Causation and Anti-Concurrent Issues in Connections with Superstorm Sandy


Superstorm Sandy caused widespread damage across several states in the Northeast. Chief among these are New York and New Jersey, followed by Connecticut. The damage has resulted in claims under policies that are written on a wide variety of forms. Each claim requires careful analysis of the specific policy covering the risk at the time of the loss in order to determine coverage.

This article is not an exercise in analyzing coverage for losses caused by Sandy under any particular policy. Instead, it is our intent to share with the industry the basic insurance law regarding causation and anti-concurrent cause exclusions in New York, New Jersey and Connecticut in order to alert the reader to these nuances.

Many claims have already raised questions concerning the cause of the loss and the application of anti-concurrent causation exclusions. Our research disclosed that the courts in these jurisdictions have developed different approaches to identify the cause of a loss for purposes of determining coverage. Further, law regarding the application of anti-concurrent causation language is not equally developed in all three states. These differences give rise to a host of questions concerning the position that insurance carriers will take regarding the cause of damage and which jurisdiction’s law will apply to Sandy claims. This is especially true in situations where the insured has multiple locations in different states.

The proximate efficient cause of an insurance loss requires a different analysis under the laws of New York, New Jersey, and Connecticut. The New York causation analysis looks to the “proximate, efficient and dominant cause” of the loss. New Jersey courts differ and look to whether the first or last step in the chain of causation resulting in the loss was a covered event. Connecticut courts take a third path, and employ a “train of events” test that looks to the first event in the chain of causation.

Anti-concurrent cause clauses have not been addressed by the highest court of any of the these jurisdictions. Lower court decisions in New York and New Jersey have upheld such exclusions. The courts of Connecticut have not addressed such language, although references to anti-concurrent cause exclusions suggest that Connecticut courts recognize the significance of the language and at least one federal court decision has applied it.


New York courts look to the “proximate, efficient, and dominant” cause of a loss when addressing issues of proximate causation. Album Realty Corp. v. American Home Ins. Co., 80 N.Y.2d 1008, 1010, 607 N.E.2d 804, 805 (N.Y. 2002); see also Wilner v. Allstate Ins. Co., 2011 N.Y. Misc. LEXIS 1424, 2011 NY Slip Op 30817U (Sup. Ct. Nassau Cty. March 28, 2011), aff’d 2012 N.Y. App. Div. LEXIS 6534 (2d Dep’t Oct. 3, 2012).

The term “proximate, efficient, and dominant cause” has not been explicitly defined. New York courts will, however, consider what a reasonable businessperson would conclude to be the cause of the loss in their analyses. Album Realty, 607 N.E.2d at 805; see Wilner v. Allstate Ins. Co., 2011 N.Y. Misc. LEXIS 1424. Only this “most direct and obvious cause” will be considered in the application of exclusionary language. Album Realty, 607 N.E.2d at 805.

When other excluded events occur closer in time or space to a loss than a covered event, the “question of whether the covered event was sufficiently proximate to the loss to require that the insurer compensate the insured will depend on whether it was the dominant and efficient cause.” Throgs Neck Bagels, Inc. v. GA Ins. Co. of New York, 241 A.D.2d 66, 69 (1st Dept. 1998).

Although the issue has not yet come before New York’s highest court, lower appellate courts have generally enforced anti-concurrent causation provisions. See, e.g., Cali v. Merrimack Mut. Fire Ins. Co., 43 A.D.3d 415, 841 N.Y.S.2d 128 (2d Dep’t 2007) (upholding anti-concurrent causation provision containing language “regardless of any other cause or event contributing concurrently or in any sequence to the loss”).

Insurers should be aware of one federal case of potential significance to Sandy claims. The Southern District of New York has declined to apply the proximate efficient cause analysis with regard to the application of sublimits. The Hit Factory, Inc. v. Royal Ins. Co. of America, 2005 U.S. Dist. LEXIS 19050 (S.D.N.Y. Aug. 26, 2005). In Hit Factory, the loss involved water damage resulting from both flood caused by rain and an apparent drain back-up. The insurer made payments exhausting the policy’s flood sublimit, which was less than the amount of the loss, but declined to make additional payments under a separate sublimit for drain back-up as it contended the loss was proximately caused by flood.

The Southern District rejected this position and declined to apply the proximate efficient cause analysis. The court reasoned that because Hit Factory was not a case “where a plaintiff is suing for coverage under a policy for damages caused by two perils, one which is covered under the policy and the other is not, and where the insurance company is denying coverage” the proximate cause analysis should not apply. The court instead held that the terms of the policy created issues of fact as to whether the loss was covered under both sublimits. We anticipate that policyholders seeking to avoid New York’s proximate efficient causation analysis in cases involving sublimits may look to Hit Factory for support.


Courts in New Jersey apply a proximate cause test derived from “Appleman’s rule” on causation, which provides:

Where a peril specifically insured against sets other causes in motion which, in an unbroken sequence and connection between the act and final loss, produces the result for which recovery is sought, the insured peril is regarded as the proximate cause of the entire loss…. In other words, it has been held that recovery may be allowed where the insured risk was the last step in the chain of causation set in motion by an uninsured peril, or where the insured risk itself set into operation a chain of causation in which the last step may have been an excepted risk.

Auto Lenders Acceptance Corp. v. Gentilini Ford, Inc., 181 N.J. 245, 257, 854 A.2d 378, 385 (N.J. 2004), quoting 5 John Alan Appleman, Insurance Law & Practice, § 3083, at 309-11 (1970).

