June 14, 2015

The art of start

Comment on Bruce Edmonds on ‘The Context-Dependency of Human (Economic) Behaviour’


“I think it is the lack of quite sharply defined concepts that the main difficulty lies, and not in any intrinsic difference between the fields of economics and other sciences.” (von Neumann, quoted in Mirowski, 2002, p. 146 fn. 49), see also (2013)

Therefore, it is of utmost importance to consequently stick to sharp definitions and to stay clear of ambiguity and vagueness.

ad (i) First you quote me correctly asserting ‘the task of theoretical economics is to explain how the monetary economy works’ and then you ask ‘does this exclude barter systems.’ The answer is evidently yes. Then you go on asking ‘does it exclude the sharing of value within families and communities?' Again, the answer is evidently yes. The focus is on the economy we happen to live in and not on issues that belong to the domains of psychology and sociology. Apart from this, to exclude some real-world phenomena at the start means only that they are to be included at a later stage. So nothing of importance is lost.

The economist's task is to incorporate valid results of PsySoc into his models but not to make ‘a fool of himself’ (Viner, 1963, p. 12) with utility maximization, rationality, bounded rationality, rational expectations, situational analysis and all the rest. Economists have no correct theory of human behavior, neither do they have a correct theory of how the monetary economy works, and it is for the latter defect that economics is a failed science. As long as Heterodoxy is preoccupied with PsySoc it will share the cheerless fate of Orthodoxy.

ad (ii) You say: “There is a way from the the understanding of individual human behavior to the understanding of the behavior of the economic system – it is called agent-based simulation.”

Generally speaking, a simulation is the proper tool of economic analysis and this has already been said in my paper on behavior (2015). However, a simulation presupposes the definition of the structural properties of the monetary economy. Hence, not any simulation will do. Here is the correct version.*

“A simulation as defined by the four structural axioms and the probability distributions is a well-defined mathematical object just like a system of equations. While they are formally on the same footing, both mathematical objects yield different kinds of outputs: the system of equations yields a solution vector, a simulation yields a bundle of paths. This bundle has a counterpart in reality.” (2015, p. 5).

In sum: every economic analysis must start with the definition of the objective structure of the monetary economy because it is this structure, a.k.a. reality, that determines the outcome of individual and collective human action. The monetary economy is the meta-context of every partial analysis. To start with a specific behavioral assumption means to get off at the wrong foot and to end up eventually in the scientific wood. Orthodox economics is a cautionary tale. Heterodoxy is expected to do much better.

Egmont Kakarot-Handtke

Kakarot-Handtke, E. (2013). Confused Confusers: How to Stop Thinking Like an Economist and Start Thinking Like a Scientist. SSRN Working Paper Series, 2207598: 1–16. URL
Kakarot-Handtke, E. (2015). Essentials of Constructive Heterodoxy: Behavior. SSRN Working Paper Series, 2600523: 1–17. URL
Mirowski, P. (2002). Machine Dreams. Cambridge: Cambridge University Press.
Viner, J. (1963). The Economist in History. American Economic Review, 53(2): pp. 1–22. URL

* For the full picture see also the cross-references