Will The Real AGW Skeptic Please Stand Up?
Normally, I don’t debate random bloggers on Anthropogenic Global Warming (AGW). However, I made an exception for Robin Hanson. For those who don’t already know of him, he was both an early proponent of decision markets and has a reasonably well known journal article on why two Bayesian rationalists can’t agree to disagree. I’m a fan of his work and have been reading his blog for years.
Yesterday, he put up a post titled CO2 Warming Looks Real. He’s not an expert. Like me, he has an economics background and did some detailed research. Yet from the title and body of the post, I though he must have reached a very different conclusion than I did. So I thought I’d try to engage him to find out where we differ. The results were interesting.
Obviously, and as Robin knows, the best way to elicit a person’s true beliefs is to observe where he puts his money. So I offered to negotiate a bet of up to $1,000. As you can see from reading the comments, he eventually offered an even-odds bet than the temperature would rise 0.1 deg C in 20 years. Yes, you read that correctly: 0.1 deg C.
So he’s as much of a skeptic as I am! I think a doubling of CO2 concentration would lead to about 1 deg C of warming. I think, it’s going to take about 200 years (from 1900-2100) to do that. So I believe that we should see warming of about .05 deg C per decade. In 20 years, that’s about… 0.1 deg C. (Yes, I know I’m linearly approximating a logarithmic function, but the amount of precision in all the input values is low enough that I don’t think it matters).
Now, as I pointed out to Robin, the official IPCC estimates are much higher. If you go to the latest Summary for Policy Makers and look at Table SPM.3, you see the rate of warming in the lowest impact scenario B1 is 2x to 5x higher. So Robin’s price reveals that he is quite skeptical of the IPCC estimates.
Moreover, if you believe that the over-under on temperature in 2100 is .45 deg C higher than today, why would you think that justifies any significant mitigation effort? To the extent that there are any costs, we could certainly pay them at a fraction of the economic cost of a carbon tax or cap-and-trade. Moreover, you’re saying that GHG emissions account for only a small part of the variance in future temperature. So it could get a lot cooler as well as a lot warmer. Seems like that implies we should save our money for adapting to whichever way the thermometer swings.
Personally, I’m pleased to know that Robin Hanson did not reach different conclusions from me and that, by my definition, he is a fellow member of the “well-informed skeptic” club.
Possible Insight 
There are a range of estimates I won’t bet against, and an even wider range of estimates I wouldn’t blog. That is, I require a higher standard of evidence to blog something than to bet it, and that betting standard is higher than my decision standard, the estimate I would use with a gun to my head. So I don’t disagree with your estimate, but that doesn’t mean I’ll claim it either.
Robin Hanson
June 5, 2009 at 5:44 pm
Thanks for stopping by Robin.
This is precisely why I offered the bet. As I am both unwilling and unable to put a gun to your head, it was the best I could do in terms of eliciting your true decision standard 🙂 That’s what I wanted to calibrate my own probability assessment.
If you do check back here, I would be curious to know what bet you would take on the other end, i.e., at what 20 year temperature increase would you take the under? .25 deg C (~1.25 deg C higher in 2100)? .5 deg C (~2.5 deg C higher in 2100)? 1 deg C (~ 5 deg C higher in 2100)?
FYI, I am currently negotiating a 20 year temperature bet with another blogger. The details turn out to be complicated. More updates as events warrant.
kevindick
June 6, 2009 at 11:24 pm
I am far more interested in, and expert at, helping to create betting markets on this stuff, than in estimating my own betting odds. A great opportunity is being missed here, so far at least.
Robin Hanson
June 7, 2009 at 4:24 am
Well, the first step in setting a market up is getting some transactions done. It seems like your making a bet or publicly issuing your odds would help facilitate this.
Of course, the prediction market is just a means to calibrate one’s probability assessment which was the primary reason to offer you a bet. Given the similarity of our backgrounds, I expected our priors to be similar. So offering you a bet helped uncover the perceived difference in those priors.
As I said, when I negotiate a bet, I’ll do a post. I think it would be great if you could figure out how to refine, parameterize, and syndicate it in such a way so that people on markets like InTrade could make the full range of bets on future temperature.
I am trying to do a principled job in the negotiations so that it is as reasonable a template as possible.
kevindick
June 7, 2009 at 10:16 am
[…] 24, 2009 by kevindick As I mentioned in the comments on this post, I am currently in the process of negotiating a bet on Anthropogenic Global Warming (AGW) with […]
Specifying a Climate Bet « The Emergent Fool
June 24, 2009 at 2:12 pm
[…] I mentioned in the comments on this post, I am currently in the process of negotiating a bet on Anthropogenic Global Warming (AGW) with […]
Specifying a Climate Bet « Possible Insight
September 27, 2011 at 3:00 am