RISKVUE ARCHIVE | FEATURE STORIES

Allocation of Continuous Damage Losses, Part 1

By Stuart Cotton and Philip C. Silverberg

Insurance coverage litigation has changed dramatically in recent years. This is especially the case when damage cannot be ascribed to a particular event or occurrence. As one might expect, the evolving concept of what constitutes an insurable event or “occurrence” has generated a considerable amount of legal analysis in which various theories have been propounded in an effort to rationalize insurance coverage in the context of complex factual scenarios. In addition to examining whether these less easily quantifiable “injuries” or “damages” are indeed subject to coverage in the first instance, when successive coverage periods are potentially at issue because of alleged ongoing damage it becomes necessary to determine which coverage is implicated. That inquiry may include analysis of the impact of noninsured time frames as well.

As articulated by many courts, the problem, quite simply, is that certain types of injury or damage “cannot be easily allocated to any particular time period or insurance policy, because the cause of the injury cannot be isolated with any degree of mathematical or scientific nicety.”1 In the case of an injury that is ongoing in nature, “the specific cause of a continuous harm cannot be parceled out with any scientific accuracy such that it could be said, for example, that a certain percentage of the ultimate injury was caused during [a particular] year of exposure.”2

It is not unusual in coverage litigation over environmental or toxic claims and damage that the insured comes to court seeking indemnity for costs and expenses related to conditions or circumstances that may have existed for a lengthy period of time. Thus, for example, an industrial concern may seek coverage to clean up a toxic dump that was present decades before the policy or policies under which coverage is sought incepted; that dump may not even have been utilized during the coverage periods at issue. In other instances, environmental damage might be the product of long-term activities that commenced prior to the period of coverage at issue and continued not only throughout the effective dates of the coverage but also for some time thereafter. In either case, a court must determine how liability for a covered loss should be distributed among the various insurers and—where periods of non-coverage may be involved—the insured.

As discussed below, many courts have held that coverage for continuous damage over multiple policy periods should be assigned to the policy or policies in effect at certain key points in time during the continuum over which the loss occurred. Other courts have held that there is no reasonable basis for limiting coverage for continuous damage only to certain policies. However the issue is approached and ultimately resolved, the analysis can vary greatly from jurisdiction to jurisdiction. Although certain trends are evident, the law is far from settled.

Theories of Coverage for Continuous Loss

Most insurance policies predicate coverage upon the existence of an occurrence. Thus, a particular insurance policy will provide coverage only if an occurrence, as defined in the policy at issue, transpired during the policy period.3 An “occurrence” is the event that implicates coverage and obligates the insurer to provide the insured with the compensation for which it contracted. The event or series of events that gives rise to the application of an insurance policy (particularly in the liability insurance context) has been referred to as the “trigger” of coverage. Traditionally, the notion of trigger was quite simple: when a covered event caused a loss, any insurer of the risk at the time of that event was potentially liable for the loss, subject to applicable policy limits.

The manner in which a policy trigger should operate has evolved in recent decades as a result of the increasing number of claims arising from losses alleged to have occurred over prolonged periods of time during which a number of different policies were in effect. Under such circumstances, a determination of when the loss is deemed to have occurred is crucial to determining which parties are potentially responsible for the loss. Courts have applied a number of theories in coverage cases involving continuous damage.4

Exposure

Under the exposure theory, it is the initial exposure to the agency or to the release of the element causing the damage that constitutes the occurrence which implicates coverage. This theory posits that no damage occurs prior to the initial exposure or release.5 In matters involving long-term environmental or toxic damage, application of this theory limits coverage to the policy or policies in effect at the time of the exposure; the insured has no rights under policies providing coverage during other periods, notwithstanding that those policies may have been in effect during periods when some of the damage occurred.6

Manifestation

The manifestation theory posits that the occurrence, for coverage purposes, takes place when the injury becomes ascertainable or identifiable. In other words, only a recognizable injury can implicate coverage under a particular policy.7 An application of the manifestation trigger thus limits coverage to the policy or policies in effect at the time of manifestation.

