[OPE-L:5533] Re: Re: William of Ockam's Razor and Political Economy

From: Paul Cockshott (paul@cockshott.com)
Date: Thu May 10 2001 - 05:02:27 EDT

On Wed, 09 May 2001, you wrote:

> It is not at all clear to many why we should
> incorporate within a critique of capitalism a
> predictive model of [market] price behavior.

Without a causal and hence predictive theory
of price one can have no basis for the theory
of relative surplus value, to take just one
instance from Marx's work. I could equally
well have said no theory of differential ground

Without a theory of relative surplus value or
of differential rent, one is unable to analyse
the basis of the incomes and class interests
of the two main exploiting classes.

I would say that that would seriously undermine
ones analysis of capitalism.

> Of those mentioned above and in this thread,
> I see the following division (I'll use your
> terminology):
> o  political-economic theories should
>     incorporate a predictive theory of price
>     behavior:
>         -- the simple labour theory of value;
>         -- Sraffa's formulation of price of
>                 production theory.
> o political-economic theory does not require,
>    nor should it include, a predictive theory
>    of price behavior:
>          -- Value-Form theories;
>          --  The TSS theory of prices.
> (Note: I don't think that it is accurate to speak
> of a "TSS theory of prices". I think this rests
> on a mis-interpretation of the purpose and
> meaning of the TSSI. 

Well if the TSS people do not have a theory of
prices what is the point of their work on the
transformation problem, whose subject matter
is the theory of price?

How do they propose to critique the Okishio theorem
if they have no theory of price?

I think you misrepresent them.

>Relatedly, I don't think
> that Marx had a predictive theory of price
> either. Indeed, I think this was not one of his
> concerns. Of course, if I am wrong in this
> assessment of TSS, then John E or someone
> else should correct me.)

If it was not one of his concerns, his theoretical
efforts with respect to the transformation problem
are incomprehensible. He was attempting, as he
saw it to address what had been seen as one of
the major flaws of the Ricardian theory of price -
its apparent contradiction between the labour
theory of value and the assumption of equal 
returns to capital in different branches of production.

Marx in many places makes predictive statements
about prices of the form

IF the labour rewuired to spin cotton falls
THEN the price obtained for cotton by spinners will fall

Without an underpinning of predictive statements
like this the whole analysis of capitalist industrialisation
falls apart. 

> So ... I guess the question becomes: is it
> a merit for a theory of capitalism to have a
> predictive theory of price?


> An odd thought occurs to me: in their rejection
> of predictive models, Hegelian-Marxism (in one
> of its guises as VFT) and Althusserian-Marxism
> (through the doctrine of over-determination) are
> in agreement. Rare indeed!

As I would regard myself as strongly influenced
by Althusserian writings, I would dispute this.
Over-determination relates to the relationship between
social levels, and has nothing to do with the 
theory of price. It is not the same as overdetermination
of a series of price equations.

> So, I'm not convinced that it is a merit to have
> a predictive theory of price. BUT, if a theory
> _does_ have such a component, then I think
> it is fair to judge the theory in part by the
> accuracy of its predictions.  AND, since there
> is a claim that one theory has superior powers
> in terms of explaining empirical matters, then
> it is fair to put such a theory to the TEST by
> asking that theory TO PREDICT price behavior
> and then compare the ex ante predictions with
> the ex post prices.  This leads me to ask: what
> does your predictive theory of price tell you
> about what will happen in the next 6 months
> and year in the UK and the US economies?
> Isn't that a fair question?

It is a fair question, but to anwser it I would need
up to date figures on labour inputs for all sectors,
plus data on product flows between sectors.
This is in principle capturable, but the data-sets
are not currently collated.
What one has to work with are I/O tables produced
typically on a 5 yearly basis. This allows one to
do hypothetical studies of the effects of changes
in labour productivity in individual industries etc,
but I have never done these myself.

> In solidarity, Jerry
Paul Cockshott, University of Glasgow, Glasgow, Scotland
0141 330 3125  mobile:07946 476966

This archive was generated by hypermail 2b30 : Sat Jun 02 2001 - 00:00:06 EDT