Uncovering Long-Tail Liability Coverage in Historical Insurance Policies

May 15, 2017

Risk Management Magazine

Recent legal decisions should give pause to policyholders across the country that believed it was uneconomical to pursue insurance coverage for long-tail liabilities. If their historical insurance policies include what are known as non-cumulation clauses, policyholders may be leaving millions of dollars of insurance proceeds owed to them in their insurers’ pockets.

Companies frequently face present-day lawsuits alleging injuries or damages arising from occurrences that happened decades ago. Many such alleged injuries or damages were not the subject of lawsuits at the time they first happened because they did not become evident until many years later. Companies have taken solace in their historical occurrence-based liability insurance policies that are triggered by, or respond to, injuries or damages even if they happened long after the policy period ended. Claims seeking insurance from historical, decades-old policies are commonly referred to as long-tail liability claims because of the latency period between the injurious or damaging event and manifestation of the adverse results.

A classic example involves lawsuits asserting bodily injuries from exposure to asbestos contained in a company’s products. There may be a long latency period after initial exposure to asbestos fibers before any adverse effects are felt or an illness is diagnosed. The defendant in such lawsuits may call upon its historical liability insurers for coverage during all policy years beginning with the first injury—frequently inhalation—even though the disease and subsequent lawsuit arises decades later. For example, a worker in a Navy shipyard in the 1950s may have been exposed to asbestos at that time, was first diagnosed with mesothelioma in 2017, and files a lawsuit against the ship’s turbine manufacturer in 2017. That turbine manufacturer may seek insurance coverage from all of its liability insurance policies from the 1950s up until the asbestos exclusion came to be in the mid-1980s.

However, many policyholders facing long-tail liabilities, such as those arising from asbestos, toxic tort, construction defects or environmental contamination allegations, choose not to pursue insurance coverage claims because they believe state law as developed through insurance coverage lawsuits provided for a pro rata allocation. Pro rata allocation takes many forms under various states’ case law, but the fundamental concept involves spreading the policyholder’s losses horizontally over multiple triggered insurance policies. The rationale on which some policyholders relied for not pursuing coverage may have arisen from their historical coverage programs that included periods of self-insurance, under-insurance, lost policies and insurer insolvencies. Under some versions of the pro rata allocation approach, the policyholder may arguably bear the consequences of these gaps and, as a result, pursuing coverage from recalcitrant insurers was an economically unsound endeavor. Recent case law, including, by way of example, cases in New York, highlights that the economics now favor pursuing coverage even in states that may be considered by most to be “pro rata states.” Failing to take a second look at their policy language may cost policyholders millions of dollars.

If a policyholder provided timely notice to its historical liability insurers, it is time to take another look at the insurance policy language and, potentially, recover significant insurance proceeds irrespective of purported gaps in coverage. Recent case law confirms that courts will, as they should, enforce the terms and conditions of the insurance policies as they were written. As the New York cases underscore, if a policyholder’s general liability insurance policies include what are commonly referred to as non-cumulation clauses, they may allocate their long-tail liabilities on an “all sums” basis. That means all triggered insurance policies are joint and severally liable and the policyholder may select the policies that will respond to its losses. In other words, if a policyholder’s insurance policies include this non-cumulation language, it can recover its liabilities from certain policy years with significant, solvent insurance.


The issue of allocation refers generally to how a large loss will be shared by multiple insurance policies that are triggered. The loss may be spread horizontally over multiple triggered policies or may be assigned to a single triggered policy year. Generally, most judicial decisions apply one of two allocation methodologies for sharing large losses: 1) pro rata; or 2) “all sums.” These two different approaches are described below.

Insurance companies generally argue for “pro rata allocation” or “pro rata by time on the risk allocation,” which may take multiple forms, but generally refers to dividing a loss “horizontally” among all triggered policy periods, with each insurance company paying only a share of the policyholder’s total damages. When courts adopt proration, they tend to rely upon general principles of equity, rather than policy language, ruling that, given the facts in a particular case, it is fair to spread the loss over several years rather than require one insurer or one policy year with liability to pay for “all sums.”

Policyholders, by contrast, frequently argue that, once a policy year is “triggered” by injury or property damage during the policy period, each of the individual insurance policies in that year must indemnify the policyholder for “all sums” for which the policyholder becomes liable, subject to each policy’s limits, regardless of when the bodily injury or property damage occurred. “All sums” allocation divides the loss among policies “vertically.” Each triggered policy is liable for “all sums” until the policy’s limits are exhausted, and then the policies that sit above the exhausted policy are called upon in the same manner. After the policyholder is paid for its loss, the paying insurance companies then may be able to pursue contribution claims against the other non-paying insurance companies whose policies are triggered in different policy years.

The rationale for allocating all sums is grounded in the policy language. Once a policy is on the risk, the unambiguous policy language requires the insurance company to pay “all sums” for which the policyholder shall become legally obligated to pay. That language defines the insurer’s duty under its policies as the obligation to pay “all sums” for which the policyholder becomes liable; not a pro-rata share.

The Old Thinking

Case law in New York provides a recent example of how policyholders’ and insurers’ thinking has changed. The New York courts’ logic is not unique to the state and is likely to be adopted in other courts across the country when faced with similar claims and policy language. Therefore, recent case law developments in New York highlight the need for policyholders across the country to reexamine their historical insurance policies.

