Monday, May 3, 2010

Boettke, Stiglitz, Hayek, and Socialism

Hypothetically, I always thought it would be great to host a dinner party for Joe Stiglitz and Friedrich Hayek and just sit there absorbing their discussion. And I call it a "discussion," rather than a "debate" intentionally - because I think they would both have a lot of very constructive things to talk about, despite their stark differences. Specifically, I've always been intrigued by both of their takes on information and knowledge. Hayek's theory of knowledge emphasizes that knowledge is decentralized, dooming any attempt to centrally solve problems of allocation and production. The price mechanism is a knowledge discovery process that leverages all of society's dispersed, tacit knowledge. Stiglitz makes these points as well, and of course adds a robust literature on information asymmetries and information imperfections. Honestly, the old "socialist calculation problem" never deeply interested me. It always seemed like a rather obvious point, and I think the fact that I was so young when Communism fell makes it less important of a question for me. So in my Hayek-Stiglitz dinner party fantasies, I've always been more curious about how Hayek would discuss information asymmetry problems with Stiglitz, because as far as I know (and that's not very far), those sorts of issues didn't come up as much in his work.

The impetus for this post was an interesting post this morning by Peter Boettke. He writes on what he calls "robust political economy", as well as Stiglitz's Wicksell lectures, titled "Whither Socialism?". This passage in his review is key:

"Whereas Hayek argues that the standard model underestimates the informational role of disequilibrium prices, Stiglitz argues that the standard model overestimates the functional significance of equilibrium prices in ensuring Pareto efficient resource allocation."

My feeling here is the feeling that I often express with respect to Austrian Business Cycle Theories: that each of these narratives is "necessary but not sufficient". I'm of the opinion that both Hayek and Stiglitz are right. Hayek's point is that socialist calculation fails because the planner lacks the knowledge that is coordinated by prices to allocate efficiently. When prices are in disequilibrium, individuals acting on their decentralized knowledge respond in the market, bringing prices back into equilibrium. This fundamental corrective isn't available to the socialist planner. The socialist planner who doesn't realize this therefore underestimates the informational role of prices. But nothing in this basic point (certainly not original to Hayek or Mises) contradicts Stiglitz's work on information imperfections. Hayek's observation of what planners are lacking is not proof of what individuals in the market place do not lack. Stiglitz's point is that individuals in the market often lack information that they need to operate efficiently in the marketplace. Asymmetries in individual information leads to strategic behavior that distorts canonical market efficiency outcomes. There is nothing in these two stories that contradicts the other. I see no reason why Hayek's critique of socialism and Stiglitz's critique of market optimization can't coexist.

This is another Boettke post on Stiglitz.

Another Austrian economist that has had an interest in Stiglitz is Jonathan Finegold Catalán at Economic Thought. You can find a list of his posts on Stiglitz here.


  1. An information asymmetry is a just another "datum" to be captured by prices.

  2. Well sure, but let's push a little harder than that Lee Kelly. No one is challenging the idea that it will be captured by prices. The concern is that it might threaten Pareto efficiency.

  3. Years later... But it's not like Pareto efficiency is the corner stone of a good society...

    Moreover, Hayek's theory of prices as information requires that the information as well as the interpretation are always correct, which is a dubious proposition at best.


All anonymous comments will be deleted. Consistent pseudonyms are fine.