Double Trigger

The double trigger combines the exposure and manifestation theories. A court applying a double trigger would hold that all policies in effect at either the time of exposure or the time of manifestation are potentially liable for the covered loss.8

Injury-in-Fact

According to the injury-in-fact (or actual injury) theory, the occurrence is the injury itself.9 This theory rests on the notion that “[a]t all times before that instant [of injury], the loss was only a potential which had not yet been realized.”10 Coverage is limited to the policy or policies in effect at the time that the injury or injuries occurred. Thus, for a continuous damage loss, coverage would be provided by all policies in effect during the time when damage took place. Because continuous damage cases often involve latent damages, however, the determination of when the damage occurred can be difficult.11

Continuous Trigger

Under the continuous (or triple trigger) theory, the occurrence that implicates coverage takes place at all times from the first exposure until manifestation or last injury.12 It follows that the continuous trigger approach implicates all policies that conceivably provide coverage under any of the theories of trigger.

The current trend in the third-party liability insurance context seems to favor application of the continuous trigger approach when there has been indivisible damage over multiple policy periods.13 Those advocating this approach argue that it furthers the “public policy” of maximizing coverage and does not demand fact determinations which are often extremely burdensome or even impossible to make under the circumstances.14

Number of Occurrences

Inasmuch as insurance contracts typically provide coverage for “occurrences,” when continuous damage is involved the determination of the number of occurrences will have a potentially significant impact upon the amount of coverage provided and the amount recoverable.15

In response to the problems presented when numerous “occurrences” are found in each policy period, most courts have held that continuous damage from one cause or at one location constitutes a single “occurrence,” notwithstanding that multiple, yet individually identifiable, “events” may have taken place. The rationale for this approach was stated with clarity by the court in Uniroyal,16 which ruled that repeated deliveries of Agent Orange were so “numerous, uniform, routinized and regularized, at such steady and frequent intervals that they merged into one continuous and repeated event.”

Allocation

Once a court has determined that it will apply a continuous trigger approach (or any other approach utilizing multiple triggers) to damage occurring over several policy periods, the court will be faced with allocation issues, including how to apportion the loss among the various insurers and determining who will bear responsibility for uninsured losses. A majority of courts utilize two basic approaches: (1) joint and several liability and (2) pro rata allocation.17

Joint and Several Liability

Under a joint and several liability allocation scheme, each triggered policy is potentially liable for all injuries sustained by the insured. Generally, courts that have imposed joint and several liability have based their decisions on the “all sums” language present in comprehensive general liability (CGL) policies.18 That language provides that the CGL insurer 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.19

Joint and several liability is an attractive allocation option for insureds. Because each triggered policy is liable for the insured’s entire loss under such a theory, a court will allow the insured to choose from which policy or policies it will recover. Moreover, with joint and several liability, the insured bears no risk for uninsured loss; each insurer is responsible for the entire loss.20

Joint and several liability can result in a number of inequities. First, taken to the extreme, a policy that may provide the insured with only a single day of triggered coverage is—theoretically, at least—as exposed to the insured’s loss as a policy that provides multiple years of triggered coverage or even a series of policies that provide such coverage. Moreover, joint and several liability gives a windfall to insureds that fail to obtain comprehensive coverage; application of joint and several liability allows an insured that obtained coverage during only one year of a twenty-year continuous loss to enjoy the same benefit of coverage as the insured that obtained—and paid for—coverage for all twenty years of the loss.21

Pro Rata Allocation

With pro rata allocation, each insurer is liable for its proportionate share of the insured’s loss. Because reasonable minds may differ as to what constitutes a “proportionate share,” a variety of approaches to this allocation scheme have developed.

