UNITED BRASS WORKS, INC., Plaintiff,
v.
AMERICAN GUARANTEE AND LIABILITY INSURANCE COMPANY and Continental Casualty Company, Defendants.
Civ. A. No. 91-97 ERIE.
United States District Court, W.D. Pennsylvania.
June 22, 1992.*466 Paul D. Shafer, Jr., Shafer Swick Bailey Irwin & Stack, Meadville, PA, for plaintiff.
Mary F. Licaris, Robert J. Bates, Jr., Michael J. Cunningham, Chicago, IL, for American Guarantee.
Stanley G. Berlin, William J. Kelly, Erie, PA, for Continental Cas.
MEMORANDUM OPINION
MENCER, District Judge.
Plaintiff United Brass and defendant American Guarantee have filed cross motions *467 for summary judgment. The central issue presented in both motions is whether American Guarantee, an insurance company, is obligated under the terms of a comprehensive general liability policy to defend and indemnify United Brass against claims that it is liable for the costs of cleaning up a hazardous waste dump.
I. FACTS
In 1973, American Guarantee, an insurance company located in Chicago, Illinois, issued a policy to United Brass that provided comprehensive general liability insurance. United Brass had purchased the policy through an insurance agent, John Hawkins of the Greensboro Insurance Agency, Greensboro, North Carolina. Mr. Hawkins' stamped signature appears on the policy, and the policy was presumably delivered to Mr. Hawkins, who countersigned the policy in North Carolina. At that time, United Brass, a North Carolina corporation, had its principal place of business and a manufacturing plant in Randleman, North Carolina. The American Guarantee insurance policy provided coverage for the facility in Randleman, North Carolina, and for two other United Brass warehouse and sales facilities in California and in New York. On January 27, 1975, United Brass purchased the assets of Keystone Brass Works. In October 1975, the parties agreed to include the former Keystone Brass facility located in Erie, Pennsylvania, as an insured risk on the American Guarantee policy. The American Guarantee policy was effective until January or February of 1976, when another insurer's policy became effective. Hawkins dep. at 73.
Since 1975, the Keystone foundry has employed a process involving molds formed from sand to produce manufactured cast brass valve components. Keystone uses brass alloys that contain primarily copper, tin, lead, and zinc. During the casting process, microscopic metal particles adhere to the molding sands. After a number of uses, the molding sands become suffused with these particles and can no longer be used as molds. Keystone then has an outside party pick up this used sand to haul it away.
During the first three months of 1975, Keystone employed Sitter Trucking Company to transport the waste sand away from its facility. In April 1975, Max Silver & Sons offered to haul away the sands for free. Mr. Silver believed that the sands contained valuable metals that could be recovered and that the profit gained from recovering metals would justify his bearing the expense of hauling the sand.
The record indicates that Silver hauled the waste sand to the Millcreek Dump Site, located in the outskirts of Erie.[1] There is, however, no record of which particular dates or on how many occasions Silver came to take the waste away, Silver continued to haul the waste away until 1981, when Independent Iron & Metal Company began to transport the sand away, also to the Millcreek site.
In August 1980, the Pennsylvania Department of Environmental Resources notified Albert Fuchs, the lessee of the Millcreek site, that his use of the property as a slag dump-site without a permit was illegal. In August 1983, the United States Environmental Protection Agency (EPA) gave notice to Ralph Riehl, the owner of the site, that it was planning a clean-up of the site based on testing that it had conducted. Riehl v. Travelers Ins. Co., 772 F.2d 19, 21 (3d Cir.1985). On April 18, 1985, the EPA wrote to the Keystone Foundry Division of United Brass and requested information concerning Keystone's involvement with the site. On October 16, 1989, the EPA filed a civil action against several parties it considered responsible for the site. On January 4, 1990, United Brass was joined as a third-party defendant by certain of the defendants.
II. ANALYSIS
The insurance policy at issue, a "Special Multi-Peril Insurance Policy," provides comprehensive general liability insurance to United Brass. The policy states:
*468 [American Guarantee] will pay on behalf of the insured all sums which the insured shall become legally obligated to pay as damages because of bodily injury or property damage to which this insurance applies, caused by an occurrence ... and [American Guarantee] shall have the right and duty to defend any suit against the insured seeking damages on account of such bodily injury or property damage, even if any of the allegations of the suit are groundless, false, or fraudulent ...
An "occurrence" is defined as:
... an accident, including injurious exposure to conditions which results, during the policy period, in bodily injury or property damage neither expected nor intended from the standpoint of the insured.
