Monday, February 28, 2011

Post Keynesian Concept of Uncertainty goes Mainstream?

There is an interesting article in the Economist attacking the efficient market hypothesis, citing work by Roman Frydman and Michael Goldberg in Beyond Mechanical Markets: Asset Price Swings, Risk and the Role of the State (Princeton University Press, 2011), which relies, to some extent, on the concept of uncertainty in economic systems and the inability to calculate objective probabilities for important factors in the investment decision:
“A New Attempt to Explain Market Inefficiency,” Economist, February 24th, 2011.
This follows some good work by the New Keynesians G. A. Akerlof and R. J. Shiller in Animal Spirits: How Human Psychology drives the Economy, and Why it Matters for Global Capitalism (Princeton University Press, 2009) that clearly approaches the Post Keynesian concept of subjective expectations.

All this is good news: the mainstream is catching up with Post Keynesians!

Addendum: The Concept of Uncertainty in Other Schools

Uncertainty in the sense of not being able to assign objective probabilities to economic outcomes and events in the future is an idea fundamental to Post Keynesian economics.

Daniel Kuehn in a post called “More on Keynesian Uncertainty” (March 1, 2011) makes some comments on my post, and wonders why I connect “Keynesian uncertainty to Post-Keynesians specifically.”

It is perfectly true that the concept is held by other economists. The neoclassical Chicago school economist Frank Knight drew the distinction between risk and uncertainty. But have the neoclassicals ever adopted his ideas? They have not. The New Classicals and monetarists adopt the ergodic axiom and rational expectations, which are not compatible with the Keynesian/Knightian concept of uncertainty which is not measureable.

Radical uncertainty – uncertainty in the Knightian sense – was also stressed by Keynes and made a fundamental part of his macroeconomic theory in the The General Theory of Employment, Interest, and Money (1936).

G. L. S. Shackle (1903–1992) also adopted the idea from Keynes, and Shackle was generally classified as a hybrid Austrian/Post Keynesian (or someone who drew “Keynesian conclusions from Austrian premises”).

The Austrian radical subjectivists in the tradition of Ludwig Lachmann also use the idea of uncertainty in economic systems, and correctly draw from this the conclusion that expectations are subjective. But it is quite clear that Lachmann was influenced to a considerable extent by Keynes and Shackle in this, as well as Mises, and that Lachmann criticised Mises for failing to extend subjectivism to the realm of expectations (Gloria-Palermo 1999: 128).

Lachmann and the Austrian radical subjectivists supported the subjectivist nature of value, expectations, and knowledge, and Lachmann denied that free market systems tend to equilibrium or have coordinating processes, which follows logically from understanding the role of money, time and uncertainty in economic processes.

But then Lachmann’s position is rejected by other Austrians like Kirzner as “nihilism,” so there are clear limits to the extent to which Austrian economics really takes these ideas seriously.


Gloria-Palermo, S. 1999. The Evolution of Austrian Economics: From Menger to Lachmann, Routledge, London and New York.


  1. Wow - nice review of additional treatments of uncertainty.

    And I was even simply thinking within Keynesianism. I don't understand why post-Keynesians have this sense of ownership. I'm not sure what Keynesianism would be without (1.) liquidity preference theory of the interest rate, and (2.) effective demand. You can't have a solid understanding of liquidity preference without Keynesian uncertainty, and effective demand gets confusing without liquidity preference. So I'm not sure what Keynesianism in general would be without this.

  2. Thanks for your comments.

    Yes, all fair enough remarks.

    But still New Keynesianism seems flawed and not quite in line with the concept of uncertainty in Keynes. Most New Keynesians accept rational expectations. I dont know whether you do.

    Rational expectations is not compatible with the logical consequences of Knightian/Keynesian uncertainty:

    "The arguments of the New Keynesian school, which emphasize the significance of sticky prices and wages, are also inadequate from a Post Keynesian perspective, as they also draw on rational expectations. Fundamental uncertainty is seen etc ..."

    See Steven Pressman, A new guide to post Keynesian economics, p. 61.

  3. Hi -
    I'm reviewing a paper that's discussing uncertainty. He discusses Knight, and then mentions Keynes as something of an afterthought. That is a mistake in and of itself, I think - to minimize Keynes's role in the development of these ideas. But one thing in particular that he wrote was that Keynes derived his view of uncertainty from Knight.

    This struck me as blatantly wrong - I wasn't under the impression either of them derived their ideas from each other. But I wanted to confirm that there wasn't some circuituous way that people have traced Keynes to Knight. It seems doubly implausible since they both published the same year, but perhaps there was an older piece I'm not aware of.

    I think it's simply mistaken, but was curious if you'd heard anything to that effect.

  4. I thought Keynes' work on uncertainty was essentially original, and not just derived from Knight - though of course give Knight his due.