Wednesday , September 20 2017
Home / John Cochrane - Grumpy Economist / In the name of Science

In the name of Science

Source:"Climate Feedback" has produced a "scientific review" of my WSJ oped with David Henderson on (Oped ungated full text here, see also associated blog post.)In the blog post, I wrote,"If it is not clear enough, nothing in this piece takes a stand on climate science, either affirming or denying current climate forecasts. I will ...

John H. Cochrane considers the following as important: , ,

This could be interesting, too:

John H. Cochrane writes Duet

John H. Cochrane writes Tax Reform Again

John H. Cochrane writes On climate change 2

John H. Cochrane writes Yellen at Jackson Hole

"Climate Feedback" has produced a "scientific review" of my WSJ oped with David Henderson on (Oped ungated full text here, see also associated blog post.)

In the blog post, I wrote,
"If it is not clear enough, nothing in this piece takes a stand on climate science, either affirming or denying current climate forecasts. I will be interested to see how quickly we are painted as unscientific climate-deniers."
Now we know the answer. 

To recap, the oped said nothing about climate science, nothing about climate computer model forecasts, and did not even question the integrated model forecasts of economic damage. We did not deny either climate change nor did we argue against CO2 mitigation policies in principle. For argument's sake we granted a rather extreme forecast (level of GDP reduced by 10% forever) of economic costs. We did not even question the highly questionable cost-benefit analyses of policies subject to cost benefit analysis. We mostly complained about the lack of any cost benefit analysis, and the quantitative nonsense of many claims.

So, it's curious that there could be any "scientific" review of a purely economic article in the first place. How do they do it? 
Aaron Bernstein, Associate Director of the Center for Health and the Global Environment, Boston Children’s Hospital, Harvard: writes 
"Although many claims in this op-ed don’t mesh with reality, [no example stated] the most concerning delusion presented is that the health costs of climate change are both known and manageable. Legitimate economic analyses have put the costs of climate change at 2100 to GDP at several percent to more than 20%[1], with the variability largely due to different discount rates." 
We did not say known. We cited estimates, which have standard errors. We cited 10% of the level of GDP, forever. The response cites the discounted cost of all future GDP loss, in terms of one year's  GDP. Our number is much larger. 10% of GDP forever has a discounted value of 10%/(interest rate - growth rate). If interest rate - growth rate is one percentage point, then 10% of GDP forever is worth 10 times annual GDP, 1000% a lot more than 20%. If we took his number, total discounted costs only 20%, then climate change would truly be trivial. Even if he were answering our 10% with 20%, a factor of two is couch change in this business. OK, two tenths of a percentage point of growth.

(The quote is only about losses up to 2100, so you don't get the full r-g effect, but you see the point -- apples to oranges. The lesson is don't divide a present value by one year's flow. The discounted costs are an even larger fraction of a minute's GDP.)  

Bernstein  continues: 
"Even these higher damage estimates may fail to capture the full costs of extreme events over time, as Martin Weitzman’s work has shown. But there’s another, and more difficult, rub. What if we don’t understand the full consequences of greenhouse gas emissions? "
and continues with a standard list of things that might go wrong. We had written, 
"... some advocate that we buy some “insurance.” Sure, they argue, the projected economic cost seems small, but it could turn out to be a lot worse. "
and addressed the issue. 

"Science" and "scientific" review is supposed to include the ability to read and basic quantification. 

David Easterling, Chief of the Scientific Services Division, NOAA's National Climatic Data Center writes:
"This is a very simplistic, almost naive op-ed on climate change impacts." ...
It wasn't an oped on climate change impacts. It was an oped on cost-benefit analysis of policies to address climate change impacts, and never questioned any climate change impacts. 
"The idea that Miami is going to build a dike like Rotterdam is almost laughable. Of course climate change is not the only risk to society, but it is the biggest environmental risk. And most large buildings (e.g. Empire State Building) are not rebuilt every 50 years, only smaller, more expendable ones are."
Just why is building dikes, or other adaptations laughable? Miami is 7 feet above sea level, Rotterdam about the same below sea level, and 7 is greater than most estimates of sea level rise. Rotterdam did it. Climate change is the biggest environmental risk? More than nuclear war, chemical pollution, the crap in the water that most people in the world drink, malaria, loss of habitat, poaching, all put together? A citation or two comparing climate change to the others would be nice. And the total value of smaller more expendable buildings is far larger than the total value of Empire State buildings. 

