[ExI] libertarian (asteroid) defense

[Anders Sandberg]

Kelly Anderson wrote:
> (Sorry for being dense, what is a GCR? Galactic Cosmic Ray?)
>   
Global Catastrophic Risk. Sorry, our inhouse terminology (see 
http://www.global-catastrophic-risks.com/ )


> While the science points in the direction that climate fluctuation can
> occur faster than we previously thought likely, it still happens over
> a period of years or decades. Great climate change may lead to great
> suffering, but it is not going to be an extinction event.
When I used the term I was thinking of nuclear winters, impact winters 
and supervolcanic winters. These happen much faster. If current climate 
models are correct they can be *very* bad - *no* agriculture for several 
years in practically all places (so far Tasmania seems to be the only 
place that survives, but I would like to run a more fine-grained sim 
since it could be surviving mainly due to mesh coarseness). While this 
is not necessarily an extinction event we should expect at least several 
gigadeaths as well as some pretty nasty conflicts over necessary resources.

Climate change is indeed a reshuffling of the cards, in itself fairly 
neutral but on one hand breaking down structure (all the winery 
infrastructure will be in the wrong place, and it takes money, time and 
expertise to build it somewhere else) and on the other affecting people 
differently depending on their resilience (the dirty secret of climate 
impacts on society research: the developed world is fairly likely to 
withstand even pretty big climate effects, while the undeveloped won't). 
In fact, you can almost define being poor as "getting screwed when there 
is a crisis" - radical change favors resilient systems, and generally 
being well-off means being embedded in a resilient civilian 
infrastructure while being poor means you have no reserves to use when 
things go pearshaped. This is worth thinking about for us who like 
radical, transformative change.


> If you can sell insurance against climate change, then I see no reason
> you couldn't sell insurance against mid sized asteroid strikes.
> Insuring against a global extinction event is silly since there would
> be nobody left to pay or collect, and nothing to do with the money
> after that point anyway. I suppose it could mean something if some
> part of humanity lives off world...
>   

You can set up near miss gambles, and the pricing of these might give 
you useful information.

Insurance systems can handle uncorrelated disasters well, but correlated 
disasters are bad news. There was apparently a short period after 
Katrina when the world's reinsurance system was so badly strained that 
had a second Katrina-disaster happened it would have crashed globally - 
think of the financial crisis but in insurance instead. This is why I am 
a bit worried about even "mere" 1 km impacts. An Atlantic tsunami would 
cause correlated insurance claims far in excess of Katrina, and the 
industry would not be able to handle it. Life would go on, but quite 
likely with a seriously crashed global economy for a long while.

Still, finance people are clever. I would love to have more of them 
think seriously about financial vehicles that can deal with this and 
other GCRs. There are already longevity bonds, helping insurance 
companies trade on the "risk" that people live a lot longer. There 
should be GCR-vehicles.



-- 
Anders Sandberg,
Future of Humanity Institute 
James Martin 21st Century School 
Philosophy Faculty 
Oxford University