Orleans v. Commercial Union Insurance

Brock, C.J.

The plaintiff, David Orleans, seeks a determination that our holding in Cacavas v. Maine Bonding & Casualty Co., 128 N.H. 204, 512 A.2d 423 (1986) should be applied retroactively. Cacavas overruled prior precedent preventing intra-policy “stacking” of uninsured motorist coverage. Seeking to “stack” the uninsured motorist benefits for two vehicles under a single policy issued by the defendant, Commercial Union Insurance Company, the plaintiff appeals from an order of the Superior Court (Goode, J.) ruling that Cacavas should not be applied retroactively. For the reasons that follow, we reverse.

The facts giving rise to this action are undisputed. In January, 1985, the plaintiff suffered serious injuries in an automobile accident caused by the negligence of an underinsured driver. The motorist had an insurance policy with a liability limit of $50,000, and the plaintiff has recovered that amount.

When the accident occurred, the plaintiff was riding as a passenger in a car owned and operated by David Jasinski, and was an “insured” within the meaning of the uninsured/underinsured motorist protection (hereinafter “UM benefits”) of Mr. Jasinski’s automobile insurance. The policy was issued by Commercial Union and provided coverage for two automobiles owned by Mr. Jasinski, including the vehicle he was operating when the plaintiff was injured. Commercial Union charged a separate premium for each car insured under the single policy, and each car was covered by UM benefits in the amount of $100,000 per person and $300,000 per accident. The policy contained a standard “Limits of Liability” clause as follows:

“Limits of Liability.

(a) The limit of liability for uninsured motorists coverage stated in the declarations as applicable to ‘each person’ is the limit of the company’s liability for all damages, including damages for care or loss of services, because of bodily injury sustained by one person as the result of any one accident and, subject to the above provision respecting each person, the limit of liability stated in the declarations as applicable to ‘each accident’ is the total limit of the company’s liability for all damages, including damages for care or loss of services, because of bodily injury sustained by two or more persons as the result of any one accident.”

Claiming the right to recover under the UM provisions insuring both automobiles owned by Mr. Jasinski, the plaintiff filed a claim with Commercial Union requesting payment of $200,000, an amount reflecting the sum of the intra-policy limits for each car. Commercial Union denied liability, arguing that the “Limits of Liability” language contained in its policy prohibited “stacking” of UM benefits.

Following a hearing, the trial court ruled that stacking was not permitted. Although acknowledging that the holding in Cacavas established “a new principle of law by overruling past precedent” prohibiting intra-policy stacking, the court ruled that the Cacavas holding required “nonretroactive application,” and that the limit of UM coverage available to the plaintiff was $100,000, minus any payments already received. This appeal followed.

The plaintiff seeks a determination that the “Limits of Liability” clause does not limit Commercial Union’s liability to the amount of UM benefits declared per vehicle, but instead, permits an insured to aggregate the liability limits for each of the two vehicles, or, in short, to “stack” the UM benefits, for total available coverage of $200,000 per person and $600,000 per accident. Relying upon our decision in Cacavas, the plaintiff claims that the standard “Limits of Liability” clause is ambiguous and should be construed against the defendant-insurer in order to honor the reasonable expectation of the policyholder that a policy which required payment of double premiums provided double coverage. The only issue we consider is whether our decision in Cacavas has retrospective application.

In Hampton National Bank v. Desjardins, 114 N.H. 68, 314 A.2d 654 (1974), we referred to a three-part test applied by the Supreme Court of the United States in its cases dealing with the question of retroactivity. Id. at 75, 314 A.2d at 658; see Opinion of the Justices, 131 N.H. 644, 650, 557 A.2d 1364, 1368 (1989). The United States Supreme Court considers three separate factors in determining whether its decisions should be applied prospectively. Chevron Oil Co. v. Huson, 404 U.S. 97, 106-07 (1971). To be applied prospectively a decision must, first, “establish a new principle of law, either by overruling clear past precedent on which litigants may have relied ... or by deciding an issue of first impression whose resolution was not clearly foreshadowed.” Id. Second, “we must weigh the merits and demerits in each case by looking to the prior history of the rule in question, its purpose and effect, and whether retrospective application will further or retard its operation.” Id. Finally, “where a decision could produce substantial inequitable results if applied retroactively, there is ample basis . .. for avoiding the ‘injustice or hardship’ by a holding of nonretroactivity.” Id. (citations omitted); cf. Hampton Nat’l Bank v. Desjardins supra.

Arguing that we have adopted the Chevron test, the plaintiff contends, inter alia, that our decision in Cacavas was “foreshadowed in several decisions of this court,” thus precluding prospective application under the first Chevron factor. Although we have never formally adopted the Chevron test, we nevertheless find its application useful in the case before us. See Opinion of the Justices, 131 N.H. at 650, 557 A.2d at 1368.

The first prong of the Chevron test questions the extent to which there was “clear past precedent” from which the new ruling worked an abrupt and clear break. Although our decision in Cacavas did overrule past precedent, we cannot confidently state that at the time of our decision in Cacavas New Hampshire law on the issue of intrapolicy stacking had become so settled that it could be accurately characterized as “clear.” To the contrary, at the time that Cacavas was decided, the law with respect to stacking was in ferment, discernibly shaken by a change in the standard of construction which had governed our former decisions interpreting insurance contracts.

