My sister sent me a link to an op-ed by Henry Paulson, which got me thinking about an old pet peeve of mine, which is the framing of climate-change policy as a matter of insurance, as for example here.
It’s an understandable temptation. With insurance, you pay a modest amount each year so that, in case something bad happens, you’ll be OK. With climate change, the idea is that we should pay a modest amount each year (take actions that cost the economy something) so that we’ll be OK.
But insurance isn’t actually what we’re talking about, and I wonder if thinking about climate policy in terms of insurance misleads us in a dangerous way.
The key is the difference between insurance and risk reduction.
Insurance doesn’t reduce the chance that a particular bad thing will happen. Instead, it puts together a bunch of people who are vulnerable to that same bad thing, but whose risks are (to some extent) independent, or uncorrelated. What that means is that one person having a bad year doesn’t make it more likely that someone else is also having a bad year.
When you have those two properties—a bunch of people with the same kind of risk, and their risks being independent—an amazing thing happens, which is that the risk for the group as a whole basically goes away.
Say each of you has a house worth $200,000, and you each face a small chance of your house burning down, something like 0.1% (a one-in-a-thousand chance). For you as an individual, the cost of fire will almost certainly be $0. But there’s a small chance that it will be $200,000. But for the group of you, the average cost of fire is right around $200.
As an individual, the only way you can prepare financially for fire is to have an extra $200,000 stashed away. If that’s beyond your means, then you have no way to prepare financially. But as a group you can each pay a mere $200 per year. When someone’s house burns, they’ll still have the emotional loss and the inconvenience of having their house destroyed, but they can be sure that they will be financially protected. Premiums from the people whose houses didn’t burn this year provide the means for financially compensating the few whose houses did burn and nobody in the group faces any financial risk at all.
But remember that the insurance doesn’t reduce the risk that houses will burn. What it does is ensure that when a bad thing happens, you’re not financially ruined by it.
Climate change isn’t like that at all.
The idea of climate insurance comes up in the context of the possibly catastrophic outcomes of climate change: complete loss of the Greenland ice sheet (24 feet of sea-level rise), or even the Antarctic ice sheet (200 feet of sea-level rise); massive drought or flood; losses of economically vital species; etc. (representative cheery reading can be found here).
When people bring up catastrophic outcomes, they’re talking about the kind of thing that Harvard economist Marty Weitzman describes here: “Thus a temperature change of »10°C would appear to represent an extreme threat to human civilization and global ecology as we know it, even if it might not necessarily mean the end of Homo sapiens as a species.” (p. 8 of the PDF file, p. 282 of the article)
OK, that sounds bad. What to do?
One response to a scenario like that is to argue that it’s a very low-probability event, so it’s not worth doing much about, and that’s where the insurance framing comes in: Sure, catastrophe isn’t likely (though it seems like it’s becoming less unlikely the longer we do nothing), but since it is, you know, catastrophe, isn’t it worth taking some actions as insurance?
Except that it’s not insurance. If there were such a thing as climate insurance, we would pay some portion of global GDP each year, and if a catastrophe struck, the insurance company would give us the financial means to recover.
That would make sense if there were a population of planets roughly like ours, with comparable risks of destroying their terrestrial support systems through anthropogenic climate change. The Assicurazioni InterGallactici (AIG) would collect premiums from each of these planets. Roughly once a millenium, the inhabitants of some planet or other would undermine the basis of their own existence. Each time that happened, AIG would use the premiums paid by the other planets to pay for the rehabilitation of the planet that just turned itself into a wasteland.
But of course there is no such population of planets. There’s nobody from whom we can draw the means to rejuvenate our one Earth if we irredemably screw it up.
Think of it this way. Having health insurance doesn’t mean you won’t get lung cancer. It means if you do get lung cancer, the premiums paid in by you and a bunch of other people will cover some large portion of the costs of your care, so that you can actually afford to be treated. But you will still have gotten lung cancer, and there’s a pretty good chance you’ll die.
But it’s worse than that. It turns out you’re the only person in the world, so there’s nobody with whom you can pool your risks. Either you don’t get lung cancer, and you’re fine, or you do get lung cancer, and have no way to provide for your treatment. Oh, and you’re a heavy smoker.
What we need is not insurance against climate change, which nobody can possibly sell us anyway. We need to stop smoking. Phrasing it as buying insurance may lull us into a false sense of security by makiong the problem seem more familiar—and thus more tractable—than it actually is.