Easterling falls neatly into our trap. We accused the politicized climate policy community for leaving quantitative, cost-benefit policy analysis behind and he... leaves quantitative cost benefit policy analysis behind.  

Frank Vöhringer, Dr. rer. pol, Scientist, Ecole Polytechnique Fédérale de Lausanne (EPFL), 
"The article plays down impacts of climate change that most studies consider to be highly important: e.g. the death toll of heat waves, hazards to coastlines, costs and friction of migration and other adaptation.... economic studies suggest that the risks of climate change are important, especially in certain economic segments (e.g. agriculture, health) and for low income countries with low capacity for adaptation. The article fails to mention that hazards and distributive issues of climate change increase all the other risks that the authors itemize, “nuclear explosions, a world war, global pandemics, crop failures and civil chaos”, even if it is not yet clear to what extent."
Verena Schoepf, Research Associate, The University of Western Australia, 
"The authors seem unaware of many consequences of climate change, particularly related to the ocean. The increase in ocean acidity and temperature, due to uptake of atmospheric CO2, will have tremendous consequences for many marine organisms and thus ultimately humans via sea level rise, impacts on weather and climate, food security, etc."
Wolfgang Cramer, Professor, Directeur de Recherche, Mediterranean Institute for Biodiversity and Ecology (IMBE) continues in the same vein.  

This is all simply untrue. We didn't "play down" any costs, and certainly not "economic studies," which we fully acknowledge. We do take for granted all the scientific, computer modeling and economic model estimates (though there is plenty to argue with there, but that's for another day). Nothing in the oped questions any of this. And "fails to mention" has to respect our limits: the WSJ gives us 900 words. We can't mention everything. 

Moreover, we acknowledge and consider
"Yes, the costs are not evenly spread. Some places will do better and some will do worse...."
We acknowledge and consider that
"Migration is costly. But much of the world’s population moved from farms to cities in the 20th century...."  
Not bad for 900 words.

Wolfgang Cramer, Professor, Directeur de Recherche, Mediterranean Institute for Biodiversity and Ecology (IMBE) continues, but I'm running out of steam. You get the idea.

Bottom line

Our main charge for the climate-policy community was, 
"Scientific, quantifiable or even vaguely plausible cause-and-effect thinking are missing from much advocacy for policies to reduce carbon emissions. " has nicely illustrated exactly such flights from scientific, quantifiable, or even vaguely plausible cause and effect thinking. Notice not one counterexample in my quotes or the whole post. Along with a striking inability to read, and a fascinating will to put words in people's mouths that aren't there.

Let me offer a little "scientific review" of this "scientific review." N=5 is a small data sample. There is this little concept called "selection bias." Offering highly interested people a chance to blast an oped is not a "scientific review."

Blogging, opedding, publishing your political opinions is what democracy and free speech are all about. Just don't call it "science." 

Like most people, I revere "science." Its dispassionate quest for the truth has brought us unimagined prosperity. But, dear climate policy "scientists," be careful,  if you are going to invoke the imprimatur of "science" you had darn well better be right. If you end up saying "never mind," as the food establishment has done with the 1970s advice to eat margarine and sugar instead of animal fats, the public prestige of science, and all the good for policy it has brought, will come crashing down. You will be treated no more seriously than economists. And that will be a great tragedy. The fact that you are using such unscientific method in your policy analysis is an early warning sign.

I wrote to the climatefeedback editor, requesting that they post a link to this response on their "review." It will be an interesting test of what ethics remain part of "science" to see if they do that, or answer my email.

Update: climatefeedback answers, in the true spirit of dispassionate transparency that "science" demands:

Hello John,
Thank you for reaching out. We could agree to add a link in our review acknowledging
 your reply; we only require that The Wall Street Journal adds a link to our review from your article.
Thank you,
Emmanuel Vincent

John H. Cochrane
In real life I'm a Senior Fellow of the Hoover Institution at Stanford. I was formerly a professor at the University of Chicago Booth School of Business. I'm also an adjunct scholar of the Cato Institute. I'm not really grumpy by the way!

Leave a Reply

Your email address will not be published. Required fields are marked *