Thursday, November 23, 2006

More Stern Assessments

Stoat and Prometheus link to Nordhaus's comment on the Stern report, so I'll go one better and point out Gary Yohe's editorial too (via Richard Tol's comment). Both Nordhaus and Yohe point out the critical dependence of Stern's results on the unusually low discount factor (0.1%) used, and Nordhaus goes further than this in arguing that even with such a low discount factor, adjusting another model parameter (which he argues is necessary for the modelled economics to match observable reality) suggests much less support for aggressive mitigation.

It's worth pointing out that both Nordhaus and Yohe agree that some mitigation is appropriate, even though they find the detailed calculations of Stern at best hard to support.

Yohe also points to the climate science in Stern as being "perhaps the most persuasive contribution of the Stern Review", and says things look considerably more serious than they did in IPCC TAR. He has made his own contribution to this, (along with Michael Schlesinger), with his claim that "there is a one in five chance that we would lose the Gulf Stream before 2050." I don't think that claim is reasonable and as far as I'm aware the only place it has appeared in the literature is in the procedings of the "frenzied week of "climate change is worse than we thought" news reporting and group-think" otherwise known as the "Avoiding Dangerous Climate Change" workshop in Exeter.

A comment on the Stern Report from Paul Baer is also promised shortly. Meanwhile, RealClimate's comment on (the climate science in) Stern remains conspicuous by its absence...

18 comments:

Anonymous said...

The 0.1 per cent isn't the discount rate or discount factor, it's the inherent rate of time preference. The other parameter Nordhaus refers to is the intertemporal elasticity of substitution, which (along with the inherent rate of time preference) determines the discount rate. So it's a bit misleading to say "Nordhaus goes further".

JQ

James Annan said...

Well, it wouldn't surprise me unduly to find I'd been a bit misleading - some of this terminology is a bit obscure to me, but both Yohe and Nordhaus explicitly refer to 0.1% as the "discount rate" numerous times (they also use more long-winded forms, eg "pure rate of time preference that discounts logarithmic utility in per capita consumption"). So I'm not sure if you actually disagree with them on this, or (more likely) in what way I've misinterpreted what they wrote.

Brian Jackson said...

If you feel so strongly about it, perhaps you could offer to write a guest commentary about the Stern report? I'd be in interested in reading it at least...

EliRabett said...

It's called the social discount rate which is very different than the ones for bonds. Before going off on Stern you should look at the background papers that he commissioned.

EliRabett said...

Take a look at this paper by Arnell which is the base for Arnells report to Stern on the probability and costs of abrupt climate change. You really should look at things like this if you you want to figure out Stern

James Annan said...

Eli,

That paper seemed virtually content-free, a methodological description and little more. Feel free to make your point more clearly. As for the economics, I claim no expertise and am only pointing to what others have said (NB to Brian - hence I have nothing much to add to what I've already written on the climate science), and Tol in particular seems very sure of himself with his latest

"a supposedly eminent economist made a fool of himself in the public eye ... climate policy would have been in a better place without Nick Stern"

Even Gary Yohe, who has put his name to some of the implausibly alarmist climate science from Michael Schlesinger, struggles to excuse Stern's economics.

EliRabett said...

For those who have not read the position paper and its backup, Arnell and colleagues selected a panel of experts and asked then about the probability of rapid warming and disruption of the thermohaline circulation (not the Gulf Stream, that appears to be a result of the usual telephone game) Expert panel methods have a long and useful history in determining policy for such issues. The content, such as it is, is in the skill of the expert panel they selected. While most of the panel agreed with you (mostly by not participating I suspect), others did not.

As to whether you think Stern made a fool of himself or not, you can read John Quiggen on Crooked Timber or on his own blog and the discussions which follow. Quiggen does not appear to think so, neither does Nordhaus, although Nordhaus disagrees with Stern's choices he does say that the choices were within bounds, just that Nordhaus prefers others for stated reasons. Tol is pushing his own peanut up the hill.

My own summary would be given a choice, Stern made them in a single direction and sometimes the evidence was strained. Did he make a fool of himself, no, but it is in the interests of some to so claim.

John Quiggin said...

In the discussion me on my blog, Tol backed off a fair way, agreeing that my counterarguments were sound.

James Annan said...

A comment emailed in from Richard Tol:

Eli:

Nordhaus literally writes:
"I am reminded here of President Harry Truman’s complaint that his economists would always say, on the one hand this and on the other hand that. He wanted a one-handed economist. The Stern Review is a Prime Minister’s dream come true. It provides decisive and compelling answers instead of the dreaded conjectures, ontingencies, and qualifications.
However, a closer look reveals that there is indeed another hand to these answers."