Some people believed that the law of New York on allocation prior to 2016 was settled. New York’s high court had adopted a pro rata approach. In the 2002 case, Consolidated Edison Company of New York v. Allstate Insurance Company, the Court of Appeals applied a pro rata allocation. The court held that pro rata allocation is consistent with the language of the policies that were at issue in that case. The court explained its conclusion as being premised on the fact that the insurance policies agreed to indemnify the policyholder for liability that arose from an event happening only during the policy period and not outside of the policy period. Notably, the court did not reach its holding in Consolidated Edison by adopting a blanket rule, based on public policy concerns, that pro rata allocation is always the appropriate method of dividing losses among multiple insurance policies that are triggered. Instead, it relied on principles of contract interpretation, and made clear that the insurance policy language, and not some notion of equity, controls the question of allocation. Further, the court recognized that if the insurance policies contained different policy language than the language that was at issue in Consolidated Edison, any such different language might compel “all sums” allocation.

New York courts following Consolidated Edison applied pro rata allocation in many cases addressing allocation of liabilities over multiple insurance policies that are triggered. As a result, policyholders with large periods of self-insurance or high deductibles may have put their insurers on notice of potential liabilities and then decided not to pursue coverage because spreading their losses horizontally across the triggered policy years may have resulted in a minimal insurance recovery. In other words, under some variations of pro rata allocation, the policyholder may be stuck paying out of its own pocket for the losses spread to policy period that do not provide dollar-one insurance coverage from a solvent insurer.

The New Approach

Despite what some policyholders and insurers alike may have thought about the allocation methodology that a given court was likely to adopt, some policyholders challenged the conventional thinking and reaped huge rewards. Policyholders focused on a provision included in many historical insurance policies and successfully argued that these provisions are squarely at odds with pro rata allocation. Courts agreed and policyholders have secured significant amounts of coverage for long tail claims under all sums allocation in jurisdictions where insurers traditionally relied on pro rata allocations.

New York case law from 2016 provides an illustrative example. New York courts held that all sums allocation must be applied to losses if the insurance policies at issue include either one of two versions of a non-cumulation provision. Generally speaking, non-cumulation provisions acknowledge that the insurance policy responds to an occurrence giving rise to injury or damage that occurs partly before and partly within the policy period. By its plain language, the non-cumulation provision (sometimes called the “prior loss” provision, or the London Condition C provision) cannot be squared with the very essence of pro rata allocation—that no two insurance policies may indemnify the same loss or occurrence.

In 2016, New York’s high court issued its decision in In re Viking Pump, Inc., which adopted the all sums allocation approach for all insurance policies containing non-cumulation provisions. The Viking Pump court concluded that the policy language at issue there, by inclusion of the non-cumulation clauses, was substantively distinguishable from the language at issue in Consolidated Edison. The court held that non-cumulation provisions were fundamentally inconsistent with pro rata allocation and held further that policyholders are entitled to allocate all sums to any triggered year of coverage with policies that included non-cumulation provisions. This means that if the policyholder has insurance policies including these provisions, it can seek payment from the insurers in certain years that do not include any deductibles, missing policies or insolvent insurers.

Policyholders immediately took advantage of the non-cumulation policy language and uncovered significant insurance proceeds to pay their losses. In Liberty Mutual Insurance Co. v. The Fairbanks Co., the federal district court for the Southern District of New York had initially held that all insurers were entitled to a pro rata allocation. The court reconsidered its decision with regard to Liberty Mutual after the decision in Viking Pump because Liberty Mutual sold Fairbanks historical policies with non-cumulation clauses. Fairbanks now has the right to seek all sums from Liberty Mutual for the asbestos bodily injury claims that trigger Liberty Mutual’s umbrella coverage from 1974 to 1982, which amounts to tens of millions of dollars of coverage.

Additionally, insurers have acknowledged that non-cumulation language may be located in the insurance policies’ “other insurance” provision. If located in this provision, the language has the same importance to allocation as it does if located in a non-cumulation or prior insurance provision. When stated in an “other insurance” provision, non-cumulation language recognizes that a loss may be covered by more than one policy in a line of successive policies sold by the same insurer. This language, whether located in a non-cumulation provision or in an “other insurance” provision, is incompatible with pro rata allocation and calls for all sums.

Policyholders across the country, like Fairbanks, who have historical liability policies with non-cumulation language should seek to allocate all sums vertically in years providing solvent insurance, without large deductibles and without prior settlements or missing policies. By doing so, policyholders may avoid unfavorable years of coverage and collect significantly more insurance proceeds that they would pursuant to a pro rata allocation.

Do not leave insurance proceeds on the table based on the old thinking that was (incorrectly) prevalent in New York jurisprudence for many years. All policyholders with dormant long-tail insurance coverage claims should dust off their general liability policies and look for non-cumulation language that entitles them to “all sums” under any triggered years of coverage. All sums allocation may allow policyholders to avoid periods of self-insurance, periods in which it purchased coverage excess to large deductibles or self-insured retentions, periods that include insurance sold by insurers that are now insolvent, and periods in which the purchased policies are missing.

Uncovering Long-Tail Liability Coverage in Historical Insurance Policies,” by Jared Zola was published in Risk Management Magazine on May 15, 2017. To view the article online, please click here. Reprinted with permission from Risk Management Magazine. Copyright 2017 Risk and Insurance Management Society, Inc. All rights reserved.