence, and to hold in such cases that both companies are liable, subject of course in both cases to any rateable proportion clause which there may be.
32 C. Brown, in Insurance Law in Canada (loose-leaf ed.), vol. 1, agrees with the reasoning of Rowlatt J. and states the principle as follows at p. 14-9:
If the competing policies each contain such a clause the appropriate way to resolve the apparent impasse is to adopt the approach of Rowlatt J. in Weddell and the Canadian decisions that have taken the same approach and view the clause as not applying to insurance, which is itself subject to such a clause. This sensibly avoids an absurdity and gives effect to the reasonable expectations of the insured.
33 However, this step is only to be taken where there is a true impasse. Where the respective wording of two “other insurance” clauses can be reconciled so as to give effect to both policies while providing coverage for the insured, then no mutual repugnancy exists, and the process is simply one of giving effect to the intent of the insurers: Rae J., Seagate Hotel Ltd. v. Simcoe & Erie General Insurance Co. (1980), 22 B.C.L.R. 374 (S.C.) (upheld on appeal (1981), 27 B.C.L.R. 89). In resolving a conflict between a pro rata “other insurance” clause and an excess “other insurance” clause, Rae J. correctly pointed out, at p. 378, that “one first turns to the wording of the clauses referred to and the interpretation thereof”. On this basis, he concluded the two clauses could be reconciled, and held that the policy containing the pro rata clause was primary. This general method has been followed on numerous occasions: see Simcoe & Erie v. Kansa, supra, and McGeough, supra.
34 In my view, the approach just described not only represents the prevailing view in Canada, but is the preferable one. It serves to respect the intentions of both insurers while simultaneously respecting the insured’s contractual right to full indemnity. What must be kept in mind is that policies of insurance are not written in a vacuum. Insurers most certainly are aware that the insured may obtain coverage for the same risk elsewhere and that, in such a case, the law of equitable contribution will necessarily arise. Thus, rendering the irreconcilable clauses inoperative will promote certainty and thereby serve to reduce the “drafting tag” of which McEwan J. spoke.
35 In the present appeal, the intention of each insurer, as evidenced by the respective policies, is to limit its liability to excess coverage in the event that other insurance covering the same risk is available. The Court of Appeal agreed that the usual course in such circumstances is to find the offending clauses mutually repugnant, but it was of the view that the policies could nevertheless be reconciled by looking to the surrounding circumstances. In my view, the court erred in this respect. Although Hall J.A. began, at least implicitly, with an examination of the insurers’ intention to limit their liability to contribute, the inquiry did not stop there. Rather than acknowledge that the insurers’ intentions were mutually repugnant, Hall J.A. unnecessarily broadened the inquiry to include the overall insuring intent as a means of determining which of the insurers was more predisposed to providing primary and excess coverage. Relying on his interpretation of the type and purpose of the respondent Lombard’s policy, as reflected in the extrinsic evidence and the surrounding circumstances, Hall J.A. concluded that Lombard’s foremost intention was to provide excess coverage whereas Family’s foremost intention was to provide primary coverage. In my view, then, the Court of Appeal erred in taking into account underwriting considerations in order to determine the true intent of the insurers and thereby overcome the fact that the two “other insurance” clauses were clearly repugnant on their terms.
36 In essence, the Court of Appeal endorsed the intentions of Lombard over Family, relying on evidence of Lombard’s subjective desire to provide primarily excess coverage. Lombard’s subjective desire, however, is not evident in its contract with the insured. In Eli Lilly & Co. v. Novopharm Ltd.,  2 S.C.R. 129, Iacobucci J. provides the following reasoning at paras. 54-56:
The trial judge appeared to take Consolidated‑Bathurst [Consolidated‑Bathurst Export Ltd. v. Mutual Boiler and Machinery Insurance Co.,  1 S.C.R. 888] to stand for the proposition that the ultimate goal of contractual interpretation should be to ascertain the true intent of the parties at the time of entry into the contract, and that, in undertaking this inquiry, it is open to the trier of fact to admit extrinsic evidence as to the subjective intentions of the parties at that time. In my view, this approach is not quite accurate. The contractual intent of the parties is to be determined by reference to the words they used in drafting the document, possibly read in light of the surrounding circumstances which were prevalent at the time. Evidence of one party’s subjective intention has no independent place in this determination.
. . .
When there is no ambiguity in the wording of the document, the notion in Consolidated‑Bathurst that the interpretation which produces a “fair result” or a “sensible commercial result” should be adopted is not determinative. Admittedly, it would be absurd to adopt an interpretation which is clearly inconsistent with the commercial interests of the parties, if the goal is to ascertain their true contractual intent. However, to interpret a plainly worded document in accordance with the true contractual intent of the parties is not difficult, if it is presumed that the parties intended the legal consequences of their words.
37 The Lombard policy merely indicates that it will provide the insured with primary coverage unless other insurance is present, in which case it will provide only excess coverage regardless of the type of coverage provided by the other insurer. On its face, then, the Lombard policy provides the same primary coverage as that of Family. To endorse the intentions of one insurer over another, where both parties have sought to limit their liability to contribute and where the offending clauses, on their face, are irreconcilable, does violence to the intentions of the insurers and does not respect the obligation of both insurers to contribute. The better approach is that endorsed by English and the majority of Canadian courts.