In other words, Nordhaus thinks that Stern is Blair's poodle and failed to do his duty as an economist.

I agree, except that Stern is in Brown's pocket.

As to my own peanuts on the hill, I provided some 10 pages of reasons why the Stern Review is peculiar, biased, misleading, or downright wrong. I have yet to see my arguments refuted. Instead, I have seen others make the same points.

Richard Tol

EliRabett said...

Drat. Now I have to add Tol to Nordhaus, Yohe, Stern and theSternlings. All this and reports and papers to write...

However, having located Richard Tol's paper and started to read it allow me to reflect a simple doubt to his sunny assumption:

"It assumes that society will never get used to higher but stable temperatures, changed rainfall patterns, or higher sea levels. This is a rather dim view of human ingenuity. It contradicts what we know about technological progress, adaptation, and evolution."

Consider the fall of the Roman empire. It took about 700 years to recover from that sucker. Consider Mayan civilization. Progress is to be wished, but it is not guaranteed. Stern was somewhat elliptical but in the end clear on the point. Sometimes us pessimists are pleasantly surprised, often not.

BTW, I certainly have no desire to argue exponentiated poodles, but if Blair is Bush's poodle, and Bush Cheney's poodle, why is Blair's poodle taking a bite out of Cheney's poodle? Guess he must be Brown's poodle.

Anonymous said...

Eli:

Exactly my point. It took Europe 700 years to recover from the collapse of the Roman Empire. 700 years is a lot shorter than forever. You may argue that this is nit-picking, and it would be with a reasonable discount rate. With a low discount rate, it does matter.

Besides, climate change is nowhere near as monumental as the breakup of the Roman Empire.

Richard Tol

EliRabett said...

While as far as I am personally concerned 700 years is forever, I don't believe that even Stern's forecasts went out quite that far, so forever is a bit of a strawman. However, forever turned out to be forever for the Mayans and there is some evidence that the dissolution of their empire was an environmental disaster as was almost certainly the case for the Anastazi.

This is neither here nor there being meer academic point scoring. The point is that assuming progress is swallowing a rather large assumption.

Anonymous said...

Eli:

With a 0.1% discount rate, you need to go out quite far. Gary Yohe tried to reconstruct the numbers in the Stern Review. Although Stern used a time horizon of 2200, 50% of the estimated damage was beyond 2200, that is, in the transversality conditions.

Transversality is not a word you would read in the Guardian, but with 2.5-10% of GDP, this is not a mere technicality.

Richard Tol

EliRabett said...

Richard, as I said, that is neither here nor there. Among other things, since Stern accepts that sea level rise committed to by 2200 will take some hundreds of years to reach equilibrium values of meters, and since sea level rise will be one of the major drivers of costs (unless of course you want to blame those fools who built in Florida) you don't have to appeal to esoterica to come up with a large future price beyond 2200 contracted for by 2200 (or even 2100).

Stern argues that even a 0% social discount rate could be justified. Stern does not accept that progress is inevitable, you appear to do so. Because of that he places high value on avoiding risks characterized by small possibility but huge costs (see for example Quiggen's blog for more along this line, but then you and he have).

Anonymous said...

Eli:

I do not assume progress. Stern does. Or rather, Stern assumes progress to drive up emissions, and no progress when it comes to vulnerability to climate change. That is inconsistent.

Having your transversality conditions dominate your results is bad modelling.

You may think that internal consistency and quality modelling are mere academic points. You should have noted that Stern derives his authority from his academic credentials.

Richard Tol

EliRabett said...

Stern assumes progress until the costs of that progress appear in full. He then sees the possibility of an economic collapse (small, but not to be ignored) or a large decline.

I would assume that Stern derives most of his authority from having been chief economist of the European Bank for Reconstruction and the World Bank as well as head of the UK Government Economic Service. Me, I have no authority. What's your excuse?

Anonymous said...

Eli:

The only authority that should count is the authority of the argument, not of the one who makes it.

If you want to know who I am, and why I dare speak about the economics of climate change, google Richard Tol.

As to your other point, Stern is again inconsistent. He predicts a new Great Depression, but does not adjust his emission scenario. Did you actually read the Stern Review?

Not only is Stern inconsistent, he is also out of step with the literature. These feedbacks have been modelled by others. There is no excuse not to include them.

EliRabett said...

Yes, I know who Richard Tol is, I also know that his model was discussed in the Stern report and I read that part and I read his reply and I've read other stuff by him.

Na nu? RTFR

BTW Beta Blogger does not work on the first word verification, you have to have the first one rejected and then use the second. A right royal pain