September 18, 2015

Predictably confused

Comment on Lars Syll on ‘Sir David Hendry on the inadequacies of DSGE models’


“The future is unpredictable.” (Feynman, 1992, p. 147) Four words! Compare this to what economists have uttered about this issue without getting one iota further. But wait, are physicists not famous for their accurate predictions? Could it be that economists have gotten something badly wrong?

Let us have a closer look. If you show a physicist an apple tree and ask him when every single apple will fall, then he will tell you that this kind of event is not predictable. But he can tell you something else. If an apple has started to fall then he can tell you exactly its location and velocity after t seconds. The physicist ‘predicts’ the coordinates with high precision and everyone can test it. (OK, you did not want to know this to benign with, and exactly these layman's expectations regularly cause the irritations with science.)

Likewise: imagine somebody throws blindly three coins into a large sandbox. Clearly, the three coins form a random triangle and no one can predict its form and size. Yet, the mathematician can ‘predict’ with certainty that the sum of angles is 180 degrees (if the sandbox is euclidean).

While the future is ‘unpredictable’ certain aspects may be ‘predictable’ with high precision. Therefore, we can agree with Keynes that “the price of copper and the rate of interest twenty years hence” is unpredictable without accepting his famous all-round capitulation “We simply do not know.” (Keynes, 1937, p. 214)

Example: an elementary consumption economy can be described by this deductively derived formula. This formula holds in every single period from past to future (2014, eq. (12)). In other words, we have a testable economic law. Test it twenty years hence and you will find out that it is true. Where, then, does the difficulty with prediction come in? The crucial point is that the variables that underly the four rhos are unpredictable random variables.

Perhaps it sounds a bit paradoxical: there are deterministic economic laws which hold ‘on the whole’ while the components vary at random or are even uncertain in Keynes's sense. So the concept of uncertainty can coexists with the concept of economic law. Hence, Keynesian uncertainty should not stop us from looking out for deterministic and testable economic laws. Attention, they are with absolute certainty not to be found where misguided Orthodoxy has looked for in the past!

The otherwise redundant DSGE debate shows that the representative economist is still a bit confused about the different aspects of prediction.

Egmont Kakarot-Handtke

Feynman, R. P. (1992). The Character of Physical Law. London: Penguin.
Kakarot-Handtke, E. (2014). The Synthesis of Economic Law, Evolution, and History. SSRN Working Paper Series, 2500696: 1–22. URL
Keynes, J. M. (1937). The General Theory of Employment. Quarterly Journal of Economics, 51(2): 209–223. URL