The Chronicle of Higher Education weighs in with an interesting article about the fight to make room for economic thinking outside the neoclassical consensus. Reporter Peter Monaghan makes some typical mistakes of oversimplification, especially when he writes that the Austrian approach "disputes the neoclassical truism that economies tend toward equilibrium"–Ludwig Lachmann might agree, but not Israel Kirzner. In general, the Austrians don't deny the tendency toward equilibrium, though they would deny such equilibrium is ever reached, or that obsessive studying and modeling of that never-never-land should trump the analysis of human choices in the real world.
Reason contributing editor Deirdre McCloskey is prominently featured. Monaghan highlights three major critiques of mainstream neoclassical economics from McCloskey: misuse of statistical significance tests, imaginary modeling with no connection to the real world, and "the arrogance of social engineering."
Hayek scholar Bruce Caldwell in the article looks at the problem of stifling orthodoxy this way: "Everyone is trying to be a little MIT or a little Harvard, and look exactly the same because that's the way you get scientific prestige," says Bruce J. Caldwell, a historian of economic thought at the University of North Carolina at Greensboro. That approach, he points out, ignores basic economic theory about the benefits of diversification, specialization, and niche marketing.
It would be easy to respond to Caldwell that if there were a worthwhile market niche in defying the neoclassical orthodoxy, surely someone would be doing it. But that would be falling into the silliest types of intellectual blindness that a pop-neoclassical obsession with equilibrium and perfect information can cause, best exemplifed in the joke about the econ professor walking across the quad with a student, who declares, "Look, Professor, a $50 bill on the ground!"
"Nonsense, dear boy," he says, not looking down. "If there were, someone would have picked it up already."
The whole article is worth a look.