This is a great piece, by the author of the exceptional [book: Globalization Paradox] (which I read years ago and never reviewed; shame on me). One of the basic themes underlying that book — and, I gather, underlying Rodrik’s [book: One Economics, Many Recipes] — is that reality is complicated, and that economics can give you diverse conclusions depending upon your assumptions. Different assumptions are appropriate for different contexts. Free trade doesn’t always, for instance, make the world a better place in the short run; there are winners and losers, and it’s not clear that the gains to the winners outweigh the losses to the losers (particularly if we attach ethical weight to a more-equal distribution of income). Only if you introduce the assumption that the winners can compensate the losers does this conclusion start to make sense.

There’s nothing wrong with assumptions; as the great John Tukey said, “Without assumptions there can be no conclusions.” If you’re going to argue that mathematics or science or economics has a poor track record in decision-making, the natural reply is, “Compared to what?” Compared to your gut, it’s not at all obvious that economics has done poorly.

Where the discipline does sin, according to Rodrik, is in telling a different story to the outside world than it does to its students. Graduate economics seminars, says Rodrik, very carefully tease out all the assumptions that make the conclusions true; what shows up in the newspapers does not (“free trade good; free markets good; industrial planning bad”).

Naturally, though, if you’re going to use economics for real-world decision-making, you need some way of testing which assumptions apply in a particular case. You need data. But doubly-blinded controlled experiments in economics are few and far between, if not outright impossible. So the discipline will always suffer from an abundance of models and a dearth of practical advice on how to use them.

Here I’m reminded of the very excellent “Deconstructing the argument for free trade”, particularly this bit:

> President Truman [allegedly] got so tired of hearing economists tell him “on the one hand…” that he wished for a one-armed economist. But frequently the best advice we can give is a menu of effects that flow from different choices. Trying to come up with a valid measure of the *net* effects is above our pay grade.

It’s a plea that economists show some humility. But since that seems to be in short supply, perhaps we need a belt-and-suspenders approach: the public, and policymakers, need to understand the limits of the discipline that they rely on.

(Rodrik link at the top via Cosma Shalizi’s Pinboard.)