June 2, 2016

Question Models' Authority

Feel free to read the headline in your best Eric Cartman voice.

Megan McArdle compares climate models to econometrics models. Very similar -- except one has been abandoned and one is accepted as gospel, capital-T truth.

They could make some pretty good guesses from that data, but when you built a model based on those guesses, it didn't work. So economists tweaked the models, and they still didn't work. More tweaking, more not working.

Eventually it became clear that there was no way to make them work given the current state of knowledge. In some sense the "data" being modeled was not pure economic data, but rather the opinions of the tweaking economists about what was going to happen in the future. It was more efficient just to ask them what they thought was going to happen. People still use models, of course, but only the unflappable true believers place great weight on their predictive ability.

This lesson from economics is essentially what the "lukewarmists" bring to discussions about climate change. They concede that all else equal, more carbon dioxide will cause the climate to warm. But, they say that warming is likely to be mild unless you use a model which assumes large positive feedback effects. Because climate scientists, like the macroeconomists, can't run experiments where they test one variable at a time, predictions of feedback effects involve a lot of theory and guesswork. I do not denigrate theory and guesswork; they are a vital part of advancing the sum of human knowledge. But when you're relying on theory and guesswork, you always want to leave plenty of room for the possibility that your model's output is (how shall I put this?) ... wrong.


McArdle links to a nine part series on Coyote Blog and compliments its seriousness and measured tone. I would say teh same for her piece.

Deleterious Anthropogenic Warming of the Globe Posted by John Kranz at June 2, 2016 12:44 PM

How did economists know that the econometric models "didn't work?"

And then the whole effort was basically abandoned, because the models failed to outperform mindless trend extrapolation -- or as Kevin Hassett once put it, "a ruler and a pencil."

The model predictions were able to be compared side-by-side to other predictions, and to observed reality. When that is attempted with climate models, and the model predictions are found lacking, we're told that the timescale isn't long enough. "We must wait longer to see the true predicted effect." But you can bet your Aunt Petunia that the timescale would be "long enough" if there were any agreement whatsoever between the models and the actual climate.

Posted by: johngalt at June 2, 2016 2:50 PM | What do you think? [1]