[OPE-L:2135] Re: Science and Ideology

Paul Cockshott (wpc@cs.strath.ac.uk)
Fri, 10 May 1996 04:51:08 -0700

[ show plain text ]

>This is entirely the wrong question to ask, and so long as one asks this
>one remains in the terrain of scholastic ideology. The scientific question
>is whether the theory is a correct portrayal of reality.
>(1)Please explain why this is the wrong question to ask. Since this
>is basic, it needs clarity. In particular please define scholastic ideology.
The purpose of this group is not exegesis of Marx, but work on an
outline of political economy. By the term scholastic ideology I am compressing
two insults into one phrase:

by 'scholastic' I allude to the schoolmen and the prescientific practice
of argument from authority,

by 'ideology' I imply that political requirements dictate what answers must
be arrived at before the investigation begins - marx must be proved consistent
in his handling of the transformation problem vis a vis critics
even if we have to repudiate much of his writings to prove it

>(2)Please also define 'portrays reality'.

In this context, I mean a theory of prices and profits in a capitalist 
economy that can be put to statistical tests against real data from 
such economies, and come up with good predictions.

>(3)You seem to be defining a criterion for what is a *correct* theory. >But this isn't what is at issue. The question is whether the TSS view >is a theory at all. As far as I can make out, you deny this. This makes >for an extremely unequal (and probably heated) debate. > Paul ---- I dont deny that it is a theory, but that to defend it, you should be able to do so without reference to Marx. It could be defensible on the grounds of the realism of its assumptions, on the grounds that it is logically coherent, that when simulation models based on it were constructed their behaviour was informative, that it gave good empirical predictions etc. But what Marx said can not be evidence for or against the theory.

As a theory, I have found that, as one would expect, your version of prices of production theory gives more accurate predictions than the orthodox Sraffian theory. I say, as one would expect, since it is actually a weaker theory - closer to surface appearances than either the labour theory of value or other prices of production theories. To arrive at its predictions one essentially sets the prices of outputs to what they would have been had all industries returned the same rate of profit. Since this is not in fact the case, the prices so predicted inevitably deviate from actual prices, but, deviate less than they would if one also transformed input prices.

>(3)The one claim *nobody* makes for v=va+l is that it portrays reality. > Paul ---- This is not the case. I contend, following Farjoun and Machover, that this is actually a remarkably accurate portrayal of real prices.

>If it portrayed reality, there would be no need for transformation, no >need to worry about what happens when goods do not sell at values, no need >for most of the debate of the last century, and probably no need for OPE-L. > Paul ---- It is true that from the standpoint of the stochastic justification for the labour theory of value derived by F and M, one can see that the debate of the last century on the problem of transformation has been entirely misplaced, based as it has been upon the erroneous assumption of the formation of an equal rate of profit.

As to whether this abolishes the need for OPE, no, because political economy is much wider than debates on the transformation problem. We have the world market, finance, state revenues, the dynamics of accumulation etc to consider.

>Indeed there would have been little need for Marx. Proudhon would have >said most of what was required with a little refinement from Ricardo. > Paul ---- On value theory, I think Marx added little to Ricardo, but on the analysis of exploitation alone, for example, he opened up a whole new field.

>You and workers such as Ochoa, Petrovic and Chilcote, working in the >tradition established by Anu, have suggested that vertically-integrated >labour embodied magnitudes (henceforth VILE magnitudes) are a good empirical >predictor of *aggregate* prices of the sectors defined by the classifications >adopted by I/O statisticians. > >This tells us nothing about *unit* prices, which is what the debate >about values and prices is all about. I think it is an important finding >that VILE magnitudes are good predictors of disaggregated total output. > Paul ---- What do you mean by unit prices at this point. Are you meaning the disaggregation of steel prices into prices for different grades of steel?

>But the fact that the total output of steel in money correlates with the >total VILE in this output, in now way proves that the price per ton of >steel correlates with the VILE per ton.

Paul ---- One can only obtain correlations between two ordered sets of figures, so that where one only has two figures - price per ton of steel, labour content per ton, then there is of course no question of correlation. If on the other hand, one performed time series analysis then one obtains figures for price and labour content at different dates. After making allowances for inflation one can then attempt a correlation. Allin and I have not done this, but other workers in the field have, and their results are support the hypothesis that changes in money prices are largely explained by changes in labour content.

>In order for VILE per unit to correlate with price per unit, steel >would actually have to sell for its VILE. Society would have to actually >work like Engels' 'simple commodity production' with profits directly >proportional to hours worked in every sector, etc. > >And it doesn't. > >Otherwise we wouldn't be having this discussion. > Paul ---- No, this is a misapprehension. One must distinguish various sources of deviation of prices from values. One is common to all modes of commodity production, and consists of random uncorrelated deviations produced by imbalances of supply and demand. This would apply even under simple commodity production. A second source of deviations arises due to rents, which cause prices to be proportional to marginal rather than mean labour contents for some commodities.

Both of these sorts of deviations are uncontroversial.

The controversial one is that which is supposed to arise due to a tendancy for profit rates to equalise. My contention is that this source of deviations is in practice less important than it has been credited with. The tendancy of profit rates to equalise is relatively weak, and in consequence, as a source of deviations of prices from values this specifically capitalist form of transformation is less marked than the others. There appears to be no significant increase in the predictive power of the theory when this form of transformation is taken into account. Paul Cockshott

wpc@cs.strath.ac.uk http://www.cs.strath.ac.uk/CS/Biog/wpc/index.html