Construing Collapse Under a Homeowners’ Insurance Policy

By Larry P. Schiffer 

Homeowners’ policies have become more complex as more and more homes have been built around the country. With the increase in natural and other disasters, including construction defect claims, homeowners have looked to their policies for coverage when disasters have destroyed or nearly destroyed their homes. A recent case highlights a couple of the issues that courts have faced in construing whether a homeowners’ policy must respond to a collapse or near collapse.

In Fabozzi v. Lexington Ins. Co., No. 15-911.cv (2d Cir. Mar. 7, 2016)(Summary Order), the Second Circuit Court of Appeals construed a homeowners’ policy after the district court had granted judgment in favor of the insurance company.
Two issues were discussed by the court in its Summary Order. First, who had the burden of proof on the cause of the collapse, the policyholder or the insurance company. Second, whether the policy provided coverage if the collapse was caused by one of the enumerated causes or must the collapse have been caused exclusively by one of those enumerated causes. If you are keeping score, the insurance company won the first issue, but the policyholder won (enough to get the judgment vacated) the second issue.

The first issue was about whether the policy, for purposes of the collapse claim, provided all-risk coverage or named-perils coverage. The difference, as discussed by the court, is critical. Construing New York law, the court noted that under all-risk coverage, the insurance company bears the burden to prove that the loss was caused by a peril specifically excluded from coverage. All the policyholder has to do is demonstrate the existence of an all-risk policy and that there was a loss. This, of course, is consistent with the majority view that it is the insurance company that has the burden to demonstrate that an exclusion applies. With a named-perils policy, the policyholder has to demonstrate that the enumerated peril caused the loss.

The interesting thing in this case was that the court (and all parties) agreed that under Coverage A of the policy, all-risk coverage was provided. But the collapse claim arose under Additional Coverage 8, which the court construed as named-perils coverage. The coverage grant provided that the policy covered “direct physical loss to covered property involving collapse . . . caused only by one or more of the following, . . .” The policy then listed six casual agents that triggered coverage. Collapse was excluded from Coverage A, but was written back into the policy in Additional Coverage 8, which the court read as a named-perils provision.

On the second issue, the court held that the phase “caused only by” in Additional Coverage 8 was ambiguous and, accordingly, required that the judgment be vacated. While agreeing that the provision obviously limited in some fashion the perils against which the policy provides coverage, the court found that each parties’ reading of the clause was equally sound. The court looked to the law of proximate cause and found that in a case with covered and non-covered perils the efficient proximate cause rule will entitle the policyholder to coverage if the covered peril is the predominant cause of the loss — not just the only cause. 

The court also pointed out other policy provisions where the policy made it clear that exclusivity of the cause was required. Finally, the court found that the reasonable expectations of a policyholder whose home collapses is that if the collapse was caused predominantly or the most important cause of the collapse was a covered peril that there would be coverage. Thus, because the provision was ambiguous, it should have been construed in favor of the policyholder.

On remand, assuming the matter does not settle, the insurer will have to prove that the collapse was caused by an excluded peril and the policyholder will have to show that the collapse was caused predominantly by a covered peril so that it meets the test of efficient proximate cause of the collapse.

Original