American Guarantee contends that the facts of this case do not indicate that there has been an "occurrence" within the period that the policy was effective and that it has no duty to defend or to indemnify United Brass. United Brass argues that the facts establish exactly the opposite conclusion.
A. Choice of Law Analysis
The first substantive question presented is whether Pennsylvania or North Carolina law should be applied to resolve the issues raised in this case. In diversity cases such as this one, a federal court must apply the choice of law principles of the forum state in which it sits. Klaxon Co. v. Stentor Elec. Mfg. Co., Inc., 313 U.S. 487, 61 S. Ct. 1020, 85 L. Ed. 1477 (1941); Melville v. American Home Assurance Co., 584 F.2d 1306, 1308 (3d Cir.1978). Therefore, the Pennsylvania choice of law principles must be consulted to determine whether Pennsylvania or North Carolina law governs the remainder of the case.
The courts are divided over the question of what precisely the Pennsylvania choice of law principles are. Pennsylvania was one of the first jurisdictions to abandon the common law doctrines of lex loci delicti and lex loci contractus. The case that rejected these old approaches, Griffith v. United Airlines, Inc., 416 Pa. 1, 203 A.2d 796 (1964), was decided soon after the first states began to adopt new approaches to questions of choice of law. See, e.g., Grant v. McAuliffe, 41 Cal. 2d 859, 264 P.2d 944 (1953); Kilberg v. Northeast Airlines, Inc., 9 N.Y.2d 34, 211 N.Y.S.2d 133, 172 N.E.2d 526 (1961). The Griffith decision announced Pennsylvania's adoption of a "flexible rule" that "permits analysis of the policies and interests underlying the particular issue before the court." 203 A.2d at 805.
The problem with this case is that it cites cases that adopt several different types of choice of law analysis: the substantial relationship approach of the Second Restatement, the interest analysis method, and possibly the "center of gravity" approach. Although there have been a number of choice of law cases decided since Griffith, the Pennsylvania Supreme Court has never made a definitive statement regarding which approach is to be used in which types of cases. The attendant confusion among courts applying Pennsylvania choice of law rules has led one commentator to remark that "[n]o state has a more convoluted, eclectic approach to choice of law than Pennsylvania." Smith, "Choice of Law in the United States," 38 Hastings L.J. 1041 (1987).
Despite the confusion, the federal courts appear to have settled on interpreting the Griffith language to mean that a court applying Pennsylvania law should analyze a choice of law problem using two of the approaches, rather than simply one. This inclusive approach has been termed a "flexible" analysis. See Compagnie des Bauxites v. Argonaut-Midwest Ins. Co., 880 F.2d 685, 688 (3d Cir.1989); American Contract Bridge League v. Nationwide Mut. Fire Ins. Co., 752 F.2d 71, 74 (3d Cir.1985). Because the Griffith court discussed the Second Restatement and applied some of its rules in resolving the choice of law question presented by the case, several federal courts have used the Second Restatement approach as the starting point for their choice of law analysis. See Compagnie des Bauxites, 880 F.2d at 689-691; Melville v. American Home Assurance Co., 584 F.2d 1306, 1308-09 (3d Cir.1978); American Int'l Underwriters Corp. v. Zurn Indus., Inc., 771 F. Supp. 690, 694-95 (W.D.Pa.1991). In addition to applying the Restatement analysis, these courts have also *469 employed some sort of interest analysis because the Griffith opinion indicates that its rule requires "analysis of the policies and interests underlying the particular issue before the court" and because, in announcing its new rule, Griffith cites language from Babcock v. Jackson, 12 N.Y.2d 473, 240 N.Y.S.2d 743, 191 N.E.2d 279 (1963), the opinion that adopted a type of interest analysis for New York. Griffith, 203 A.2d at 805-06.
In the absence of any definitive statement from the Pennsylvania Supreme Court regarding choice of law rules, this court will use the federal approach, which provides a satisfactory accommodation of the Griffith language. Under the federal approach, a court should first look to the Second Restatement.
The Second Restatement is a set of rules that range from rules of very general application, which are to be applied to all choice of law problems, to rules that apply to only very specific cases and that essentially preempt the more general rules. Section 193, which applies to "contracts of fire, surety or casualty insurance," is of the latter type. Section 193 creates a presumption that the state which contains the principal location of the risk has the most substantial relationship to any disputes regarding insurance coverage for that particular risk.