38 Accordingly, were the Court to give effect to each insurer’s intention, the appellant Family would provide only excess coverage and the respondent Lombard would provide only excess coverage to that, leaving the insured without primary coverage at all. This result is obviously absurd. Thus, where the competing policies cannot be read in harmony, the most sensible course, and that which accords with the interest and expectation of both the insured and the insurers, is to treat the conflicting clauses as mutually repugnant and inoperative.
39 With the irreconcilable clauses rendered inoperative, the insurance policies of both the appellant and the respondent provide the insured with primary coverage. As a result, each insurer is independently liable to the insured for the full loss, as if the other insurer did not exist. Thus, what remains to be determined is the method by which the extent of each insurer’s obligation to contribute is calculated. The overwhelming view in Canada is that where liability is shared among insurers covering the same risk, the loss is borne equally by each insurer until the lower policy limit is exhausted, with the policy with the higher limit contributing any remaining amounts: Dominion of Canada v. Wawanesa, supra; McGeough, supra; Simcoe & Erie v. Kansa, supra; Guardian Insurance of Canada v. New Hampshire Insurance Co. (1997), 193 N.B.R. (2d) 111 (Q.B.).
40 The appellant argues that because the respondent’s policy provides a sharing formula, that formula should be followed and should apply to both insurers. The Family policy is silent with respect to the method of sharing whereas Lombard’s policy provides as follows:
If all of the other insurance permits contribution by equal shares, we will follow this method also . . . .
If any of the other insurance does not permit contribution by equal shares, we will contribute by limits. Under this method, each Insurer’s share is based on the ratio of its applicable limit of insurance to the total applicable limits of insurance of all Insurers.
I note that the Lombard policy limit is $5 million whereas the Family policy limit is $1 million. Were I to adopt the appellant’s view, Family would be liable to contribute one-fifth of the loss and Lombard would contribute four fifths.
41 It is trite law, however, to say that the provisions of a contract cannot unilaterally bind a person who is not a party to it. In the absence of privity between the two insurers, there is simply no basis for allowing Family to benefit from Lombard’s provision on sharing by limits. The beneficial impact of that provision on Family does not compensate for the lack of privity between the insurers.
42 Moreover, in my view, the “independent liability” approach to apportionment has become the prevailing view in Canada for good reason. In Dominion of Canada v. Wawanesa, supra, Proudfoot J. (as she then was) canvassed the authorities on the issue of apportionment and found the approach of Lawton L.J. in Commercial Union Assurance Co. v. Hayden,  1 Q.B. 804 (C.A.), most persuasive. Lawton L.J. provides an insightful review of the policy considerations at work that underlie the rationale for equal sharing of liability (at p. 822):
What is the burden under an indemnity liability policy? It is the claim which is made, not the claim which could be made. This kind of policy may be unlimited as to the amount of the indemnity, as it always is under motor car policies and often is under employers’ liability policies, or limited as it almost always is under professional negligence policies. Even when a policy limits the amount of the indemnity, it is a matter of judicial experience that most claims are well below the limit. To ascertain the proportions of contribution by reference to the limits of indemnity would, in my judgment, be an odd way of sharing the burden in equity between insurers; and in cases where there was a limit under one policy but none under another, it would be a difficult judicial task, probably an impossible one, to assess, as Mr. Dehn suggested should be done and Donaldson J. agreed (I quote from the judgment):
“the maximum cover which will ever be likely to be provided by the unlimited policy and to compare this figure with the sum specified in the other policies.”
Further, using the limits as the basis for apportionment of contributions would be unfair to the insurer who gave a much higher limit for a small increase in premium. This case provides an example. The plaintiffs gave the assured a limit of £100,000; the defendant a limit of £10,000. The difference in premium was £1.
It is a matter of my experience both as a judge and a practitioner that the assessment of premiums under liability policies is based on underwriting experience backed up by statistical information and actuarial projections. The risk underwritten is at its greatest with small claims and at its least with large claims. It follows that the upper limits for claims can be increased with only a small increase in premiums. When there are two insurers with differing upper limits for claims, the inference I would draw is that they were both accepting the same level of risk up to the lower of the limits. If this be so, in my judgment, “a rateable satisfaction,” to use Lord Mansfield C.J.’s phrase, would be an equal division of liability up to the lower limit; the burden of meeting that part of the claim above the lower limit would fall upon the insurer who had accepted the higher limit.
43 I find the reasoning of Lawton L.J. equally compelling and agree with Proudfoot J. that equal sharing of liability to the respective policy limits represents the fairest method of apportionment. Moreover, it accords with the principles on which the doctrine of equitable contribution is based. As Lord Mansfield C.J. put it so directly in Godin, supra, at p. 420:
As between them [the insurer and insured], and upon the foot of commutative justice merely, there is no colour why the insurers should not pay the insured the whole: for they have received a premium for the whole risque.
44 Finally, it is worth reiterating that the parties agreed at trial to an equal sharing of liability to the respective policy limits. Regardless of the status of that agreement when the parties appeared before this Court, it nevertheless offers some indication that this approach accords with the expectations of the parties.
45 In my view, McEwan J. applied the correct principles to this contest between insurers. He was correct in finding the two excess insurance clauses mutually repugnant and inoperative vis-à-vis each other. The appeal is allowed and the trial judge’s decision is restored with costs to the appellant throughout.
Appeal allowed with costs.
Solicitors for the appellant: Clark, Wilson, Vancouver.
Solicitors for the respondent: Branch MacMaster, Vancouver.
Solicitors for the intervener: Black, Sutherland, Crabbe, Toronto.