Choose Your Heterodoxy (Wonkish)

I’m pretty sure Roger Farmer is subtweeting me here, when he says

There are still a number of self-professed Keynesian bloggers out there who see the world through the lens of 1950s theory.

And it’s true! In fact, quite a lot of what I use is 1930s economic theory, via Hicks. And I should be deeply ashamed. I am, however, not the worst offender. After all, there are plenty of physicists who still use Newtonian dynamics, which means that they’re seeing the world through the lens of 17th-century theory. Fools!

OK, Farmer wants us to think in terms of models with

an infinite dimensional continuum of locally stable rational expectations equilibria

or maybe

a continuum of attracting points, each of which is an equilibrium.

But why, exactly? Saying that it’s “modern” is no answer; so, for a while, was real business cycle theory, which proved to be a huge wrong turn.

In part, I think, Farmer is trying to explain an empirical regularity he thinks he sees, but nobody else does — a complete absence of any tendency of the unemployment rate to come down when it’s historically high. I’m with John Cochrane here: you must be kidding.

But let me not try to figure out what Farmer wants, and instead ask what the rest of us should want.

Clearly, models with rational expectations, markets continuously in equilibrium, and unique equilibria don’t cut it. But which pieces of such models would you want to modify or replace? Farmer wants to preserve rational expectations and continuous equilibrium, while introducing multiple equilibria. That strikes me as a bizarre choice. Why not appeal to behavioral economics, behavioral finance in particular, to make sense of bubbles? Why not appeal to the clear evidence of price and wage stickiness — perhaps grounded in bounded rationality — to make sense of market disequilibrium?

The 1950s theory Farmer derides actually follows more or less that agenda, albeit informally. Wage stickiness is just assumed, but loosely justified in terms of psychology; New Keynesian models, with explicit modeling of price setting and menu costs, make this a bit less ad hoc but not much. Demand for goods and assets is based on plausible descriptions of behavior, with allowance for possible deviations from rational expectations. Obviously you want to go deeper than this if you can; but has this approach been proved useless as compared with more modern theory?

Surely the answer is a resounding no. As I’ve written many times, economists who knew their Hicks have actually done extremely well at predicting the effects of monetary and fiscal policy since the 2008 crisis, whereas those who sneered at this old-fashioned stuff have been wrong about almost everything.

I’m all for new ideas, indeed for radical heterodoxy, if it solves some problem. Attacking ideas that seem to work pretty well simply because they’ve been around for a while, not so much.