Owens-Illinois/Carter-Wallace Revisited: N.J. Court Addresses Allocation Yet Again

Determining precisely when property damage or personal injury occurred in a case involving exposure to chemicals or other toxic substances can be extremely difficult, if not impossible.  In most cases, the acts or practices giving rise to potential liability occurred decades before the injuries manifested themselves.  It is often the case that a commercial insured had many different insurance policies in place during the relevant time periods.  Indeed, large commercial insureds typically purchase several layers of insurance coverage from multiple insurers   Thus, when an insured is sued for such injuries, it is necessary to determine which policies may potentially provide coverage.

In Owens-Illinois, Inc. v. United Ins. Co., 138 N.J. 437, 479, 650 A.2d 974 (1994), the New Jersey Supreme Court adopted the continuous trigger for determining which policies of insurance provide coverage.  Trigger is “a shorthand expression for identifying the events that must occur during the policy period to require coverage for losses sustained by the policyholder.”  Id. at 441.    Under the continuous trigger theory, all policies that were in effect from initial exposure to the injury-producing substance through manifestation of injury provide coverage.   The court summarized its holding as follows:

To recapitulate, we hold that when progressive indivisible injury or damage results from exposure to injurious conditions for which civil liability may be imposed, courts may reasonably treat the progressive injury or damage as an occurrence within each of the years of a [commercial general liability] policy.  That is the continuous-trigger theory for activating the insurers’ obligation to respond under the policies.

Id. at 478.

While Owens-Illinois involved claims for bodily injury and property damage resulting from exposure to asbestos, a continuous trigger has been applied to other claims involving progressive, indivisible injury resulting from exposure to toxic substances, including both property damage and personal injury claims.

The Owens-Illinois court also addressed the issue of how to allocate the losses among the various policies that had been triggered.  The court decided to adopt an allocation method based on “both time on the risk and the degree of risk assumed.”  Id. at 479.  The court further held that an insured is required to assume a portion of the risk during those periods when coverage was available but the insured chose not to purchase it.  The allocation method adopted by the court is referred to as a pro-ration by years and limits method of allocation.

One issue that the court did not address was how to allocate the loss among the various primary and excess policies that were triggered.  That issue was addressed several years later in Carter-Wallace, Inc. v. Admiral Ins. Co., 154 N.J. 312, 712 A.2d 1116 (1998).  The Carter-Wallace court adopted an allocation method based on the amount of insurance purchased or risk retained in any given year.

In simple terms, under the method adopted by the court, the amount of coverage available in a given year is divided by the total amount of coverage available for the entire period at issue.  That number (which represents a percentage of the total coverage available) is then multiplied by the total damages to arrive at a loss figure allocated to that particular year.  The same calculations are repeated for each year in which coverage was available.

Under the method adopted by the court, the proportion of the loss allocated to any given year will depend on the amount of coverage purchased in that year in relation to other years.  In other words, a larger portion of the loss will be allocated to the years in which the insured purchased or could have purchased the most coverage.

Once a particular amount is allocated to a given year, the policies that were in place during that year are exhausted from the ground up.  The allocation method adopted by the court requires the depletion of any underlying insurance in a given policy year before an excess carrier is required to pay.

Despite the fact that Owens-Illinois was decided twenty years ago, and Carter-Wallace was decided sixteen years ago, disputes still arise as to how a loss should be allocated.  In 2013 alone, the New Jersey Supreme Court issued two decisions just eight days apart concerning Carter-Wallace/Owens-Illinois allocation issues.  See Farmers Mutual Fire Ins. Co. v. New Jersey Property-Liability Ins. Guaranty Assoc., 215 N.J. 522, 74 A.3d 860 (2013); Potomac Ins. Co. v. Pennsylvania Manufacturers’ Assoc. Ins. Co., 215 N.J. 409, 73 A.3d 465 (2013).

