Much has been written about the judicial inquiry into whether separate incidents may be aggregated as a single accident or occurrence under various formulations of accident or occurrence language in insurance policies. This issue has been litigated heavily in the asbestos context and in other latent and continuous injury settings. But sometimes looking at the issue in the context of a more straightforward situation helps explicate the rule.

In Nat’l Liab. & Fire Ins. Co. v. Itzkowitz, No. 14-3651-cv, the Second Circuit Court of Appeals, in a Summary Order without precedential effect, had the opportunity to discuss New York’s “unfortunate event test” for determining whether a series of incidents may be aggregated under the language of the operative insurance policy.

This case involved events that many of us have seen all too often. A truck, this time carrying a dump box, was driving on a highway with overpasses. Most overpasses are shorter on the outside lanes and have more height on the inside lanes. Here, the driver hit the overpass with the box and the box careened off the truck and landed on the highway. In a terrible series of events that followed, two separate carloads of families returning from upstate New York hit the box with their cars causing catastrophic results.

The insurer brought an action to declare that the three incidents were part and parcel of one accident (which would limit its liability to one policy limit). The insureds wanted a declaration that these were three separate accidents, thereby tripling the limits available (one policy limit per accident). The district court granted summary judgment to the insureds and the Second Circuit affirmed.

In affirming, the circuit court held that New York’s unfortunate event test applied to determine how the incidents should be categorized for insurance coverage purposes. The court explained the two prongs of the test: (1) identifying the operative incident giving rise to liability and (2) determining whether there is a close temporal and spatial relationship between the incidents and whether the incidents can be viewed as part of the same causal continuum.

The court found that the policy language was similar to language that the New York Court of Appeals had determined was not the type that evinced an intent to aggregate incidents. If policy language shows an intent to aggregate incidents, the unfortunate event test does not apply. That was not the case here. The court went on to explain how these were three separate incidents that could not be aggregated.

The court adopted the practical approach that has been used by the New York Court of Appeals and found that the collision with the overpass was separate and apart from the two car collisions with the box. The incident that damaged the overpass played no role in the car accidents. The court also found that the first car collision with the box played no role in the second car’s collision with the box moments later. Although there was spatial proximity between the second and third incidents, the latter two were not part of the same causal continuum because each incident was its own independent causal chain. In other words, each collision was unrelated to the preceding collision.

Lots of issues arise from this analysis, including what kind of language must an insurance company use to evince an intent to aggregate incidents and how close does a collision have to be short of a chain-reaction collision to be considered part of a single causal continuum? It comes down to what the parties intended when they entered into the insurance policy with language providing that bodily injury or property damage resulting from continuous or repeated exposure to substantially the same conditions will be treated as one accident or occurrence. We know that the insurance company believed it meant aggregation for incidents that arose from the same event and that the policyholders believed that separate limits were required for separate incidents.