The practical application of Appleman’s rule is that “with regard to sequential causes of loss, [New Jersey] courts have determined that an insured deserves coverage where the included cause of loss is either the first or last step in the chain of causation which leads to the loss.” Simonetti v. Selective Ins. Co., 372 N.J. Super 421, 431, 859 A.2d 694, 700 (N.J. App. Div. 2004).

Where covered and excluded perils act concurrently to cause a loss, “it is for the factfinder to determine which part of the damage was due to the included loss and for which the insured can recover.” Id. One New Jersey court contemplated concurrent causation issues relating to a storm under a named-perils policy and held that the insured bore the burdens to prove: (1) that the excluded peril (i.e. the sea) was not the “sole or substantial cause of the loss;” and (2) what damage was actually caused by the covered peril (i.e. wind). Newman v. Great Am. Ins. Co., 86 N.J. Super. 391, 403, 207 A.2d 167, 175 (N.J. Super. Ct. 1965), citing Brindley v. Firemen’s Ins. Co. of Newark, N.J., 35 N.J. Super. 1, 113 A.2d 53 (N.J. App. Div. 1955).

Although there are few published cases in New Jersey addressing anti-concurrent causation provisions, it appears likely that New Jersey courts will uphold them. See Grossberg v. Chubb Ins. Co. of New Jersey, 2012 N.J. Super. Unpub. LEXIS 1981 (N.J. App. Div. Aug. 20, 2012) (upholding as unambiguous anti-concurrent causation provision containing language “however caused”); Petrick v. State Farm Fire and Cas. Co., 2010 N.J. Super. Unpub. LEXIS 1964 (N.J. App. Div. August 13, 2010) (upholding anti-concurrent causation provision containing language “concurrently or in any sequence”); Simonetti, 859 A.2d at 700 (suggesting that anti-concurrent causation provision would be enforceable if included in policy); Assurance Co. of Am., Inc. v. Jay-Mar, Inc., 38 F. Supp. 2d 349, 353-54 (D.N.J. 1999) (predicting that the New Jersey Supreme Court would uphold anti-concurrent causation provisions).


Connecticut takes a sequential approach to the proximate efficient causation analysis. Connecticut courts attribute causation to the “active efficient cause” and have also defined proximate cause to be “an actual cause that is a substantial factor in the resulting harm.” Frontis v. Milwaukee Ins. Co., 156 Conn. 492, 499 (Conn. 1968); see also Sansone v. Nationwide Mut. Ins. Co., 47 Conn. Supp. 35 (Sup. Ct. New Haven 1999).

The Connecticut analysis includes a “train of events test” that looks to the first event in time which sets off a sequence of events resulting in the loss as the “active efficient cause” of the loss. The Connecticut Supreme Court, quoting Couch on Insurance, explained:

In the determination of what caused the loss or damage, the cause or agency which is nearest in time or place to the result is not necessarily to be chosen. The active efficient cause that sets in motion a train of events which brings about a result without the intervention of any force started and working actively from a new and independent source is the proximate cause….In the determination whether a loss is within an exception in a policy, where there is a concurrence of two causes, the efficient cause – the one that sets the other in motion – is the cause to which the loss is to be attributed, though the other cause may follow it and operate more immediately in producing the disaster.

Frontis v. Milwaukee Ins. Co., 156 Conn. at 499 (citations omitted) (emphasis added).

Put another way, the proximate cause “is not that which is last in time or place, not merely that which was in activity at the consummation of the injury, but that which is the procuring, efficient and dominant cause.” Id. at 497-498 (citations omitted).

The Connecticut Supreme Court has not yet had the opportunity to address anti-concurrent causation provisions. There are, however, a few decisions where anti-concurrent causation language is quoted, but not applied due to facts at hand, suggesting that Connecticut courts may not be hostile to the concept. Cornish Contracting and Real Estate, LLC v. The Travelers Indem. Co., 2008 Conn. Super. LEXIS 840 (Sup. Ct. Conn. April 2, 2008) (holding that wind damage to roof would be covered loss subject to provision of exclusions including anti-concurrent causation language but deciding case on late notice grounds); Dlugokenski v. Hartford Ins. Co. of Illinois, 2010 Conn. Super. LEXIS 3346 (Sup. Ct. Conn. Dec. 14, 2010) (applying Ordinance or Law exclusion prefaced by anti-concurrent causation language without analysis of the anti-concurrent causation language).

In addition, the U.S. District Court for the District of Connecticut, in explaining Connecticut’s “train of events” test, held that “under an all risk policy where the efficient cause of loss is a covered risk, coverage is not defeated merely because an excluded risk contributed to the loss or constituted the loss (unless the policy expressly excludes such loss regardless of the antecedent cause).” Farrell and Reinhold v. Royal Ins. Co. of Am., 989 F. Supp. 159, 164 (D. Conn. 1997).


This basic research is only the beginning of the analysis that must be conducted in determining coverage for any insurance claim. The specific policy language in question must be compared to the facts of the loss and often further legal research must be done to determine if there is additional precedent that would govern your specific situation.

Further, a claim may require the carrier to conduct a choice of law analysis, particularly where the insured’s claim involves locations in multiple jurisdictions. We have encountered policies that contain choice of law provisions as well as choice of jurisdiction provisions. These two concepts should not be confused. A choice of law provision is intended to identify the jurisdiction whose law will apply to specific disputes. A jurisdiction clause may be limited to only an agreement as to the state where any lawsuit will be filed. Simply requiring a lawsuit to be filed in a specific state does not by itself require that the law of that state be applied to determining coverage for the claim.

© 2013 Robins, Kaplan, Miller & Ciresi L.L.P.

The articles on our Website include some of the publications and papers authored by our attorneys, both before and after they joined our firm. The content of these articles should not be taken as legal advice.