I’ve tried [to read Farmer's stuff], a couple of times, but found it very hard to penetrate and gave up — and several other economists I’ve talked to had the same reaction.
Krugman is on to something here. Though not always a paragon of transparency myself, I have at times found Farmer's work confusing. But there can be a nugget in there that is worth taking the trouble to dig for. Here's a tip for Krugman: Ignore the words that Roger writes, and just try to sort through the model. Typically, he's not doing anything that's technically difficult for the average economist with a PhD - the models typically have standard features.
A couple of years ago, I was confused by this paper by Roger. But, I think I figured it out, and the idea is fairly straightforward, and interesting enough that I wrote this paper, to explain Roger's idea and extend it. Krugman says:
Sorry, but I won’t commit to sitting through your two-hour movie if you can’t show me an interesting three-minute trailer.So, here's the "trailer" for Roger's idea (different from the subject of his complaint - that's another idea):
Workers and firms are in the business of bargaining over the surplus from exchange. But how they split that surplus is indeterminate - we don't have good theories of bargaining power. What this can lead to is indeterminacy in real wages, aggregate economic activity, and the unemployment rate. There are multiple equilibria. If the unemployment rate is high, the expected payoff from searching for work tends to be low, and if the real wage is high, the expected payoff to searching for work tends to be high. In the model, equilibria with high unemployment rates and high real wages coexist with equilibria with low unemployment rates and low wages, and across these equilibria, the expected payoff to searching for work is equal to the best alternative. This encapsulates the basic Keynesian idea - private economic agents find it difficult to agree on socially beneficial terms of exchange - but in the model there is nothing left on the table. Everyone is optimizing in equilibrium, and no prices or wages are fixed. The government can fix things - this is an essential element of Keynesianism - but they do it in non-Keynesian ways. For example, Taylor rules don't work.
Though Krugman has a point, we can get carried away with sound-bite economics. As with good music, sometimes you have to live with it for a while before you get it. Then, there's no turning back. Examples:
1. Expectations and the Neutrality of Money was part of a revolution, but almost nobody got it when it was first written. And the people who got it initially were not waiting for the trailer to convince them. They were moving the research forward, and getting credit for it. Ideas progress quickly, and young economists who want to make a splash are not waiting for a three-minute spiel to convince them what the big new ideas are.
2. The economics job market: It's well-known that lazy recruiting strategies don't work. Every January, 10,000 economists converge on some city in the United States, and various academic and non-academic institutions conduct 30-minute interviews of newly-minted (or about-to-be-newly-minted) economics PhDs. Recruiters can show up for interviews without preparation, and screen candidates based solely on the 30-minute interview. That guarantees that the people who are hired are the ones who can give a 30-minute interview. Of course it's no guarantee that the candidates can actually do good research and teach effectively. The conclusion of this paper seems to be that: (i) the research payoff from the median economics PhD is pathetic; (ii)quality falls off quickly in the top-ranked schools: the top graduate from the University of Toronto is roughly as good as the #3 from Yale. Everyone knows that it is worthwhile to spend time reading job market papers - carefully. A job candidate can be an inarticulate nerd, but have the power to create beautiful research that will pay off big-time in the future.