The problem is that in deciding Owen-Illinois and Carter-Wallace, the New Jersey Supreme Court was trying to fashion a method for dealing with complex losses that were not contemplated by the parties to the insurance policies at the time they were issued.  The method selected by the court was less than perfect.  As new issues arise, the New Jersey courts try to mold the facts and policy language at issue to fit the Carter-Wallace/Owens-Illinois allocation methodology.  When the policy language does not fit in with the method adopted by the court, the policy language  simply is not enforced.

In support of the argument that policy language may be ignored, New Jersey courts often cite to Benjamin Moore & Co. v. Aetna Cas. & Sur. Co., 179 N.J. 87, 105, 843 A.2d 1094 (2004), in which the New Jersey Supreme Court stated that “the basic insurance policy provisions apply so long as they are not inconsistent with Owens-Illinois.”   Courts have taken that to mean that if the terms are inconsistent with Owens-Illinois they should not be enforced.  Interestingly, the Benjamin Moore court actually enforced the policies as written, holding that the amount payable in any given year was subject to the full per-occurrence policy deductible.

Recently, the Carter-Wallace allocation method was revisited once again by the New Jersey Appellate Division in IMO Indus. Inc. v. Transamerica Corp., 2014 WL 4810047 (App. Div. Sept. 30, 2014).  IMO involved claims for insurance coverage arising out of asbestos-related personal injury claims.  At issue was a total of $1.85 billion in insurance coverage purchased by the insured, IMO Industries, Inc. (“IMO”), over several decades.  The policies, which were issued by a number of different insurers, were in effect during the period from 1935 through 1986.  IMO was seeking coverage under those policies for approximately 75,000 asbestos-related claims asserted against it.

The IMO case had a fairly complex history.  IMO originally sued some of its insurers in August 2003.  In 2004, IMO amended the complaint to add additional claims against some of the insurers and to add additional insurers.  The case was reassigned to several different judges over the course of the litigation and both a special allocation master and a special discovery master were retained by the court to assist with the case.  In June 2009, the first of three mini-trials was held.  A second and third mini-trial were held in the spring of 2010.  In January 2011, a new trial judge was assigned to the case.  He issued a series of rulings, culminating in the issuance of a final judgment, and an appeal followed.

The Appellate Division addressed a number of issues pertaining to the allocation of IMO’s asbestos-related losses.  The more significant questions are addressed below.

Allocation of Defense Costs

One of the issues addressed by the IMO court was the allocation of losses to policies that provided defense costs “outside the limits.”  Under such a policy, an insurer is required to pay an insured for all costs it incurs in defending certain actions in addition to the full policy limits.  Such a policy provides unlimited coverage for defense costs until coverage is exhausted through the payment of judgments or settlements.  This is in contrast to an “inside the limits” policy where the payment of defense costs erodes or reduces the policy limits.  Under an “inside the limits” policy, an insured cannot recover more than the policy limit for defense costs and indemnity payments combined.

TIG Insurance Company (“TIG”) issued a series of “outside the limits” policies to IMO that were in effect during the period from 1977 through 1981.  The coverage limit for each policy was $1 million.  However, TIG was required to reimburse IMO for defense costs in addition to the limits of those policies.  TIG argued that its policies were exhausted because it had already paid IMO an amount in excess of the policy limits.

IMO, on the other hand, argued that the policies were not exhausted because only the payment of indemnification costs would exhaust the limits of the policies.  IMO claimed that TIG still owed $48 million in defense costs under the policies.  IMO’s arguments were summarized as follows:

The crux of [IMO’s] theory is that, with defense costs being paid outside the policy limits, TIG’s obligation to cover defense costs for claims that could be attributed to those policy years would continue until TIG actually paid $1 million in indemnity costs from the policy in each of those years.  With the small amount of indemnity payments on the liability that IMO has for injured plaintiffs in asbestos cases (many of those cases settling for only several thousand dollars from IMO), the limits of the TIG “outside the limits” policies would not be reached for many years. TIG’s obligation to continue paying for defense costs would continue indefinitely.

Id. at *6.

