Aggregating Losses From the COVID-19 Crisis for Reinsurance Purposes

Onions at the farmers market

The question of whether hundreds or even thousands of COVID-19-related losses can be aggregated together as one “loss” or “occurrence” for reinsurance purposes is one that both ceding companies and reinsurers are pondering.  Expressly putting aside whether COVID-19-related business interruption losses are covered by underlying insurance policies (see our other blog posts), this post discusses how both cedents and reinsurers might prepare for the possibility.

This question is relevant because many individual COVID-19-related claims may be too small on their own to reach the attachment point of most property excess-of-loss or catastrophe reinsurance treaties.  Insurance companies purchase catastrophe reinsurance protection for accumulated losses from hurricanes, wildfires and other natural and man-made disasters, which allows them to aggregate a pool of smaller individual losses into one single loss or occurrence for reinsurance purposes that would not otherwise be reinsured.

Most property excess-of-loss and catastrophe reinsurance contracts contain provisions that allow for aggregation.  For example:

The term “Loss Occurrence” shall mean the sum of all individual losses directly occasioned by any one disaster, accident or loss or series of disasters, accidents or losses arising out of one event which occurs anywhere in the world.

Additionally, most property excess-of-loss and catastrophe reinsurance contracts include an hours clause within the definition of “Loss Occurrence,” which limits the loss to a specific, but limited, time period.  For example, a common hours clause limits the loss to 168 consecutive hours.

As seen above, many aggregation clauses rely on event-based language, which requires that the individual losses to relate to each other temporally and spatially.  For COVID-19-related losses, which spanned the country and took place over many months, the key question is what caused the underlying losses:  was it a civil authority order; was it multiple civil authority orders; was it the novel coronavirus itself; or was it a series of losses separated by time and space that cannot be combined for reinsurance purposes.

Determining what individual losses can be aggregated and to how many events are complex issues.  It depends on the reinsurance contract wording and, especially, the definitions used in the aggregation provision.  A critical review of aggregation language will prepare both cedents and reinsurers for potential cessions of COVID-19 losses as one event.

Aggregation clauses in reinsurance contracts are similar to provisions in insurance policies that allow for aggregation within the definition of “occurrence” or “accident.”  In other words, there are quite a few cases addressing whether a collection of individual losses are deemed “one occurrence” or “one accident” or “one event” in the direct insurance context.

In the direct insurance context, one Texas federal court identified three approaches for deciding whether losses arise from a single occurrence:  the causation view; the effect view; or the liability triggering view, with the majority of jurisdictions applying the causation view.  Sec. Ins. Co. v. Lubrizol Corp., No. 1:06-CV-215, 2009 U.S. Dist. LEXIS 147818, at *13 (E.D. Tex. Aug. 27, 2009).  Your jurisdiction will determine which test to apply, but because the majority of states apply the causation view, we will focus on that approach.

Under the causation view, the parties look at the cause of the damage or injury.  Id. (finding that damage caused by contaminated dumping at one site resulted in only one occurrence); see Big Lots Stores, Inc. v. Am. Guar. & Liab. Ins. Co., 240 F. Supp. 3d 725, 732 (S.D. Ohio 2017) (stating that Ohio like the majority of jurisdictions applies the proximate cause test).  Courts consider whether there is one proximate, uninterrupted, and continuing cause that resulted in all of the injuries and/or damages. Id. (citing Mich. Chem. Corp. v. Am. Home Assurance Co., 728 F.2d 374, 379 n.5 (6th Cir. 1984)).

If there is only one proximate cause, then there is only one occurrence.  Id. (concluding that there were multiple incidents, explaining that each time Big Lots sold a defective torch there was a “new exposure and another occurrence”); see Golden Eagle Ins. Corp. v. Moon Marine (U.S.A.) Corp., No. 3:12-cv-05438-WHA, 2013 U.S. Dist. LEXIS 197531, at *6 (N.D. Cal. Nov. 15, 2013) (after twenty-five people developed salmonella infections, the court said that “whether . . . the insured events arise from a single occurrence will depend upon how the bacteria got into the fish in the first place” and thus whether there was a single proximate cause); see also Appalachian Ins. Co. v. Gen. Elec. Co.  863 N.E.2d 994 (N.Y. 2017) (describing the proximate cause test as the “unfortunate event test,” which asks “whether there is a close temporal and spatial relationship between the incidents giving rise to injury or loss, and whether the incidents can be viewed as part of the same causal continuum, without intervening agents or factors.”).