As a general (although not universal) rule, courts that impose a pro rata method of allocation require the insured to bear responsibility for uninsured or underinsured portions of its loss. Similarly, many of these courts have rejected the argument that the “all sums” language in CGL policies renders each insurer on the risk liable for the entirety of the insured’s loss.22 In other instances courts have interpreted the language of the policies to require pro rata allocation.23

An attempt at a “middle-of-the-road” approach to insurer liability for underinsured or uninsured portions of the loss was undertaken by the court in Owens-Illinois.24 There, the court held that if insurance was available but the insured declined to avail itself of such insurance, the insured should bear the liability for its failure to obtain such coverage. On the other hand, if insurance had not been available for some reason, then the insured should not bear responsibility for the uninsured portions of the loss. Notwithstanding this distinction considered by the court, the logic supporting the decision in Owens-Illinois demands that the insured bear responsibility for underinsured or uninsured portions of its own loss if insurance was available.

As discussed below, there are a number of variations of pro rata allocation.

Policy Limits

One approach to pro rata allocation is to apportion the insured’s losses among the various insurers based upon each insurer’s policy limits.25 This allocation method does not account for the number of years during which each insurer provided coverage. For example, an insurer providing only one year of coverage with a relatively high limit could end up with a much greater liability than an insurer with a lower limit but coverage of the risk for several years.

Per Capita

Under per capita allocation, each insurer (and the insured if allocated an uninsured portion of the loss) is responsible for an equal share of the loss.26 This pro rata by party allocation does not account for the number of years during which each insurer provided coverage. Thus, when the insured suffered six years of continuous damage and Insurer X provided five years of coverage to the insured and Insurer Y provided one year of coverage, each insurer would be liable for a 50 percent share of the insured’s loss.

Time on Loss

Under pro rata-by-time on loss allocation, each insurer is responsible for a percentage of the damage based on time on the risk relative to the entire time over which the damage occurred.27 Many courts consider this method of pro rata allocation to be the most sensible and fair. To the extent that the insured purchased insufficient coverage in any given year, the insured assumes its pro rata share of the loss.

Owens-Illinois Approach

The Owens-Illinois court adopted a hybrid approach to pro rata allocation. The court determined that the fairest method would be to allocate the losses “on the basis of the extent of the risk assumed, i.e., proration on the basis of policy limits multiplied by years of coverage.28 The court recognized, however, that its decision required complex calculations beyond its own capabilities. Accordingly, a special master “skilled in the economics of insurance” was appointed to determine the specific allocation among the parties.29 Moreover, the court recognized that even a “skilled” special master could provide only a “rough measure” of what the proper allocation should be.30 Although there is some surface attractiveness to the Owens-Illinois approach, in light of the problems recognized by the very court that rendered the decision, one must conclude that the solution proposed by the court was not a practical one.

