Schinckus’ … econophysics

Schinckus Economic uncertainty and econophysics Physica A, Vol 388, n° 20, pp.4414-423, 2009

Schinckus’ abstract makes the significant claim that:

“[Knightian uncertainty] is paradoxically excluded from the economic field. … By presenting econophysics as a Knightian method, and a complementary approach to a Hayekian framework, … econophysics can be methodologically justified from an economic point of view.”

 To me, the paper only justifies the much less significant:  

‘From an econophysicist’s perspective, economics seems to be about Gaussian distributions whereas Knightian uncertainty seems to be about unknown distributions, and in this sense econophysics seems to handle uncertainty better than economics. Moreover, econophysics seems to supply the analytic form for Hayek’s concept of emergence.’

 As I see it, economics is still largely about the issues that Adam Smith considered, rather than short-term trading, and it is the long-term issues where the uncertainties of Keynes, Knight and Hayek primarily arise. Thus while one might be critical of economists involved in finance, it is but a small part of the picture. Actually, I would be critical of economists on the whole in not getting across these uncertainties, but the paper does not address this issue.

Schinckus’ interpretation of Knightian uncertainty is that it is about unknown probability distributions. This is not my reading of Knight. On re-reading I see that the issue is not clear, and Knight’s examples do not meet the point, but Schinckus’ assertion that Knightian uncertainty is nothing but unknown probability distributions needs justification. I doubt it. Schinckus, in arguing against Aunt Sally economists quotes Keen:

“Imagine that you are very attracted to a particular individual and that you know this person has gone out with 20% of those who have asked him or her out in the past. Does this mean that you have a 20% chance of being lucky if you pop the question? Of course not. Each instance of attraction between two people is a unique event, and the past behavior of the object of your desires provides no guide as to how your advances will be received. How he or she will react cannot be reduced to some statistical predictions based on past apparent regularities. From our perspective, their reaction is truly uncertain and this uncertainty is at the root of much of the angst that romantic attraction generates” 

Schinckus supposes that there is actually a real probability, but that it is unknowable. But he provides no justification for his view. Could there be ‘true uncertainty’ in Keynes and Knight’s sense? Perhaps a romantic entanglement involves emergent phenomena, and is at least as complex and uncertain as particle physics? Schinckus says:

“Hayek … described higher-level social phenomena as emergent from, although reducible to, individual actions. This stance implicitly refers to the notion of emergence used in econophysics in which it appears as a macro-phenomenon arising from non-simple interactions between lower-level entities”

This is a much narrow notion of emergence than that due to Bergson or Whitehead, for example. Is it really the case that, as in thermodynamics, emergent phenomena in economics are reducible, as Schinckus supposes, to ‘individual actions? Whitehead’s student, Keynes, thinks that they might not be, as in some chemical reactions associated with life. Keynes thus has an extra type of uncertainty that Schinckus disregards, because he thinks that it is not real. Keynes’ view of romance is perhaps richer than Schinckus’. Were they made for each other? ‘Destined’ to go out? Might a romantic meeting not be as complex and uncertain as ‘the collapse of the wave function’ in quantum physics?

 Schinckus does make some interesting observations, of importance to economics, not just finance:

 “The origin of this reduction of uncertainty in economics comes directly from the works of von Neumann and Morgenstern [Game Theory] in which individuals appear to be choosing among alternatives with probabilistic outcomes to maximize the expected amount of some measure of value termed “utility”. … The definition of utility they created refers to how decisions should be made when only the probabilities of some events are known. This definition of utility excludes uncertainty from the economic field, and this exclusion leads to confusion between this concept and risk in economic models.”

To support this view, which I found surprising, Schinckus provides these selective quote:

“Probability has often been visualized as a subjective concept more or less in the nature of an estimation. Since we propose to use it in constructing an individual, numerical estimation of utility, the above view of probability would not serve our purpose. The simplest procedure is, therefore, to insist upon the alternative, perfectly well founded interpretation of probability as frequency in long runs.”