As is typical of most liability policies, the TIG policies provided that TIG would “not be obligated to pay any claim or judgment or to defend any suit after the applicable limit of the company’s liability has been exhausted by payment of judgments or settlements.”  Id. at *9 (emphasis in original).  IMO argued that formal payment of judgments or settlements was required to exhaust coverage.  In other words, according to IMO, the allocation of losses to the policies under the Owens-Illinois/Carter-Wallace methodology, although it required payment by TIG, was not sufficient to exhaust coverage and, therefore, TIG’s obligation to pay defense costs continued indefinitely.

The court began its analysis by noting that while the New Jersey Supreme Court never directly addressed the issue, “the undeniable implication of Owens-Illinois is that defense costs are also allocable, subject to policy terms, in the same manner as indemnity payments.”  The court also concluded that the allocation of costs to a policy under Owens-Illinois/Carter-Wallace can exhaust coverage.  Thus, the court sided with TIG.   However, the court’s holding did not mean that the insured would recover less.  Rather, the payment obligation would be shifted to the excess insurers.

Essentially acknowledging that its holding was not consistent with the plain language of the policy, the court noted that “the Owens-Illinois and Carter-Wallace allocation model supersedes contrary terms of an insurance policy.”  Id. at *11-13.  The court went on to note, however:

To allay some of the fears expressed by amicus curiae, the exhaustion decision in this case is closely tied to its facts. We reach no general conclusion that an insurer’s obligations to cover defense costs and other litigation expenses through an “outside the limits” policy is limited by the maximum amount of indemnification coverage provided in that policy.

Id. at *13.  The court stressed that the case before it dealt with application of “the supervening effect of the Owens–Illinois and Carter–Wallace allocation methodology,” pursuant to which “inconsistent” policy terms “drop out” and/or are “displaced.”  Id.

Coverage Available Under Multi-Year Policies

Another issue addressed by the court was the coverage available under multi-year policies.  Specifically, several policies were in effect for a period of more than one year.  Those policies provided a per occurrence limit on coverage.  In other words, the policies provided that the insured was not entitled to recover in excess of the per occurrence policy limit during the entire period that the policies were in effect.  Thus, although the policies were in effect for more than one year, the insured was able to recover the per occurrence limit only once.  Under a single-year policy, in contrast, the insured would be entitled to recover a separate per occurrence policy limit each year.

The court summarized the issue before it as follows:

IMO does not dispute that the plain language of the policies would impose per-occurrence limits on a term rather than annual basis, but it sought a blanket ruling that every year of a multi-year policy should be treated as if a separate annual limit is available for asbestos claims.

Id. at *17.  The court agreed with IMO on this issue, noting:

Were it not for the pro-rata methodology adopted in Owens–Illinois, each asbestos claim filed against IMO that triggered the [multi-year] policies would be treated as a separate occurrence subject to the per-occurrence limit for the entire multi-year terms of the policies.  The aggregate limits of the policies would control the insurers’ total liability on the claims.  Owens–Illinois changed the ground rules and classified all asbestos claims made in a year as a single occurrence.  If all three years were to be viewed as a single occurrence, the insured would be deprived of the annual aggregate limits of the policies.

Thus, the court concluded that “[b]ecause the imposition of per-occurrence limits in multi-year policies contravenes the goals of the pro-rata methodology established in Owens–Illinois, such limits are unenforceable as specifically written.”  Id. at *18.  In other words, the court disregarded the plain language of the policy because it did not “fit” with the Owens-Illinois/Carter-Wallace allocation theory.  By doing so, the court greatly increased the liability of those insurer who issued multi-year policies.  Rather than ignore the clear and unambiguous policy language, the court could have simply divided the total per occurrence limit of each multi-year policy by the number of years the policy was in effect to arrive at a “per year” occurrence limit.  Doing so would give meaning to the per occurrence limits in the policy while also being consistent with Owens-Illinois/Carter-Wallace because a separate, albeit reduced, limit would apply in each policy year.

Coverage Under a “Stub” Policy

One of the policies at issue was in effect for only eleven months.  A policy that is in effect for less than a full year is referred to as a “stub” policy.  An insurer will sometimes issue a “stub” policy in a situation where an insured is attempting to make the terms of all its policies coincide or where a short term extension is needed for some other reason.  Because the policy was in effect for only eleven months, the insurer argued that IMO was entitled to recover only eleven-twelfths of the full policy limit.

