Sunday, December 31, 2017

Betting With Catastrophe Bonds

Betting on climate carries some risk for realists, not so much losing the bet but if, as  here and others there, a bunny bets that global temperature will increase, perhaps that is not a bet an ethical hare would care to win.

Bets do have their virtues because as Steve Schneider put it
On the one hand, as scientists we are ethically bound to the scientific method, in effect promising to tell the truth, the whole truth, and nothing but — which means that we must include all the doubts, the caveats, the ifs, ands, and buts.
and you can get tied up in knots trying to convince others that that last detail is not very likely.  The virtue of bets is they simplify things
On the other hand, we are not just scientists but human beings as well. And like most people we'd like to see the world a better place . . .
and win a few bucks.  Bill Foster, the only physicist in the US Congress put it another way
On the campaign trail, I learned that there is a long list of neurons that you have to deaden to convert a scientist's brain in to a politician's.  When you speak with voters, you must lead with conclusions rather than complex analysis of underlying evidence -- something that is very unnatural to a scientist. 
and even some of the most obdurate denialists recognize this virtue of bets, for example  Gosslein from the No Tricks Zone
If the 2011/20 decade averages to be warmer than 2001/10, then I will concede that the earth is indeed warming. But if the next decade turns out to be cooler or the same, then you will have to concede that the theory that CO2 is driving the climate is bunk. We can work out the details in the days and weeks ahead.
That, as Eli would say, looks like a losing bet

Now in some cases the losers bets pay to charity and in other cases, some bunnies simply are looking for a counter party and not finding any takers.

An interesting presentation at this year's AGU was by Mark Roulston from Winton Capital, a hedge fund that plans to set up a climate betting market in 2018 (not open to US inhabitants) in order to take advantage of the wisdom of crowds (they are familiar perhaps with the results of elections in the US and UK, maybe not).
 The initial market will allow bets to be placed on the atmospheric concentration of carbon dioxide and the global mean temperature anomaly. It will thus produce implied forecasts of carbon dioxide concentration as well as global temperatures. If the initial market is successful, additional markets could be added which target other climate variables, such as regional temperatures or sea-level rise. These markets could be sponsored by organizations that are interested in predictions of the specific climate variables.
If Eli plays the IPCC chalk and wins, that leaves neither Eli nor the world in a good place.  So what could one do.

While thinking about the issue the Bunny came across Catastrophe Bonds, a high risk high interest investment where the issuer pays the buyer interest, but if catastrophe strikes the buyers don't get their capital back.  For example, the New York has issued catastrophe bonds to cover flooding in the subway tunnels, and there are lots of them in Florida covering hurricane damage.

Well, that is an investment for Roger P Jr, and he would have done well for the last seven years or so, and lost his shirt last, but what about Eli

Eli has an idea, a catastrophe bond market where the interest is split according to climate outcomes.  When a catastrophe occurs part of the capital is used for relief.  These bonds could be issued by governments, or perhaps organizations such as the World Bank which has issued catastrophe bonds covering hurricane and earthquake damage in Mexico.  The bonds could specify whether the capital would be used for amelioration, adaptation, conservation, substitution or mitigation.

A new ethical playground for financial engineering                         


William Connolley said...

> don't get their interest back

*Principal* not interest.

William Connolley said...

Looking at, cat bonds are effectively a form of reinsurance. Indeed so much so that it isn't clear to me what extra they bring over reinsurance. Perhaps only a different risk profile of investor? But as wiki says, many of them are actually issued by Re companies, so no not that.

Fernando Leanme said...

The question in my mind is: who gets to plug in the temperature? Once there's big money involved, there's going to be pressure to arrive at a figure some high roller wants to have.

Lately I've been trying to figure out a better way to measure energy changes, concluded an array of 100 anchored buoys measuring temperature all the way to the sea floor, and 100 mountain top stations away from urban areas, which measure both the location temperature as well as the clear sky temperature (they have their sensors looking up) may be more reliable and serve to calibrate everything else.

Bernard J. said...
This comment has been removed by the author.
Bernard J. said...

Lately I've been trying to figure out a better way to measure energy changes...

Apparently FL hasn't heard of radiosondes, or the Argo network, or even about random sampling.

EliRabett said...

Thanks Weasel. Answer to your q They are less expensive than reinsurance against significant risks

EliRabett said...

FL you ever read a bond offer? The specification is done in the offer.

Fernando Leanme said...

Eli, I only buy bond funds. The point is that if there's big money in the way temperature is used, then there will be pressure to cheat. The idea you discuss was already proposed by Pielke Junior in the carbon tax field, he thinks it will lead to better climate models.

Somebody asked me if I knew about Argo etc. I guess this person doesn't understand that bond clauses are best when tied to simpler easier parameters, and it's also clear it has never looked at the number of Argo buoys which fail to deliver data.

I'm also interested in seeing temperatures measured al the way down, as well as up, from fixed stations, because these can really help reanalysis. And as we all are supposed to know, reanalysis is the best tool to measure ocean energy content. And that's all I have to say about that.

Bernard J. said...

"...and it's also clear it [sic] has never looked at the number of Argo buoys which fail to deliver data.

I repeat, FL, you seem to be oblivious to the power of random sampling and also of the Argo network despite any issues that might occur with individual buoys (refer to first point...). I am satisfied that the network accurately reflects the volumes that it samples, and I once had a long 'phone conversation with Susan Wijffels about this very topic, back around the time that I was lugging around an earlier generation bouy with the innards removed, to show to students. The Argo system is state-of-the-art, and the newer iterations of the bouys are going to be manna for data-pigs everywhere.

I am certainly satisfied that the Argo network provides a far superior solution to profiling global ocean temperatures than would your string of a hundred static buoys. However, you've made a testable claim so out with it - what is your evidence that the Argo network is somehow unreliable, and that your solution is somehow better?

" clauses are best when tied to simpler easier parameters..."

Argo data are simple enough and easy enough to fathom* for the technical data interpreters of financial companies - after all, these folk are themselves eminently capable of "reanalysis". Or do you think that they can handle nothing more complicated than a toe dipped into the sea shore?

[*Yes, that was deliberate...]

EliRabett said...

Really deep of you