"Models and Predictions"
Will Wilkinson notes that “a solid background in the philosophy of science is especially useful when it comes to explaining why many economic theories fail to meet the basic standards of adequate science.” That certainly seems true to me.
To oversimplify a bit for the sake of polemic, a lot of economics work seems to put more emphasis on “doing work that superficially resembles physics and therefore counts as science-like” rather than on doing work that actually resembles scientific endeavor in the sense of leading to useful predictions or technologies or what have you. You get the sense that some practitioners of economics would pick up The Origin of Species and dismiss it as too narrative to count as real science. This guy’s just arguing from a bunch of anecdotes!
These thoughts seemed relevant to me as at Paul Krugman’s suggestion, I read James Morley’s critique (PDF) of Dynamic Stochastic General Equilibrium modeling as a way of explaining macroeconomic outcomes:
However, in terms of really predicting the crisis, the award obviously goes to theories of endogenous financial crises inspired by the ideas of Hyman Minsky. Formal evaluation of these more narrative approaches is hard and there may be an element of the “stopped-clock syndrome” at play. But it would be foolish to dis- miss such theories out of hand. In particular, a ludicrous notion sometimes expressed in the ivory towers of academia is that, for Minsky to be taken seriously, his ideas need to be put into a DSGE model. Instead, the converse is true. For DSGE models to be taken more seriously outside of academia, they need to explain and predict as well as Minsky.
In these terms, there’s good reason to think it would be better if ideas as good as Minsky’s could be put into a more formal framework since that would make them more useful. But the correct ordering is that you start with theories that are doing some good and try to make them formal to make them more useful. Starting with formalisms that would be useful if they were accurate, and then endlessly tweaking the inputs so that in retrospect they give you the right answer is in some ways an interesting exercise but there’s a reason policymakers don’t rely on this work and it’s not that the policymakers are stuck in the past.