The court rejected this argument.  In a prior decision involving a similar issue, United States Mineral Products Co. v. American Ins. Co., 348 N. J. Super. 526, 792 A.2d 500 (App. Div. 2002), the court held that a “stub policy created an additional set of aggregate limits that were available to the insured for the term of the policy.”  IMO, 2014 WL 4810047 at *21 (citing Mineral Products, 348 N.J. Super. at 550).  The IMO court noted that “[t]reating a stub policy as providing a pro-rated limit would result in the loss allocated to the policy being reduced twice, once by its time on the risk and a second time by the pro-rating of the policy limit.”  2014 WL 4810047 at *21.  The court went on to note:

Our holding in United States Mineral Products is clear. If the annual aggregate limits of a stub policy are to be pro-rated, specific language in the policy must so provide.  Nothing in the [stub] policy indicates that the annual aggregate limits are pro-rated.

Id. Thus, the court ruled in favor of IMO on this question.  In resolving the issue, the court did not disregard the policy language.  As noted by the court, nothing in the policy indicated that the limits were to be pro-rated.  In addition, had a loss occurred prior to expiration of the policy term, there is no question that the insurer would have been liable up to the full policy limit.

Application of “SIRs”

Some of the policies at issue contained a “self-insured retention” or “SIR,” which is an amount that must be “retained” by the insured before coverage for a loss will be triggered.  If the loss is less than the SIR, the policies will not come into play.  Thus, the insured acts as if it is self- insured with respect to its retention.  A SIR is different from a deductible in that a deductible is an amount that will subtracted from the total amount of a covered loss.  Consequently, “the insurer has the liability and defense risk from the beginning and then deducts the deductible amount from the insured coverage.”  In In re September 11th Liability Ins. Coverage Cases, 333 F. Supp. 2d 111, 124 n.7 (S.D.N.Y. 2003).  Significantly, “[w]hile the limits of a policy are reduced by a deductible, they remain intact in the case of a self-insured retention.”  IMO, 2014 WL 4810047 at *21.

The distinction between a SIR and a deductible had not been directly addressed by a New Jersey court prior to IMOId. at 22.  As noted by the court:

New Jersey courts have recognized that an insured’s deductible erodes the policy limits.  On the other hand, federal courts have stated clearly that a SIR does not reduce the limits of an insurance policy.

Id. (citations omitted).  Without explicitly saying so, IMO adopted that distinction between a SIR and a deductible.  The court then found that the language in the policies at issue supported the conclusion that they were subject to a SIR and not a deductible.

The difference between a SIR and a deductible is fairly significant under the Owens-Illinois/Carter-Wallace allocation methodology.  Specifically, if a primary policy contains a $1 million limit, and $1 million is allocated to that policy under Owens-Illinois/Carter-Wallace, the amount recoverable by the insured will be reduced by any deductible.  If, for example, the deductible is $250,000, the insurer will  have to pay the insured only $750,000.  If, however, the policy contains a SIR, and not a deductible, the insured will be entitled to recover the full $1 million policy limit.

One issue that was not discussed by the court was whether the SIRs will be allocated to the insured.  Because a SIR must be satisfied before coverage comes into play, a court should treat the SIR the same as primary insurance for purposes of the Owens-Illinois/Carter-Wallace allocation, thereby reducing the total liability of the insurers accordingly.

The Excess Insurers’ Ability to Contest Coverage

Another issue addressed by the court was whether the excess insurer could contest the underlying coverage determinations made by IMO.   The court began its analysis of this issue by noting that the insured ordinarily bears the burden of establishing that a claim is covered under its insurance policy.  Id. at *23.  The court then observed:

A primary insurer that refuses its obligation to defend claims against its insured without first timely challenging coverage forfeits the right to hold an insured to that burden at a later time.  Excess insurers, on the other hand, generally have no duty to participate in the defense and may rely on the good faith of the primary insurer in settling claims against the insured.