In this case, the insured risk, the Millcreek site, is located in Pennsylvania. However, the Millcreek site is not the principal location of the risk insured by the American Guaranty policy. That policy insures several different risks located in several different states. Accordingly, there is no principal location of the risk and section 193 is inapplicable. See Zurn Industries, 771 F.Supp. at 695-96. Furthermore, Comment b to section 193 indicates that less weight should be conferred to the location of the risk if the policy insures risks located in more than one state.
As noted above, section 193 indicates that in the area of casualty insurance contracts, the law of the location of the principal risk insured under the contract should ordinarily be given determinative weight. A reader of the Restatement might well be surprised to find that the location of the risk is given determinative weight if it is the principal risk insured under the contract, and yet very little weight is given to this same factor if the contract insures multiple risks scattered over several states. One would expect that the reasons for conferring such weight upon the location of the subject matter of the contract would apply equally well regardless of whether the subject matter of the insurance contract is the primary subject of the contract or not. And indeed, the reasons cited for the adoption of section 193's presumption stem from the fact that all states have an interest in regulating property located within their boundaries.[2]
Nevertheless, section 193 and its comments make clear that its provisions are largely inapplicable if the risk in question is not the policy's primary insured risk. The reason for the distinction appears to be a preference that only one set of laws govern a given insurance contract, and a disapproval of the possibility that the laws of different jurisdictions might apply to different risks under the policy. Although some courts have suggested that the latter rule would be little more trouble than the former, and have indicated that requiring that the law of only one jurisdiction apply to a contract does not further uniformity or predictability of the laws, see e.g., Johnson Matthey Inc. v. Pennsylvania Mfr. Association Ins. Co., 250 N.J.Super. 51, 593 A.2d 367, 371 (1991), such a conclusion is contrary to the considered opinion of the drafters of the Restatement. Furthermore, there is little reason to suspect that the Pennsylvania Supreme Court would not adopt this particular section of the Restatement.[3]
*470 Therefore, under the Restatement approach, section 193 is inapplicable and this question is governed by the "contacts with the jurisdiction" inquiry under section 188. Restatement section 188 identifies five contacts that a court should evaluate in deciding which law to apply: the place of contracting; the place of negotiation of the contract; the place of performance; the location of the subject matter of the contract; and the domicil, residence, nationality, place of incorporation and place of business of the parties.
In this case, most of the contacts are with North Carolina, not Pennsylvania. The insurance contract was countersigned, and thus finalized, by United Brass' insurance agent, Mr. Hawkins, in North Carolina. The record does not disclose where the contract was negotiated, although it was probably negotiated either in North Carolina (the headquarters of the insured and the insured's agent) or Illinois (the headquarters of the insurer). Some of the subject matter of the contract, in other words the insured risks, are located in North Carolina, but others are located in Pennsylvania, New York, and California. However, the subject matter that is at the heart of this dispute is the location in Erie, Pennsylvania. The headquarters and state of incorporation of the insured is in North Carolina; and the headquarters and state of incorporation of the insurer is Illinois.
Although a simple tally of the contacts indicates that North Carolina has the greatest relationship with the contract, the Restatement cautions that such an analysis is not the intent of the drafters of the Restatement because an undifferentiated application of the formula to all situations would yield unjust results in too many circumstances. "These contacts are to be evaluated according to their relative importance with respect to the particular issue." Restatement (Second) of Conflict of Laws, § 188(2). However, even with this cautionary instruction in mind, the Restatement still favors the application of North Carolina law. The only contact between the insurance contract and Pennsylvania is the location of the insured risk. And, as noted above, while this is an important contact, it is not a determinative factor standing alone.
Under the federal interpretation of Pennsylvania's choice of law rules, a court should also undertake an inquiry into whether one state has a greater policy interest in the resolution of the dispute than another. In this case, the central question is the meaning of an "occurrence" under an insurance policy. The first step in the interest analysis approach is to determine how each state would resolve the dispute and to ascertain the policy justifications underlying each set of governing rules.
