Tuesday, October 27, 2009

Why do economists always weigh in on climate change?

The following seems to articulate a lot of the feeling people have about economists taking part in the climate debate, at least when they're skeptical:

Economists of all stripes should strive to acquire some humility nowadays , given the disastrous failure to recognize an impending economic disaster.

From a comment on an environmental economics blog.

Instead of discounting the value of economists in the global warming debate, I think that this actually makes economists some of climate science's most valuable critics. Macroeconomics is actually much more similar to climate science in methodology than, say, particle physics or many of the other hard sciences. Both disciplines study extremely complicated systems that are the net result of millions of interactions, not all (not even many) of which are directly observable. Both use statistical modeling and tremendous amounts of data to develop models and make forecasts and neither are particularly suitable for controlled, laboratory based experiments.

Economics, however, has the benefit of a much, much shorter time horizon. In other words, "long term" for economics is a much shorter time period than "long term" in climate science. In fact, economists have the benefit of readily accessible, detailed, accurate data for at least the past century. They have had the opportunity to watch long term cycles play themselves out many times in many different environments. They've had the economic equivalent of watching several different ice ages and the effects of the industrial revolution many times over, all with the equivalent of satellite data gathering capabilities.

And yet, as my econometrics professor taught us, "You're doing well if you have the sign right." Economists know they can't predict the future with great precision, they try but there is just too much uncertainty about even medium term predictions, much less long term predictions.

Comparatively, climate scientists have a tremendous amount of excellent data for the previous 50 or 60 years, and very good data for the past couple centuries. Beyond that, they have data, but it's far less precise and covers a much, much smaller percentage of the world. At the same time, climactic trends much longer than economic trends, and our really good data only covers one environment: A heavily industrialized world.

So, to me, from an economics background, climate models sound like they've got data from the past week and they're using that to forecast out for the next month to two decimal precision.

I think climate scientists need to take a cue from economists' predictive failures and not over emphasize precision. I really doubt they'll be more accurate than economists have been, though I expect they have the trend in the right direction.

No comments:

Post a Comment