[OPE-L:3717] Re: Two "causes" of deviation?

Paul Cockshott (wpc@cs.strath.ac.uk)
Wed, 27 Nov 1996 02:56:48 -0800 (PST)

[ show plain text ]

>1. Capitalist competition is a lot more complex than what is captured by the
>perfect competition-pure monopoly axis, and the orientation about equilibrium.

I conceed this.

>2. What bothers me about the dualist position, in this context, is the causal
>and/or intentional links presumed to be operating between your price system and
>your labour value system.

The position that I defend on value theory is broadly speaking the formalisation
given by Farjoun and Machover in their book The Laws of Chaos. In this version
prices are presented as random variables with a kurtotic probability density
function such that the probability of a selling price being below the cost
price is very low, but with a long tail in the upward direction.

It can be shown by statistical arguments of a very general nature, that given
such a PDF there will be a close correlation between prices and vertically
embodied labour contents. The question then arises as to whether the
correlation between prices and labour contents will be better than that
between prices and 'production prices' understood as the price system that
would prevail under assumptions of an equal rate of profit. It turns out that
this depends upon the relative coefficients of variation of the profit/wage
and profit/capital ratio.

Farjoun and Machover predict that the coefficient of variation of profit/wage
ratio will be small. Allin and I have published data for the UK economy which
tends to confirm F and Ms predictions. This was in the Bergamo conference on
the 100th anniversary of Capital III and called 'Does Marx need to Transform'.
Under conditions of low variation in the
profit/wage ratio, one obtains close price to value correlations.

Our view is that the value vector constitutes an attractor for the price vector.
Our empirical work indicates that the price of production vector also consitutes
an attractor for the price vector, and that actual price vectors 'orbit' around
both these attractors.

I can understand that if you are looking for causal/intentional links, the
Farjoun Machover presentation of value theory will be unsatisfying, since it
presents no such links. This is one reason why it constitutes a significant
paradigm shift in the mode of explanation of economic phenomena. It completely
dispenses with methodological individualism and attempts to draw its
conclusions by resting on oly the weakest of assumptions.

>3. It is not the 'model' (or whatever they are being called these days) that
>bothers me but its interpretation in order that it can be claimed to be
>asserting something about real capitalist economies.

I consider that one of the great advantages of the stochastic approach is
that it makes clear testable predictions about what is going on in real
economies, and as such lends itself to emprirical test. We are very concerned
about what is happening in real economies.

>4. In these terms, intra-industry price equalisation and inter-industry profit
>equalisation are abstract tendencies, the real, let alone the empirical,
>manifestation of which we may expect to be complex and changing.


>5. I have yet to see a persuasive account of how the complex dynamics of
>competition in the face of incentives and constraints expressed in monetary
>terms, and orientated towards equilibria on only the most transient definition
>can usefully be interpreted as imposing the kinds of trends in terms of the
>labour value which are sufficiently stable and cumulative to manifest the kind
>of predictions of instability (as the basis for an argument of
>that you wish to make.

Alexandro's argument was couched in the language of volume III which is
still deterministic rather than stochastic. In order to meet his concerns
at this level, I was conceeding this model. The conventions of such models
require one to have equal rates of profit and equal prices, which, I agree,
would not occur in the real world. I would say, that once one moves to
a stochastic interpretation the whole formalism adopted in the debates over
prices of production become irrelevant.

My concern was to show that prices are a misleading indicator of values, this
is obvious as soon as one adopts a stochastic approach, but slightly harder
to see if one retains the problematic of prices of production. I was
trying to demonstrate that even within that problematic prices of production
can be misleading in that they can mask wastes of social labour which,
subject to any noise in the system will tend to be revealed.

The assumption that I made to show the instability was only that a lower cost
producer will tend to gain market share versus a higher cost one. This is a
pretty weak assumption. If one refuses this assumption, there is very little
of classical political economy or of Marx's Capital that would stand.
In particular in this context, the mechanism for the establishment of prices
of production relies upon this assumption. In short, the argument is that
if a price is above the price of production, new entrants into the market
will be able to obtain capital sufficiently cheaply to come in and gain
market share by undercutting existing producers. This has the effect of
lowering prices until they reach the price of production. I would agree
that such mechanisms are weak, but they are what the whole theoretical
framework of the price of production theory rests on.

One can also turn your objection on its head and ask what assumptions one
has to make in order that the type of system presented by Alexandro
in Table A would actually be stable, implying thereby that stability
is what requires a special explanation rather than instablity.

Paul Cockshott