Re: [OPE-L] price of production/supply price/value

From: Ian Wright (wrighti@ACM.ORG)
Date: Sun Jan 29 2006 - 13:00:03 EST

Hi Rakesh (and Fred -- glad you joined in)

Rakesh, I am interested in your original suggestion that prices of
production are proportional to labour values. Not many theorists do
this. I can at the moment only think of two: Farjoun & Machover, who
reject the theoretical concept of prices of production, and argue that
in a probabilistic framework that prices are proportional to labour
values; and Krause, who if I recall correctly, maintains that there is
a kind of proportionality of prices and values, but only if the
assumption of homogeneous labour is dropped (i.e. that the real-cost
of all concrete labour kinds is identical). If prices of production
are proportional to labour values then the transformation problem is
solved, in the sense that all Marx's conservation claims trivially
follow. F&M take the attitude that the TP is "dissolved" in a
probabilistic framework and revert to Marx's Vol I. theory of prices;
Krause basically says the TP is a mis-specified problem. But neither
of these solutions to the TP would be acceptable to Marx (the text)
because he resolutely maintained divergence of prices of production
and labour values.

Divergence of PoP and value is a major premiss of the TP, both
classical (pre-Bortkiewicz) and modern (post-Bortkiewicz). This
property is very much preserved in the Sraffian formalisations of the
problem (except Krause).

> Marx however never did himself say that he had left his transformation from
> values to prices incomplete by failing to extend the transformation to the
> inputs (see Moseley [2000] and Ramos M. [1998-9])).  There has been much
> controversy over the nature of Marx's admission of failure in Capital,
> volume 3, p. 265. I argue that the problem to which Marx was calling
> attention was the exact opposite of what it has commonly  understood to be
> since von Bortkiewicz wrote his famous criticism one hundred years ago. In
> my reading, Marx admits that he had wrongly assumed in writing the tables
> that he could infer the value transferred from the means of production from
> the machine's flow price as recorded in the cost price. In other words, Marx
> recognized that he had failed to transform the prices of production of  the
> input means of production backward from prices of production to  values in
> the determination of the values of outputs.  Once the value-price
> proportionality assumption is dropped in the third volume, the assumption
> embedded in the input value columns were no longer tenable, for those values
> were derived from the prices to which they were falsely thought to be
> proportional.

Yes, that's pretty much how I see it, once we do the Bortkiewicz thing
and abstract from the process of the formation of the general rate of

> Moreover,  let us stipulate that Marx had assumed that each
> wage good has a labor value of one and allows for the employ of one worker.
> But after Marx carries out the transformation, we know that wage goods may
> have sold above or below their value. If they sold above value, then the
> variable capital in the cost price would not have allowed for the hire of as
> many workers and thus as high rate of exploitation as Marx had assumed in
> transformation tables. If wage goods sold below value, then the variable in
> the cost price would have allowed for the hire of more workers and thus a
> higher s/v (the ratio of surplus value to variable capital) than Marx had
> assumed.

Yes, if we don't assume simultaneous determination, and view Marx's
two-step procedure as an embryonic account of the process of the
formation of the general rate of profit, then there will be such
mismatches because the economy is in a state of disequilibrium.

> My interpretation offers, then,  the problem of an inverse
> transformation. That is, the problems are not in the failure to transform
> the inputs from values into prices but in terms of the assumptions Marx
> actually did make in his tables about the value transferred and s/v as he
> moved from cost prices to output values and prices of production. Ultimately
> value transferred and s/v cannot be known directly but only inferred from
> price data. In constructing his tables Marx made it seem as if such value
> related variables are something we can know directly and explicitly and
> write into the transformation tables. That in my opinion is the problem to
> which Marx was calling attention.

OK, Marx is saying that the cost prices do not have this simple
relationship to labour values, as he had assumed at the beginning of
the transformation. I think that's a traditional reading of Ch.9 so

> And he had to fail since value magnitudes
> can never be known directly or explicitly, or even easily inferred from
> price data, including flow price data. The mistake in his transformation
> tables to which Marx was admitting was simply that he had treated the value
> transferred, the rate of exploitation and the value added as if they were on
> the same phenomenal level as the money sums invested as constant and
> variable capital But this violates his own theory of fetishism: value can
> only be represented in money price though money price mis-represents value.
> Representation is not necessarily not mis-representation (Mattick, Jr,
> 1991).

OK, that reading makes sense. But one way of resolving this problem is
to assume equilibrium and move to simultaneous equations and
simultaneous determination.

Marx of course couldn't do that. Which is why I said his conclusions
(and starting points?) deserve close examination.

I'm not convinced Marx's theory of commodity fetishism is directly
connected to divergence of PoP and values, as you are suggesting here.
Marx wished to resolve law of value and uniform profits -- so claimed
that money price *conserves* value, rather than misrepresents value.
In this sense, it very much represents value correctly, otherwise why
would Marx use a MELT to translate between prices and values? -- This
is really a linear conservation law, after all.

> In short, Marx never admitted to a failure to transform the inputs
> from values to prices and there is thus no gap in his theory which calls for
> the use of linear production theory by means of which the inputs and outputs
> can be transformed together into identical prices.

No, I think linear production theory can help a lot here. It helps to
address the problem you identified above.

> Simply put, the first
> table is not in values; they specify the cost price of commodities, and cost
> price is obviously a monetary phenomenon. If the inputs which enter into
> cost prices are already (as the term indicates) in the form of prices, why
> do they have to be transformed into prices? I therefore reject the idea that
> Marx himself gave his imprimatur to a problem that can only be solved by
> means of linear production theory,

Quite a few big jumps at the end of your argument, I think. Also, Marx
obviously thought there was a problem that the transformation was
intended to solve -- it was the contradiction between the law of value
and law of uniform profits that the Ricardian school tripped up on.

We will never know whether Marx would have liked linear production
theory, and whether he would have accepted the premisses of
Bortkiewicz's "special case". But I think it might be helpful to
distinguish the classical TP (pre-simultaneous determination) and the
modern TP, which is, fortunately or unfortunately, a problem
formulated in terms of simultaneous determination and linear
production theory.



This archive was generated by hypermail 2.1.5 : Mon Jan 30 2006 - 00:00:02 EST