Id.  Nonetheless, the court concluded that “[a]llowing excess insurers to contest coverage is not feasible for long-tail, multi-claim coverage cases and would compromise the allocation methodolgy mandated by the Supreme Court.”  Id. at *23.  The court once again noted that “policy terms and traditional principles applicable to ordinary coverage litigation must bend insofar as they conflict with application of the Owens–Illinois framework.”  Id. at *24.

The court’s holding was limited, however, to only those excess insurers “who have declined to associate in the defense of claims . . . .”  Id. at 25.  Thus, it is incumbent on insureds to notify their excess insurers of the claims presented against the insured and allow the excess insurers to participate in the defense of those claims.

Coverage for Uncovered Claims

Also before the court was the issue of whether IMO was entitled to reimbursement for defense costs incurred in connection with claims that were not covered under the policies.  Certain of the policies provided that coverage for defense costs was limited to those defense costs paid in connection with a covered occurrence.  Those insurers argued that their policies made “the existence of an actually covered claim, and not just a potentially covered claim, a prerequisite for indemnification of defense costs.”  Id. at *25.  IMO, in contrast, argued that the policies required that the insurers “pay for the costs of defending liabilities arising from covered ‘occurrences,’ not covered claims.”  Id.  According to IMO, “Owens-Illinois defines an ‘occurrence’ to be the decision to manufacture asbestos-containing products, not the exposure of a specific claimant to an asbestos product.”  Id.

The court agreed with IMO’s argument, noting that under Owens-Illinois, “the manufacture and sale of the asbestos-containing product should be regarded as the single occurrence triggering liability for asbestos-related injuries or damage.”  Id. at *26.  According to the court, “IMO’s decision to trade in such products resulted in IMO paying damages to claimants following litigation or settlement.”  Id.  However, those claims were not at issue.  What was at issue were costs incurred to defend uncovered claims.  The court did not seem troubled by that significant distinction.  The court held that “[t]he conclusion that the excess insurers must reimburse IMO for defense costs even if some of them were incurred to defend uncovered claims is also compelled by another aspect of Owens–Illinois.”  Id. at 27.   According to the court, “the need to segregate and classify defense costs according to each individual claim would greatly complicate the already complex allocation process.”  Id.  The court seemed more concerned with the fact that segregating uncovered claims “would increase litigation and require additional judicial attention” than it was with enforcing the policies as written.

*          *          *

The IMO decision is just another example of a New Jersey court’s struggling to make a judicially created allocation scheme fit the particular facts and policy language before it.  Unfortunately, the method adopted by the New Jersey Supreme Court in Owens-Illinois and Carter-Wallace was far from perfect and has left courts grappling with allocation issues for well over a decade.  The IMO case shows how difficult allocation of losses in long-tail toxic tort cases can be:  Several different judges were assigned to the case, both a special allocation master and a special discovery master were retained, two trials were held, and the case is still pending eleven years after it was originally commenced.  It will not be surprising if the case eventually winds up before the New Jersey Supreme Court.

However, the complexity added to the resolution of cases as a result of the adoption of the Owens-Illinois/Carter-Wallace allocation method is not the most unsettling aspect of the court’s rulings. What is even more unsettling is the fact that courts will simply disregard clear policy language when it is “inconsistent” with the Owens-Illinois and Carter-Wallace allocation method, in many instances creating much different insurance contracts from the ones the parties entered into decades ago.  Equally troubling is the fact that the IMO court thought it appropriate to ignore clear and unambiguous policy language because enforcing certain policies as written “would increase litigation and require additional judicial attention.”     


© William D. Wilson and NJInsuranceBlog.com, 2014.  Unauthorized use and/or duplication of this material without express and written permission from this blog’s author and/or owner is strictly prohibited.  Excerpts and links may be used, provided that full and clear credit is given to William D. Wilson or NJInsuranceBlog.com with appropriate and specific direction to the original content.


By William D Wilson

I am a partner in Mound, Cotton, Wollan & Greengrass, which is headquartered in New York City. I am in charge of running the firm's New Jersey office, which is in Florham Park. I have been practicing law for approximately 23 years and focus primarily on insurance related matters