Sunday, September 8, 2013

Attack of the Zombie Philosophers


The New York Times online recently ran a guest column called "What Is Economics Good For?"  It is framed as advice for the incoming chairman of the Federal Reserve from two philosophers of science. Their main advice: Don't think that economics is a science.

When we put a satellite in orbit around Mars, we have the scientific knowledge that guarantees accuracy and precision in the prediction of its orbit. Achieving a comparable level of certainty about the outcomes of an economy is far dicier.

That's because it's a more complicated problem, right? No:

The fact that the discipline of economics hasn't helped us improve our predictive abilities suggests it is still far from being a science, and may never be. Still, the misperceptions persist.... The trouble with economics is that it lacks the most important of science’s characteristics — a record of improvement in predictive range and accuracy. This is what makes economics a subject of special interest among philosophers of science. None of our models of science really fit economics at all.

This is just depressingly superficial. Apparently there's been essentially no progress in philosophy of science since I encountered it thirty-odd years ago. Here is the way a philosopher of science thinks: "Physics is science. Therefore, the more closely something resembles physics, the more scientific it is. Social science is not very much like physics, therefore it's not very scientific."

Might not the problem be, instead, with the philosophers and their "models of science"? It appears the philosophers haven't done much over the last three decades about understanding sciences other than physics, even natural sciences. Science, in their view, is all about prediction. Begone, Charles Darwin, you pseudo-scientist! Where are your predictions? Begone, paleontologists and seismologists! Come back when you've got some predictions. Meteorologists, you're improving, but you'll never be true scientists, because you'll literally never be able to know what the weather will be 30 days from now. You'll never have the accuracy and precision of physics.

The philosophers never say this, of course. They take it for granted that Darwin was a scientist, even though he doesn't fit the model. What the examples above suggest is that not all scientists make predictions; some spend their time understanding and explaining things. I'm reasonably satisfied that we understand earthquakes better than than we did a hundred years ago, even though seismologists have made no progress over that time in predicting them. And similarly for the origin of species.

Nor do I follow what the authors mean by "prediction", or, for that matter, "economics". Apparently what they mean is that economists haven't done very well in forecasting recessions-- not much better, in fact, than seismologists in forecasting earthquakes.

But there's more to economics than that; even just within macroeconomics, some theories have done very well at predicting what would happen after the recession began. And there's more to prediction than forecasting. Any time you formulate a  testable hypothesis, you're making a prediction. Chemistry had one stunning success in forecasting, when Mendeleev's periodic table predicted the characteristics of elements that hadn't been discovered yet. Since then, chemists have continued to do science, without, as far as I know, making forecasts.

Finally, one would expect that philosophers of science would have a reasonable acquaintance with the particular science they're talking about. But here's their advice to the new Fed chairman:

An effective chair ... will be one who understands that economics is not yet a science and may never be. At this point it is a craft, to be executed with wisdom, not algorithms... What made Ben S. Bernanke, the current chairman, successful was his willingness to use methods — like “quantitative easing,” buying bonds to lower long-term interest rates — that demanded a feeling for the economy, one that mere rational-expectations macroeconomics would have denied him.

Paul Krugman calls a foul:

Whoa! They apparently imagine that QE was an intuitive reaction by Bernanke, one that academic macroeconomics would never have suggested. Nothing could be further from the truth.

Philosophy of science, of course, doesn't claim to be scientific. But it does claim to be true. The philosophers need to get to work gathering some evidence for their theories, in places other than the comfortable precincts of physics. Karl Popper and Thomas Kuhn wrote about physics; what are you going to write about?


No comments:

Post a Comment