Re: [OPE-L] Okishio Theorem - do anyone think it is relevant?

From: Anders Ekeland (anders.ekeland@ONLINE.NO)
Date: Tue Oct 30 2007 - 15:55:41 EDT


Hi all,

I am not asking for a vote here. But I have always been surprised
that people inspired by Marx think that the tendency for the profit
rate to fall is refuted by it. The premises of Okishio is clearly
different from Marx'. Marx has a dynamic temporal model, where sunk
cost - the M paid for the original machines - will only gradually be
written down, be replaced by new and cheaper machines, but for a
certain period of time the profit rate will fall - which will in its
turn lead to other things happening that will counteract this
tendency. And of course it is the monetary rate of profit we are
talking about - the only rate that capitalists care about.

Okishio and Roemer is not on that dynamic playing field at all - so
it is from rather general observations  on the relationship of
dynamic systems to static equilibria  clear that only by accident
would the Okishio theorem have any bearing - as I have said before.
One had to show that Okishio's static model incorporated the essence
of Marx' dynamic model.

That there always were very strong ideological forces wanted to see
Marx proved false - not empirically, but logically/ideologically is
quite clear - and a fact one has to bear in mind. I mean - the whole
GE thing of Debreu is a construction - totally divorced from reality
in an economic sense, but with an immense ideological importance. So
important - that more empirically sound "bourgeoisie" theories like
Hayek's (Austrian economics) and Schumpeterian (evolutionary) -
remains marginal - although gaining influence.

Given the basic law of Marxist economics: no static result must be
accepted as relevant before it is proven to be valid in a fairly
general dynamic model, should make one sceptic of the relevance of
the Okishio theorem from the start.

I have not studied the literature around this theorem in depth, and
it seems to me that in Kliman's book, the chapter on this puts
forward a quite correct critique using a "dynamics light" - or
"quasi-static" model in order to disprove the Okishio theorem by the
simplest model possible - or more precisely - to show that Okishio is
not relevant for the dynamic mechanisms that Marx' discusses.

I must have missed something in Jerry's line of argument, because to
me the fact that wages are zero in Kliman's examples is only a true
simplifying assumption. I fail so see that giving workers a constant
monetary wage would change anything regarding the relevance of Okishio.

The very limited purpose of the Kliman (and Freeman) exercise
regarding Okishio was recently very well formulated in the text
forwarded to the list. What we should do is to model the mechanisms
of the tendency for the profit rate to fall - and the counteracting
tendencies - and try to validate them empirically.

Just to mention one stylised fact: The profit rates achieved after
the enormous destruction of physical capital (and the switch from
elite to mass consumption of cars, electrical articles, chemical
products etc.) that we saw in the decades after the II WW - will be
hard to repeat. That profit rates will be temporarily restored when
the workers movement suffers defeats - f.ex. by the neo-liberal
offensive - point again to the cyclical nature of the accumulation
process which the tendency of the profit rate fall in an
*organic/endogenous* interplay with the counteracting tendencies creates.

It seems for me - again based on Kliman's book - that slowly the
irrelevance of the Okishio theorem is being accepted by people who
IMO earlier had not reflected critically enough on what the theorem
actually does say.

Just to repeat - I am interested in hearing the arguments for why
Okishio's static model can say anything definitive about Marx dynamic
model on the issue of the movements of the profit rate when there are
endogenous (competition driven) labour-saving technological change?
Or am I kicking in open doors?

Regards
Anders


This archive was generated by hypermail 2.1.5 : Fri Nov 02 2007 - 00:00:19 EDT