Although it appears that no Pennsylvania state courts have ever ruled on the issue, federal courts applying Pennsylvania law have held that Pennsylvania would adopt an actual time of injury rule for triggering coverage. In Armotek Indus., Inc. v. Employers Ins. of Wausau, 952 F.2d 756 (3d Cir.1991), the Third Circuit Court of Appeals rejected the contention of an insured that an insurance policy that took effect in 1979 would provide coverage for a spill of 1600 gallons of chromic acid that occurred in 1977. The "occurrence" happened at the time of the spill, which, the court surmised, was the point at which most of the damage was done. The court dismissed the notion that the "triple trigger" or "continuous trigger" rule should be applied.[4] The Armotek court limited the application of the "triple trigger" rule to asbestos cases only, noting that "persons who suffer from asbestos-related diseases are *471 often exposed to asbestos for some period and generally do not manifest symptoms until some later point." 952 F.2d at 763. Thus the court held that, as a general rule, the "occurrence" happens at the time of the dumping.[5],[6]
In contrast to Pennsylvania, North Carolina has made a clear choice in favor of the manifestation or discovery rule for determining when an occurrence arises in environmental spill cases. In West American Ins. Co. v. Tufco Flooring East, Inc., 104 N.C.App. 312, 409 S.E.2d 692 (1991), the Court of Appeals of North Carolina held that "the general rule is that, for insurance purposes, property damage "occurs" when it is manifested or discovered." 409 S.E.2d at 695. In so holding, the court relied on two earlier federal cases, Peerless Ins. Co. v. Strother, 765 F. Supp. 866, 870 (E.D.N.C. 1990), which predicted that North Carolina would adopt the manifestation/discovery rule, and Mraz v. Canadian Universal Ins. Co., 804 F.2d 1325 (4th Cir.1986), which predicted that Maryland would adopt this rule.
The task of identifying policy rationales for these rules is not an easy one. The cases predicting that Pennsylvania would adopt the actual occurrence rule base their opinions on a "common sense" reading of the language. "An occurrence arose each instance the wastes were released onto the ... site. This is the time that the injurious effect took place because this is the time that the property was actually damaged." Centennial, 677 F.Supp. at 346. See also Continental Ins. Co. v. Northeastern Pharm. and Chem. Co., Inc., 811 F.2d 1180, 1190-91 (8th Cir.1987) (court applying Missouri law reasoned that an "occurrence" should conform to the time when the elements of a lawsuit are complete and indicated that in cases involving improper hazardous waste disposal, the wrong and the resulting damage are usually contemporaneous).
The adoption of the manifestation or discovery rule in North Carolina is based on a similar concern, but on the further observation that in hazardous waste cases, the damage may be ongoing. In Mraz, the opinion cited favorably by the West American and the Peerless courts, the court stated that the "general rule is that the time of the occurrence of an accident ... is not the time the wrongful act was committed but the time when the complaining party was actually damaged ..." The court went on to observe that "there are situations, however, in which the existence or scope of damage remains concealed or uncertain for a period of time even though damage is occurring. The leakage of hazardous wastes as in this case is a clear example. In such cases we believe that the better rule is that the occurrence is deemed to take place when the injuries first manifest themselves." 804 F.2d at 1328.
The reasons underlying both the Pennsylvania and North Carolina rules are essentially the same. Both rules seek to define an occurrence in the legal terms of when the plaintiff could theoretically have sued the insured. Both rules also recognize that defining exactly when the damage is done in hazardous waste cases can be difficult. The rules differ only in their estimation of when *472 the bulk of the damage happens. The Centennial and Northeastern courts think that most of the damage is done when the spill occurs. The Mraz court thinks that there is a sufficient amount of damage that may occur after the initial spill to warrant a rule that the occurrence does not exist until it is discovered.
Such concerns cannot be said to implicate the substantial interests of either state. In both cases, the rules have no real policy rationale. Instead, they function to "draw a line" between a situation in which an insurance company is liable to its insured and a situation in which it is not. Neither rule represents an attempt to expand or contract coverage under insurance policies. For example, if one of the jurisdictions had adopted a "triple trigger" approach, an argument could be made that the rule expresses a procoverage policy because it increases the number of times that an "occurrence" could be said to have happened. But both the Pennsylvania and the North Carolina rules fix an occurrence as happening only at one point in time.
Some courts applying an interest analysis would look at whether any interests of one state would be harmed by the application of the other's law, regardless of whether those interests underpin the applicable laws. If such a perspective were to govern this case, a court might conclude that Pennsylvania stands to lose more by the application of North Carolina's law to this dispute than vice versa because of the possibility that the cost of clean up might exceed the polluter's assets and Pennsylvania might be required to fund the clean up. This court declines to adopt such a perspective. First, the record contains no suggestion that the cost of clean up will bankrupt the plaintiff. Second, and more important, Pennsylvania's "trigger rule," as expressed by the Third Circuit, was not fashioned with this sort of state interest in mind.[7]
Thus, neither Pennsylvania nor North Carolina have any substantial interest in having its law applied. In a jurisdiction that has adopted a pure interest analysis approach, there are further sets of rules for choosing a governing set of laws when no state has a substantial interest in having its law applied. However, because it appears that Pennsylvania has adopted a mixed Second Restatement and interest analysis approach, it seems appropriate to defer to the result clearly dictated by an application of the Second Restatement, rather than apply a result reached under interest analysis's "tie breaking" rules.[8]
Accordingly, this court finds that the law of North Carolina governs the question of when an accident is said to occur under the insurance policy.
