[ExI] Existential hysteria

[Anders Sandberg]

Tomaz Kristan <protokol2020 at gmail.com> , 25/7/2014 8:42 AM:
Anders wrote:

> It also encompasses the reinsurance industry, a business that is very  careful to have as accurate risk models as possible and does lose money  when they are wrong.

I think, the insurance companies use AGW as a PR stunt and maybe as an excuse to be more expensive. No hard data proves the link between calamities and AGW, but it is a PR important stuff.
I have been to Lloyd's meetings where they have carefully looked at climate data and discussed how it affects their risks. If they are doing it as a PR stunt, the Chief Risk Officers of the industry all seem to be fooling each other and their own teams. I doubt the insurers are just naively following AGW hype: they are seriously trying to be on top of the situation in order to beat their rivals.

The effect of climate change on perils is actually quite nontrivial (one reason for those meetings: more risk is not a problem to the industry, it is uncertain risk that is a problem). If everything gets equally warmer hurricane frequency and strengths might not be affected at all since they are being powered by a temperature difference. But the sea-stratosphere gradient is complex and hard to model reliably. Even without a change in hurricanes themselves a different climate may cause changed paths, and that is serious business. One discussion I participated in was the issue of unlocking faultlines by melting glaciers: most of the time a really small-scale slow issue, but even a tiny increase in extreme tail events in areas believed to be seismically stable  is a big deal (check out the Lapland fault province - huge postglacial quakes in otherwise totally stable Scandinavia). Models of changes in precipitation distribution are very uncertain, and the flood models they feed into are fairly sensitive to changes in land cover, which also clearly and nontrivially gets affected by temperature. And most obviously, even small increases in sea level have big effect on flooding probabilities: it is not like a one centimetre increase makes the flood level one centimetre higher (exceedances are always *way* higher than expected), but that they become much more common. But the sea rises are unevenly distributed. And so on and on and on... 
If you actually think you know the risks better than the industry, you could be making billions. 

Anders Sandberg, Future of Humanity Institute Philosophy Faculty of Oxford University
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.extropy.org/pipermail/extropy-chat/attachments/20140725/2598c959/attachment.html>