## Hicks on neoclassical ‘uncertainty laundering’

from **Lars Syll**

To understand real world ‘non-routine’ decisions and unforeseeable changes in behaviour, ergodic probability distributions are of no avail. In a world full of genuine uncertainty — where real historical time rules the roost — the probabilities that ruled the past are not necessarily those that will rule the future.

When we cannot accept that the observations, along the time-series available to us, are independent … we have, in strict logic, no more than one observation, all of the separate items having to be taken together. For the analysis of that the probability calculus is useless; it does not apply … I am bold enough to conclude, from these considerations that the usefulness of ‘statistical’ or ‘stochastic’ methods in economics is a good deal less than is now conventionally supposed … We should always ask ourselves, before we apply them, whether they are appropriate to the problem in hand. Very often they are not … The probability calculus is no excuse for forgetfulness.

John Hicks,Causality in Economics, 1979:121

We can read Hicks’s Causality in Economics thanks to ANU repository:

https://openresearch-repository.anu.edu.au/bitstream/1885/114673/2/b12723149.pdf

This pdf is a good reproduction such that we can use the search function. The page Lars Syll cites is in the chapter VIII Probability and Judgement (pp.103-122).

However, let me add a comment on this chapter. I will then argue some related points on Syll’s post.

(1) John Hicks gives a short history of statistics. As a whole it is a good concise summery but he is making several errors in his chapter.

A conspicuous one is the account on two concepts of probability. There are two ideas with regards to the essential feature of probability: (1) the frequency theory and (2) the axiomatic theory. Hicks explains that the frequency theory “has become orthodox” and that “modern works on statistical mathematics take it as starting point.” He adds that “the chief proponents of the alternative approach have been Keynes, in his Treatise on Probability (1921) and Harold Jeffreys, in his Theory of Probability (1939).”

This digest of the history of probability is wrong. First, the frequency theory did not become main stream of probability theory. Second, it is not correct to characterize Keynes and Jeffreys as proponents of axiomatic theory. They were subjectivist in the sense that probability is a subjective estimate of judging person. They opposed to frequency interpretation only on the basis that probability is more subjective than a frequency that one can measure objectively. Their main focus was that the probability was a degree of confidence.

To characterize Keynes’ Treatise on Probability as axiomatic theory is questionable, because subjective probabilities are not assured that they are logically consistent. Keynes as well as Hicks are ambiguous on this point. We cannot apply axiomatic methods on probabilities considered as degrees of belief or confidence.

(2) Statistics is not a part of mathematics. It comprises mathematical reasoning but the core part of statistics is rather a kind of sciences that have specific fields of applications (do not confuse statistics with theory of probability). The peculiarity of statistics is that it has a large variety of applicable fields. It is the wonder of statistics but it means at the same time the roughness and ambiguity of statistics as a theory. Statistics rarely investigates what the underlying processes are. Universality of statistics is obtained by abandonment of pursuit of causal analyses. In this sense, it denies causality. It is usefully applicable only when such an investigation was impossible. In spite of this fact, many statisticians think erroneously that they have grasped a secret of nature including human society.

When we argue the plausibility of applying statistics, we should ask the non-mathematical part which is the core of statistics.

(3) Although Hicks’s chapter is a good piece of reflections on probability, randomness, repeatability, and uncertainty, it is not a good way of argument to cite some sentences as if some observations of a high authority are always relevant for the discussion.

What is more important is that Syll does not show (or even hint) how to think if we accept Syll’s argument. Is this a proper attitude of a methodologist? No hints other than criticism mean that his methodology is not sufficiently deep in restructuring economics. He shows the weakness of the mainstream economics but no hints for alternative economics. Isn’t it disastrous?

(4) We should rethink probability thinking itself. My opinion on this difficult question is to reemphasize the importance of routine behaviors. See my working paper: Microfoundantion of Evolutionary Economics

https://www.researchgate.net/publication/301766363_Microfoundations_of_Evolutionary_Economics

Oct. 0, 2017

Mr.. Shiozawa,

What on earth do you mean by:

“What is more important is that Syll does not show {or even hint} how to think if we accept Syll’s argument…….Isn’t it disastrous ?”. Not to mention, “Statistics is not a part of Mathematics….Do not confuse statistics with theory of probability”

Moreover, Mr. Shiozawa, There’s lots of places you can investigate to get more than “hints”

about alternative paradigms.

Have you ever heard of Ludwig Von Mises’s statement {With help from his brother Richard, who was a very capable Mathematician of Statistics & probability, among other achievements} :

“Case probability is not class probability”.

M. Sylls exposition is quite a clear explanation of “genuine uncertainty” after all.

He, of all people, cannot be accused of the “ambiguity” you assume exists in so many other thinkers on the subject.

GOOGLE: {1}Norman L. Roth {2} Norman L. Roth, Economist

Dear Norman L. Roth,

My answer to your question is simple. To emphasize “genuine uncertainty” produces no theory in economics. If Lars Syll knows any hopeful direction, he should hint it. Can you point any concrete one? I am not talking of hints about alternative paradigms in general.

By the way, Richard von Mises is famous of his frequency approach in probability theory. His approach failed as mathematics and was replaced by Kolmogorov’s axiomatic approach. Hicks is talking against frequency theory. Richard von Mises is not a right person to support “genuine uncertainty.”