Courts, however, have reached seemingly inconsistent positions on whether losses are attributable to a single cause when the injury or damage occurs at geographically separate locations.  This can be seen in the context of environmental contamination claims.  In Argonaut Insurance Co. v. Travelers Insurance Co., 800 N.Y.S.2d 342 (N.Y. Sup. Ct. 2005), the court determined that pollution claims at 140 separate sites could not be aggregated as one occurrence.  “The cause of the damages, in the case of any particular claim, was not the failure by [the insured’s] management to implement environmentally-sound, company-wide procedures but, rather, the separate environmental pollution which occurred at the site involved in the claim.”  See Cadet Mfg. Co. v. Am. Ins. Co., 391 F. Supp. 2d 884 (D. Wash. 2005) (concluding that the release of TCE at geographically distinct sites are separate occurrences).   In Sunoco, Inc. v. Illinois National Insurance Co., 226 F. App’x 104, 105 (3d Cir. 2007), however, the court determined that in 76 of 77 lawsuits, there was “one proximate, uninterrupted and continuing cause which resulted in all of the injuries and damage” arising from a single occurrence because they arose from MtBE contaminating ground water, despite the damage occurring in different regions, from leaks in different pipelines.

Like all insurance policies, the language of each reinsurance contract needs to be scrutinized because aggregation provisions vary greatly.  For example, a reinsurer may agree to pay the cedent “for each and every loss” in excess of the reinsurance contract’s retention limit or aggregate deductible, with “each and every loss” defined as “all losses arising out of any one disaster, casualty, or occurrence.”  Other aggregation provisions use the term “event” in defining “each and every loss.”  For example, “[t]he term ‘each and every loss’ as applied to ‘any one risk’ shall be understood to mean each and every loss arising out of one and the same event.”  It seems evident that a hurricane might be is “one disaster.” See SEACOR Holdings Inc. v. Commonwealth Ins. Co., No. 06-4685, 2008 U.S. Dist. LEXIS 129346, at *1 n.1 (E.D. La. Dec. 11, 2008) (“Both parties agree that an event such as Hurricane Katrina is considered a ‘single occurrence.’”).

Travelers Casualty & Surety Co. v. Certain Underwriters at Lloyd’s of London, 96 N.Y.2d 583 (2001), is the leading case addressing aggregation under a reinsurance agreement.  Travelers involved two cedents contending that pollution-related damages at several sites over several decades could be aggregated as a single casualty or event under their respective reinsurance agreements.  Each reinsurance agreement had the same language, and required the reinsurer to pay the cedent for “each and every loss,” which was defined as, “all loss arising out of any one disaster and/or casualty under coverage of any or all insureds of the Companies . . . .”  Disaster and/or casualty, was defined as,

each and every accident, occurrence and/or causative incident, it being further understood that all loss resulting from a series of accidents, occurrences, and/or causative incidents having a common origin and/or being traceable to the same act, omission, error and/or mistake shall be considered as having resulted from a single accident, occurrence and/or causative incident.

Both cedents settled the pollution-related claims, and then sought indemnification from their reinsurers as a single loss, arguing that the claims had a common origin or were traceable to the same act, omission, error or mistake.  The New York Court of Appeals held, as a matter of law, that the cedents’ single allocation of their settlements did “not fall within the ambit of ‘disaster and/or casualty’ in the reinsurance treaties.”  In other words, the losses could not be aggregated as a single occurrence under the reinsurance agreements.  In making this assessment, the court found that the parties did not intend for the cedent to simply group together all other losses into a single disaster and/or casualty, but sought to allow aggregation only where the losses were linked spatially or temporally and shared a common origin.  Neither complaint, held the court, contained any allegation that the sites bore a spatial or temporal relationship to each other.

Interestingly, the Travelers Court addressed the famous House of Lords decision in Axa Reinsurance (UK) Plc v. Field, [1996] 2 Lloyd’s Rep 233, which explained the difference between arising out of “one event” and arising out of “one originating cause.”  In that case, Lord Mustill found that the words “originating cause” opened up the widest possible search for a unifying factor, while the word “event” was something that happens at a particular place, at a particular time, in a particular way.  In other words, “event” language is narrower than “original cause” language.  In Travelers, the court found that the words “a series of” required a spatial or temporal relationship between losses with a common origin.  This only emphasizes the point that each reinsurance contract has to be construed based on its specific language.

The bottom line is that if cedents pay COVID-19 related losses, reinsurers must scrutinize the language of their treaties and carefully evaluate the basis for aggregating multiple losses as one loss occurrence.

Leave a Reply

Your email address will not be published. Required fields are marked *