B. "Occurrence" Under the Insurance Policy
As noted previously, North Carolina law provides that an accident "occurs" when it is discovered or manifested. The facts of this case are quite similar to those in the Peerless case, supra. In Peerless, the insured ran a business in which it obtained scrap metal from used electrical transformers. In this process, the insured released materials containing polychlorinated biphenyls ("PCBs") into the air and onto the ground. After the period in which the insurer provided coverage, both state and local authorities began to investigate complaints of tainted drinking water. 765 F. Supp. at 867-68. The court held that the accident had not been discovered until after the policy period, *473 and ruled in favor of the insured. Id. at 870-71.
In this case, the uncontested facts are that the dumping was ongoing throughout the 1970s, but the earliest date on which it could be said that the violations were discovered was 1980, when the Pennsylvania Department of Environmental Resources contacted Albert Fuchs. The American Guarantee policy ceased to be effective as of early 1976. Thus, there was no occurrence within the meaning of the American Guarantee policy, and American Guarantee has no duty to defend or indemnify.
NOTES
[1] There is also testimonial evidence that Sitter hauled contaminated sand to the Millcreek Dump Site. Fuchs dep. at 43-44; Gray dep. at 58-59.
[2] Comment c to section 193 lists three reasons: first, it is the state in which the risk is located that also creates the obligations and rights of the parties toward that risk; second, it is likely that the parties would have assumed that the law where the primary risk was located would control insurance coverage questions as well; and third, the state in which the risk is located has a "natural interest" in the resolution of coverage questions.
[3] The Restatement's preference for one law to apply to multiple risk policies is underscored by the narrowness of the single exception to the rule contemplated by the comments. Comment f of section 193 indicates that a multiple risk policy should be treated as separate policies subject to the laws of the various states in which the risks are located if, as is often the case in fire insurance, the various states have statutory requirements regarding the form or language of the policy. Obviously, no such statutory requirements are present in this case.
[4] In ACandS, Inc. v. Aetna Cas. and Surety Co., 764 F.2d 968 (3d Cir.1985), the Third Circuit had adopted the "triple trigger" or "continuous trigger" rule to find that bodily injury was caused by asbestos at the time of exposure to the asbestos, at the time of "exposure in residence," or when "the further progression of injury ... occurs even after exposure has ended," and at the time when the injury is manifested. Id. at 971.
[5] Some language in the Armotek decision could be construed to mean that if the insured could show that some further injury to the property occurred at some point after the dumping, the further injury might constitute a separate "occurrence" under a policy. Armotek, 952 F.2d at 763. However, just as in the Armotek decision, in this case the insured has made no such argument and so this possibility will not be considered. Id.
[6] There are a few federal cases interpreting Pennsylvania law that have suggested that Pennsylvania might adopt a manifestation or discovery rule for triggering an "occurrence" under an insurance policy if the facts do not establish when the dumping actually took place. In Centennial Ins. Co. v. Lumbermens Mut. Cas. Co., 677 F. Supp. 342 (E.D.Pa.1987), the court construed the opinion in Riehl v. Travelers Ins. Co., 772 F.2d 19 (3d Cir.1985) to mean that an occurrence arose when the effects are manifested, not when the property was actually damaged, "where the record did not disclose when the dumping commenced, occurred or when the pollution was discovered." Centennial, 677 F.Supp. at 346-47. Interestingly, the Riehl litigation involved the same site that is the subject of this case. However, this court's independent reading of Riehl indicates that the Riehl court stated that such a rule might apply if the district court determined, on remand, that there was no evidence regarding when the dumping had actually taken place.
[7] The Griffith decision made the following observation on the spate of commentary that influenced the court to reject the old place of wrong and place of contracting approaches:
Although [the various commentators] may differ slightly in the manner in which these contacts and relationships are to be evaluated, all place importance upon an analysis of the policies underlying the conflicting laws and of the relationship of the particular contacts to those policies.
(Citations omitted) (emphasis added), 203 A.2d at 802. The Griffith court perceived interest analysis as focusing on the "relationship" between a state's contacts with the dispute and the policies underlying the applicable laws.
[8] These rules can be quite arbitrary. For example, Professor Brainerd Currie, one of the founders of interest analysis, advocated that in such cases a court should automatically apply the law of the forum. See Brilmayer, Conflict of Laws, Foundations and Future Directions (1991) at 58.