In Grimes v. Concord General Mutual Insurance Co., 120 N.H. 718, 422 A.2d 1312 (1980), this court first considered whether an insured is entitled to stack the uninsured motorist benefits contained within a single policy that insures two cars. Id. at 720, 422 A.2d at 1314. Noting that the issue was one of “first impression in this State,” we held that the answer was “no.” Id. at 720-24, 422 A.2d at 1316. Although we recognized that the payment of double premiums was a factor to be considered in determining the reasonable expectations of the insured, we held that the standard “Limits of Liability” clause prohibiting stacking of benefits was “sufficiently clear that the ‘ordinarily intelligent insured’ after a ‘more than casual reading’ would understand it to limit the carrier’s liability.” Id. at 722-23, 422 A.2d at 1315 (citation omitted).

At the time that Grimes was decided, New Hampshire had not yet expressly adopted the rule of construction that ambiguous terms in an insurance policy are construed against the insurer and in favor of the insured. See Trombly v. Blue Cross/Blue Shield, 120 N.H. 764, 769, 423 A.2d 980, 984 (1980); Cacaras v. Maine Bonding & Casualty Co., 128 N.H. at 207, 512 A.2d at 425. Rather, Grimes was decided when the rule in this State regarding construction of insurance policy language created no presumption in favor of either the insurer or the insured, directing simply that insurance contracts were to be interpreted “ ‘in the light of what a more than casual reading of the policy would reveal to an ordinarily intelligent insured.’” See Trombly v. Blue Cross/Blue Shield, 120 N.H. at 770, 423 A.2d at 984 (quoting Aetna Insurance Co. v. State Motors, 109 N.H. 120, 125, 244 A.2d 64, 67 (1968)). In the interim between our decision in Grimes and our decision in Cacaras, we observed that our former rule of construction was based upon the “unfounded” assumption that the “reasonable insured both reads the policy and comes to some understanding of its terms.” Trombly v. Blue Cross/Blue Shield supra. Reasoning that “men in general cannot read and understand ... insurance documents,” we held in Trombly that doubtful language in an insurance policy should be construed most strongly against the insurer. Trombly supra.

Subsequently, we held in Descoteaux v. Liberty Mutual Insurance Co., 125 N.H. 38, 480 A.2d 14 (1984) that an insurance company might preclude the stacking of coverage under two separate policies by including clear and unambiguous policy language to that effect, but that, in the absence of any language specifically barring stacking, an other-insurance clause or limits-of-liability clause could be read merely as limiting the insurer’s proportional share of the total coverage under all applicable policies, thus entitling the insured to stack coverage under two separate policies in return for premiums paid. Id. at 42-43, 480 A.2d at 17-18; see Cacaras v. Maine Bonding & Casualty Co., 128 N.H. at 208, 512 A.2d at 426 (Souter, J., concurring specially). Given that our decision in Grimes limiting UM benefits was predicated upon a limits-of-liability clause that did not expressly bar stacking, Descoteaux effectively overruled Grimes, thereby putting the insurance industry on notice that an insured could reasonably expect to stack UM benefits unless informed by fairly specific policy language that the payment of additional premiums would not buy increased protection in the form of combined coverage.

We are therefore not persuaded that our decision in Cacavas was a novel one, signalling an abrupt change in clearly settled law. It is true that our decision in Cacavas held that a standard “Limits of Liability” clause, identical to the one incorporated into the policy at issue in Grimes, was ambiguous with respect to stacking, and in view of the ambiguity, required a construction in favor of the insured and against the insurer. Cacavas v. Maine Bonding & Casualty Co., 128 N.H. at 207, 512 A.2d at 425. Nevertheless, Cacavas did not constitute a clear break from past precedent. The standard applied in Cacavas to determine the reasonable expectations of the insured was first announced in Trombly, and the novelty of that decision had worn off sufficiently to assure its precedential weight and to render reliance upon our old rule of construction, as applied in Grimes, unreasonable. Accordingly, while Cacavas did not write on a clean slate, neither did it overrule clear past precedent upon which litigants may reasonably have relied, and the first Chevron factor weighs in favor of retroactive application.

The defendant fares no better under the second Chevron factor, which considers whether retrospective application will further or retard application of the rule. Our decision in Cacavas was based on the rationale that the purpose of an insurance policy is to provide indemnity against loss, and that clauses limiting an insurer’s liability should not be susceptible to conflicting interpretations. By holding that an insurance company “remains free to limit its liability through clear and unambiguous policy language,” we intended to promote the use of plainer language in the drafting of insurance contracts and to honor the reasonable expectation of the insured that, in the absence of fairly specific language to the contrary, UM benefits may be stacked up to the amount for which premiums have been paid. Cacavas v. Maine Bonding & Casualty Co., 128 N.H. at 208, 512 A.2d at 425. That end would be better served by retrospective application, and consequently, the second Chevron factor weighs in favor of retroactivity.

Finally, we conclude that the third Chevron factor counsels in favor of retroactive application. Well before the cause of action here accrued, subsequent precedent had drained our decision in Grimes of most of the vitality it may once have possessed. Indeed, a fair reading of our subsequent cases indicates that the rule announced in Cacavas is one which was foreshadowed, and hence, could have been anticipated. Manifestly, it cannot be said that justice demands a purely prospective application, and we conclude, accordingly, that on balance, the equities favor application of Cacavas to the plaintiff’s claim.

Reversed.

Thayer and Horton, JJ., did not sit; the others concurred.