Then Schinckus notes:

“This probabilistic approach advocated by von Neumann and Morgenstern has become a theoretical reference point in economics, in which uncertainty is considered as a measurable future. This approach is a partial and incomplete representation of a uncertainty (which economists often forget), and moreover it is not necessarily appropriate for all situations. An example from Keen … [as above.]

According to Frankfurter and McGoun, all the empirical contradictions in economic theory are attributable to an error in translating the uncertainty phenomenon into a theoretical framework.”

This last point seems credible, and it may be that von Neumann & Morgenstern was interpreted in the way that is claimed, and that it did have that baleful influence. But my own reading of von Neumann is quite different. According to Wikipedia

“Another result [of von Neumann] was the nonexistence of a static equilibrium. An equilibrium can only exist in an expanding economy.”

Thus it would seem that von Neumann was well known for two results that are actually incompatible. In fact, von Neumann & Morgenstern only assume that economic events with numerical probabilities exist, not that all uncertainties are measurable, and go the point of the book (as Wikipedia says) is to show that even in the simplest imaginable economies, as soon as you have two or more actors with non-identical aims the need for strategies arises, hence instability and the inapplicability of concepts based on long-runs or learnt subjective probabilities. Thus, von Neumann and Morgenstern show how non-probabilistic non-stochastic true ‘Knightian’ uncertainty and Bergsonian non-Hayekian emergence arises.

Schinckus also notes: “Econophysicists often use the concept of entropy to characterize the idea of uncertainty in economics and finance”. Von Neumann was one of the first to make use of this concept in this domain. But entropy and probability only really make sense in equilibria. The approach of von Neumann and Morgenstern was much broader.

Schinckus is not explicit about how he views the systems of interest, but we have some hints:

“In finance, for example, several studies have focused on aspects of the stochastic process being analyzed i.e. the shape of the distribution of price changes.”

“Econophysicists often use the concept of entropy to characterize the idea of uncertainty in economics and finance”.

“Economic and financial systems then consist of a large numbers of components whose interactions generate macroscopic characteristics totally independent of microscopic details (individual behaviors).”

“Econophysicists often use the concept of entropy to characterize the idea of uncertainty in economics and finance”.

This approach is reminiscent of the thermodynamic approach, which von Neumann and Morgenstern critique. Schinckus points to no evidence that economies really are stochastic processes rather than being more generally like those that Keynes and Whitehead allow for.

Conclusion

It may be that, viewed as a social field, economics paid too little attention to the possibility of crashes. This may be the fault of economists. Or it may be that no-one was interested in funding such work. Financiers can often take a good rent from the up-sides while being insulated from down-sides, so occasional crashes are ‘under-weighted’. And, at the time, the political class didn’t seem very concerned either. In any event, it doesn’t seem to be uniquely the fault of economists, much less of the books on their shelves.

Schinckus seems to argue that economists lost their way due to a misunderstanding of the mathematics (e.g. von Neumann’s). If so, perhaps we need some independently funded mathematicians looking at these issues.

Schinckus advocates the use of econophysics. This seems reasonable for short-term trading, but something else would seem needed in the long run: perhaps the sophistication of econophysics combined with the insights of [good] economics, underpinned by appropriate mathematics. A bit like Keynes’ approach, but updated.

See Also

Bentes’ Econophysics: a new discipline’  covers some of the same ground and is more balanced. It acknowledges where economists have influenced Physics and is more focussed in its criticisms, e.g. of neoclassical economics rather than all of economics. Its characterisation of the approach of econophysics evokes that of Bergson, Whitehead, Keynes and Smuts, with their lack of ergodicity, but makes no reference to uncertainty. It supposes that the empirical framework of Physics is universal and hence applicable to economics, but without giving any supporting detail or argument. It does not consider, for example, the application of its approach to micro and macro issues will give rise to the same paradoxes as in Physics. The treatment of uncertainty would seem to be key, so Schinckus’ paper is very much to the point. Does Physics need to look broader, as Popper suggested?

More recent work by Schinckus is more like Keynes and Whitehead, but no so radical.

 Dave Marsay

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: