Colorado Court of Appeals: anti-concurrent cause provision bars coverage where covered snow combined with uncovered rot to cause loss
Thanks to one of my loyal readers who often points out anti-concurrent cause cases for passing along the Colorado Court of Appeals July 24 decision in Colorado Intergovernmental Risk Sharing Agency v. Northfield Ins. Co.
This opinion is right on the money in its analysis of how anti-concurrent cause language works. In the case, a roof on a building containing a pool collapsed from the weight of snow, but the timbers supporting the roof were rotted from humidity and chemicals from the pool (this kind of rotting due to pools happens more often than you would think). The way the case came to trial was this: CIRSA (I hate acronyms as much as the next English major/former journalist, but in this case the name is a real mouthful, so I'll compromise my standards a bit) had the primary level of insurance, and Northfield had the second layer. CIRSA paid out on the loss, and Northfield declined to pay CIRSA for any of the money CIRSA had paid, citing an anti-concurrent cause provision in the Northfield policy. The anti-concurrent cause language was the standard Insurance Services Office clause:
We will not pay for loss or damage caused directly or indirectly by any of the following. Such loss or damage is excluded regardless of any other cause or event that contributes concurrently or in any sequence to the loss.
The trial court correctly said Northfield didn't have to pay for any of the loss that was due to excluded rot, but incorrectly allowed the jury to apportion the damages between covered snow weight and uncovered rot. Question: what do you think the jury's apportionment was? C'mon, just take a guess, knowing how juries love to stick it to insurance companies, even when the plaintiff is a government agency. That's right: 90 percent due to snow.
On appeal, the Colorado Court of Appeals got it right. This case presents a classic example of a true concurrent cause: the loss was caused by a combination of factors that arose independently, but the loss would not have occurred but for the combination of the two. That is the best, shortest definition I have come up with of what a concurrent cause is. The key, once again, and I've said this as often as I can because I've come to see that it is counterintuitive to most people, is to look first to what the loss is and define the loss. Unlike in Hurricane Katrina cases, the damage to items of property was not due to discreet and separate causes -- for example, first covered wind causes some damage to an item of property, and then uncovered flood causes some more. That is not an example of concurrent cause, because a given item of property acted upon by each force was damaged by each force separately and in its own way. The "loss" in Katrina cases was not the total damage to the house, but rather loss to specific items of property -- a house consists of many items of property.
In the CIRSA case, however, the "loss" was to the whole -- apparently there were no items of property where you could segregate out damage from covered snow from the involvement of rot in helping the snow cause damage. To put it another way: under normal circumstances where there was no uncovered rot, the roof would have held up the snow without collapsing. In Katrina cases, the wind damage was not dependent on the existence of flood, so the "loss" in those cases cannot be caused by concurrent forces.
Also, thanks to the Colorado Court of Appeals for favorably citing my work on anti-concurrent cause: the particular article they cited was Katrina in the Fifth Dimension, which appeared earlier this year in New Appleman on Insurance: Critical Issues, and which examined the U.S. Fifth Circuit's Katrina decisions. I think it's the third case -- that I'm aware of -- that has cited to my analytical methodology, the first state court case and the first non-Katrina case. I hope my methodology continues to catch on, because it truly is easier to use and brings more predictable results than any other. In reality, there really aren't any competing methods of analysis, other analysis is more like what happens when you stub your toe -- hopping around, shouting and hoping the pain will just go away.
So the analysis and the result of this case are correct: in this instance, unlike a lot of things people have come to think of as concurrently caused, the loss actually was due to concurrent forces, and therefore none of the loss was covered by the Northfield policy. Now some people will see this as crazy and unjust no matter what I say, and I will only point out that I am describing what I see in how anti-concurrent language works, without regard to which side is going to win in any given case. As it turns out, the prior two cases that cited my anti-concurrent methodology found for the policyholder. Whether anti-concurrent language exists or not is of no particular concern to me, but since it does exist, I'd like courts' analyses of it to be as focused, sharp and correct as possible, that's all. In other words: I'm just sayin'.
Colorado Court of Appeals citing your work....we've been telling you we need your blog because youre good--and now the court agrees!
I'm happy to know your logic is recognized as an authority on this subject. Congratulations, Your Precedence!
David, I don't know whether to thank you or curse you for igniting my interest in anticoncurrent cause problems. You make it sound so simple, so... obvious (which is why I would cite your Appleman's article if I were a judge, too). But it's kind of like a Rubik's Cube for me. I can't put it down, and just when I think I've got all the green squares lined up, the pesky white one always shows up out of nowhere. So I continue to be fascinated, frustrated, and constantly referring to your work to get me back on track. Thanks. I think.
I didn't do so much. On most levels, a correct analysis involving anti-concurrent cause is actually much simpler to do than to do an incorrect analysis. Like anything involving complex ideas, it's best to break up the analysis into distinct parts and go step by step. All I did was to find a methodology that could explain it, and that gave consistent results. The only real insight I had was that identifying the loss and understanding what the term "loss" means in this context are the keys to the analysis.
How many cases, percentage wise include anti-concurrent language. Interested in how often this comes up.