It may not be unreasonable to think that, if you examined a judge’s entire body of opinions, addressing a certain issue, over a decade or more-long career, a pattern would emerge. You can imagine that you would get into the judge’s head and pretty soon you’d be able to predict how he or she would rule in similar cases. Maybe this is so when it comes to certain ideological issues, such as First Amendment or the death penalty. But each time that I have tried to find a pattern in a judge’s rulings in insurance coverage cases I have found that it is a fool’s errand.

My examination of Judge Robert Smith’s opinions – majority and dissenting -- in property-casualty coverage cases, and some others insurance-related, over his eleven years on the New York Court of Appeals, was no exception. I looked at 20 such opinions as part of the preparation for my interview of the Judge (see nearby article). My conclusion -- I could not tell you how he would rule in the 21st. As I see it, neither side in a coverage dispute could have counted on his vote before seeing the opinion. I guess that means he was doing his job the way judges are supposed to. But this is not to say that I did not see a pattern.

As for policy language-based cases, they are sui generis -- at least factually, even if the same contract provisions are at issue. Each case should stand on its own. Any comparison between such cases should be apples to oranges. When it comes to this type of coverage case, Judge Smith told me that he, and the Court, followed the rule that in a close case you find coverage. There was plenty of evidence of this in his opinions. I saw no propensity, one way or another, for Judge Smith to find policy language to be ambiguous or not.

For example, in determining whether an earth movement exclusion applied to excavation, Judge Smith concluded that both the policyholder’s and insurer’s readings of the exclusion were reasonable. As a result, he was bound to adopt the reading that resulted in coverage. See Pioneer Tower Owners Ass’n v. State Farm Fire & Cas. Co., 908 N.E.2d 875 (N.Y. 2009). More recently, in Nesmith v. Allstate Ins. Co., -- N.E.3d. -- (N.Y. 2014), Judge Smith held that the language of a “non-cumulation clause” was clear and capped an insurer’s maximum total liability to only one policy limit, despite that members of different families were successively exposed to lead paint, in the same apartment, over more than one policy period.

While I saw no pattern in Judge Smith’s opinions that were tied to the interpretation of policy language, I did see a pattern in his decisions where the coverage issue was conceptual and not contractual. In such opinions he may have considered the nature of insurance, and the risk business itself, in reaching his decisions. While a few of these examples come in the form of dissenting opinions, the most impactful one, a majority opinion, used this rationale to hand a significant win to a policyholder.

In Voss v. Netherlands Ins. Co., 8 N.E.3d 823 (N.Y. 2014), Judge Smith, in a dissenting opinion, would have preferred to see the court take a narrower view of insurance agent liability in the situation before him: “It is natural for a client that has suffered a loss not covered by its insurance to blame its insurance agent; and if lawsuits by clients against their agents are welcomed by the courts, the consequence may be to make the agent into a kind of backup insurer, a result neither sensible nor fair.” [Of course, the agent’s professional liability carrier is then the backup, backup insurer.]

In Kramer v. Phoenix Life Ins. Co., 940 N.E.2d 535 (N.Y. 2010), Judge Smith, in a dissenting opinion, would have ruled that it was unlawful for one to purchase a life insurance policy on his or her own life, and then transfer it to someone without an insurable interest in that life, even if the policy was obtained for that purpose. He stated: “Even if we ignore the possibility that the owner of the policy will be tempted to murder the insured, this kind of ‘insurance’ has nothing to be said for it. It exists only to enable a bettor with superior knowledge of the insured’s health to pick an insurance company’s pocket.”

In Bi-Economy Market, Inc. v. Harleysville Ins. Co., 886 N.E.2d 127 (N.Y. 2008), Judge Smith dissented from the majority’s opinion, which held that an insured could seek consequential damages arising out of an insurer’s breach of policy obligations. He stated: “[W]here there is no agreement on what money will be paid in the event of a breach, a court must try to decide what damages the parties contemplated--what damages they would have agreed to had they considered the question when the contract was signed. But in insurance contracts or other contracts for the payment of money, the parties have already told us what damages they contemplated; in the case of insurance, it is payment equal to the losses covered by the policy, up to the policy limits.”

In Union Carbide Corp. v. Affiliated FM Ins. Co., 947 N.E.2d 111 (N.Y. 2011), Judge Smith addressed whether the aggregate limit, in a three year excess liability policy, covering asbestos claims made against Union Carbide, applied annually or continued over the three-year life of the policy. The insurers argued that one limit applied and Carbide argued for three. Judge Smith held that three aggregate limits applied. In doing so, he was persuaded less so by tedious policy language than a more fundamental issue: “UCC has the better of the argument. While the reading Continental and Argonaut give to the word ‘aggregate’ might be plausible in many contexts, here the follow-the-form clause should prevail. Such clauses serve the important purpose of allowing an insured, like UCC, that deals with many insurers for the same risk to obtain uniform coverage, and to know, without a minute policy-by-policy analysis, the nature and extent of that coverage. It is implausible that an insured with as large and complicated an insurance program as UCC would have bargained for policies that differed, as between primary and excess layers, in the time over which policy limits were spread. Under Continental’s and Argonaut’s reading, UCC could (and in fact did) reach the second and third years of its excess policies with the full limit of its primary coverage in place, but with its fifth-layer excess coverage exhausted. It is unlikely that the parties intended this result.”

Lastly, looking at New York’s grand dames of insurance law, § 3420(d) and late notice, Judge Smith mandated strict compliance with both notice-related rules. That can benefit both insurers and policyholders – depending on the issue. In Sierra v. 4401 Sunset Park, LLC, -- N.E.3d – (N.Y. 2014), he held that a contractor’s insurer, that sent a disclaimer to an additional insured’s insurer, but not to the additional insured itself, did not meet the requirement of § 3420(d). In Argo Corp. v. Greater New York Mut. Ins. Co., 827 N.E.2d 762 (N.Y. 2005), Judge Smith strictly applied New York’s (then) no prejudice required rule for a liability insurer to disclaim coverage based on late notice: “A liability insurer, which has a duty to indemnify and often also to defend, requires timely notice of lawsuit in order to be able to take an active, early role in the litigation process and in any settlement discussions and to set adequate reserves. Late notice of lawsuit in the liability insurance context is so likely to be prejudicial to these concerns as to justify the application of the no-prejudice rule.” See also Briggs Ave. LLC v. Insurance Corp. of Hannover, 899 N.E.2d 947 (N.Y. 2008) (following Argo Corp.).

But Judge Smith also held that there can be times when requiring strict compliance with a notice provision is simply not reasonable. In Executive Plaza, LLC v. Peerless Ins. Co., 5 N.E.3d 989 (N.Y. 2014), he wrote that a clause in a fire insurance policy, limiting the time in which the insured may bring suit under the policy to two years, running from the date of the fire, is unreasonable and unenforceable when the policy also says that the insured may recover the cost of replacing destroyed property—but only after the property has already been replaced. Judge Smith concluded: “Thus, if (as happened in this case) the process of replacing the property takes more than two years, the insured’s claim will be time-barred before it comes into existence.”

Perhaps it should not come as a surprise that my examination of 20 of Judge Smith’s insurance opinions left me unable to predict how he would rule in such future cases. After all, I’ve been married to my wife for seventeen years and I can’t predict how she will respond to anything I say.