Notes
1 Prolerized Schiabo NEU Co. v. Hartford Accident & Indem. Co., 990 F. Supp. 356, 364 (D.N.J. 1997).
2 Id.
3 7 Couch on Insurance 3d § 102:2 at 102-9-10 (West Group 1997).
4 See generally Doherty, M., Allocating Progressive Injury Liability Among Successive Insurance Policies, 64 U. Chi. L. Rev. 257 (Winter 1997); Kallis, Reiter & Segerdahl, Policyholder’s Guide to the Law of Insurance Coverage § 202 at 2-4–2-18 (Aspen Law & Business 1997); Robinson, R., Coverage Allocation Law: a Primer on the History, Evolution and Current State of Court Mandated Shared Immunity and Defense Obligations, 519 PLI/Lit. 495, 505–09 (1995); Lantz, C., Triggering Coverage of Progressive Property Loss: Preserving the Distinctions Between First- and Third-party Insurance Policies, 35 Wm. & Mary L. Rev. 1801, 1810–12 (Summer 1994).
5 Ins. Co. of N. Am. v. Forty-Eight Insulation, Inc., 633 F.2d 1212, 1222–25 (6th Cir. 1980); Commercial Union Ins. Co. v. Sepco Corp., 765 F.2d 1543, 1546 (11th Cir. 1985).
6 See Hickman & DeYoung, Allocation of Environmental Cleanup Liability Between Successive Insurers, 17 N. Ky. L. Rev. 291, 296 (Winter 1990).
7 See Eagle-Picher Indus. Inc. v. Liberty Mut. Ins. Co., 523 F. Supp. 110 (D. Mass. 1981), modified, 682 F.2d 12 (1st Cir. 1982), cert. denied, 460 U.S. 1028 (1983).
8 See Zurich Ins. Co. v. Raymark Indus., Inc., 516 N.E. 2d 150 (Ill. 1987).
9 See Am. Home Prods. Corp. v. Liberty Mut. Ins. Co., 565 F. Supp. 1485 (S.D.N.Y. 1983), modified, 748 F.2d 760 (2d Cir. 1984).
10 Robinson, R., supra note 4, at 508.
11 See Uniroyal, Inc. v. The Home Ins. Co., 707 F. Supp 1368, 1388 (E.D.N.Y. 1988).
12 Keene Corp. v. Ins. Co. of N. Am., 667 F.2d 1034, 1047–48 (D.C. Cir. 1981).
13 Doherty, M., supra note 4, at 262.
14 See Ingram, J., Insurance Coverage Problems in Latent Disease and Injury Cases, 12 Envtl. L.J. 317, 345 (Winter 1992); Crown Cork & Seal Co., Inc. v. Travelers Cas. & Sur. Co., Nos. L-007456-88, L-007232-93, N. J. Super. Ct., Hudson Co. and accompanying Special Master’s Report (Mealey’s Litigation Report, Insurance, Vol. 12 #28 May 27, 1998).
15 Hickman & DeYoung, supra note 6, at 296–300.
16 Uniroyal, 707 F. Supp. at 1379–80.
17 Id. at 1385. The “stacking” question, which concerns allocation of coverage among various primary and excess carriers on the risk during the same policy period presents a number of variations on these allocation schemes, but will not be analyzed here. For a discussion of the stacking issue, see Gillespie, G., The Allocation of Coverage Responsibility Among Multiple Triggered Commercial General Liability Policies in Environmental Cases: Life after Owens-Illinois, 15 Va. Envtl. L.J. 525, 532–39 (Spring 1996); Hickman & DeYoung, supra note 6, at 301–06; Caton & Brosnahan, Issues of Independent Liability and Stacking in Complex Insurance Coverage Litigation, 369 PLI/Lit. 83, 105–20 (Jan. 1, 1989).
18 See Keene Corp. v. Ins. Co. of N. Am., 667 F.2d 1034, 1047–48 (D.C. Cir. 1981); J. H. France Refractories Co., v. Allstate Ins. Co., 626 A.2d 502, 507 (Pa. 1993).
19 See, e.g., id. at 505.
20 Keene Corp., 667 F.2d at 1049; J. H. France, 626 A.2d at 508.
21 See Owens-Illinois Inc. v. United Ins. Co., 650 A.2d 974, 992 (N.J. 1994). But see Keene Corp., 667 F.2d at 1048.
22 See, e.g., Ins. Co. of N. Am. v. Forty-Eight Insulation, Inc., 633 F.2d 1212 (6th Cir. 1980).
23 See, e.g., Uniroyal, Inc. v. The Home Ins. Co., 707 F. Supp 1368, 1392 (E.D.N.Y. 1988).
24 Owens-Illinois, 650 A.2d at 992.
25 Robinson, R., supra note 4, at 518.
26 Id.; see also St. Paul Fire & Marine Ins. Co. v. Vigilant Ins. Co., 919 F.2d 235, 242 (4th Cir. 1990); Pacific Indem. Co. v. Linn, 966 F.2d 754, 768 (3d Cir. 1985) (pro rata allocation by party of defense costs affirmed as comporting with precedent requiring joint and several liability).
27 Robinson, R., supra note 4, at 519; see Forty Eight Insulations, 633 F.2d at 1225.
28 Owens-Illinois, 650 A.2d at 993.
29 Id. at 994.
30 Id.

Read Allocation of Continuous Damage Losses (Part 2)

ABOUT THE AUTHORS

Stuart Cotton and Philip C. Silverberg are partners in the firm of Mound Cotton Wollan & Greengrass, New York, NY.

riskVue | The webzine for risk management professionals
September 2005