[OPE-L:4009] Re: Re: Re: Re: Re: Re: Re: Re: Re: Re: m in Marxs theory

From: Ajit Sinha (ajitsinha@lbsnaa.ernet.in)
Date: Sun Oct 08 2000 - 06:14:09 EDT

Duncan K. Foley wrote:

> Ajit writes in part, with my comments:
> >Duncan K. Foley wrote:
> >
> >...
> >
> >
> >>  I don't see why Ajit has such trouble with the idea that money
> >  > represents labor time. It seems sort of intuitive to me.
> >_____________________
> >
> >I don't have such trouble with the idea that money represents labor time. My
> >problem is which money, and how and how much of labor time it represents. I
> >don't think that when Marx said, "And what we call price of production is in
> >fact the same thing that Adam Smith calls 'natural price', Ricardo 'price of
> >production' or 'cost of production' ..." he simply made a huge
> >error. All these
> >are basically price concepts.
> I didn't really think you had that much trouble with the idea...
> >So Marx's transformation problem is an attempt to
> >develop his theory of prices based on labor-values of commodities, which are
> >determined directly by the given technologies of production.
> This point might bear a bit more examination. Do you think Marx had
> the same idea of a "given technology of production" as Sraffa, or
> Leontief, or we do? It seems to me that in the schemes of
> reproduction, and in the transformation tableaux, Marx seems to think
> of matters at a different level from the notion of a "technique of
> production" defined in purely physical terms.


I think both Marx and Ricardo had a good sense of technology in physical terms.
This has to be the basis of the notion of *indirect labor*, which is present in
both the authors. It may be true that neither Ricardo nor Marx had an input-output
structure of Leontief or Sraffa type in their mind. They may be very well working
with some notion of vertically integrated sectors. Now the reproduction schemas are
designed to show the flow of values from one sector to another at a much more
aggregative level such as capital goods sector and wage goods sector, but these
flows are predicated on the assumption that equal values exchange. My sense is that
Marx thought that properties derived from the reproduction schemas will not be
affected even if the exchange ratios where different from the value ratios, and he
would be right on that. Transformation problem is also about showing the flow of
value from one sector to other when one moves from production to exchange or
realization of the surplus value. That's why i think that the proof that the
general rate of profit is equal to S/(C+V) is central to this problem. He needs to
show that the flow of value from one sector to another is confined to the flow of
income between one class.

> Ajit:
> >That's why I think
> >taking prices as empirically given in this context is like putting the cart
> >before the horse--it simply negates the theoretical problem posed. So what is
> >the problem of "m" in this context? Now we know that by using either
> >Seton's or
> >Sraffa's formulations we could derive Marx's prices in terms of any commodity
> >as money commodity. In Seton's formulation taking any commodity as 'money
> >commodity' directly implies the assumption that we are assuming no value-price
> >deviation for that particular commodity.
> Duncan:
> I don't understand why your last statement follows. If we think of an
> abstract economy with profit rate equalization, and a particular
> commodity (say, gold) performs the functions of Marx's socially
> accepted general equivalent, why shouldn't we assume that commodities
> express their exchange values in terms of the ratio of their own
> price of production to the price of production of gold? Why can't the
> money commodity have a "value-price" (I would prefer to say "embodied
> labor coefficient-price") deviation just like all the other
> commodities?


I would say, you are right. There should be a price-value deviation for the money
or the gold commodity too. But if you recall Bortkiewicz-Seton types formulations,
they construct their simultaneous equations in terms of labor values and attach the
unknown deviation factors x, y, z, etc. to them, where there are n numbers of
x,y,z, etc. and n+1 unknowns; n numbers of x,y,z... and one general rate of profit.
So if we solve the system by putting say x = 1, then that implies that the
commodity associated with x, say gold commodity, has no value price deviation. But
this, of course, is an arbitrary decree. So i agree with what you say above.

> Ajit:
> >...If we leave out your empirical
> >derivation of "m", and just deal with the claim that labor theory of value
> >implies that the prices of net output (and in this case the prices
> >will have to be prices of production and not the empirical market
> >prices) must equal total live labor time. Then this gives us another
> >"m". But then this is again a decree similar to saying that
> >value-price deviation is zero for the money
> >commodity.
> Duncan:
> This gets close to the heart of the difference in ways of looking at
> the problem. The NI value of money depends, as you have pointed out,
> on the actual net product, which you view as a defect, but it seems
> to me could be argued for as a virtue of the concept. Historically
> the labor expended is embodied in a particular net product.


That is quite acceptable. The question is the relation of the labor-value of the
net product to the market prices or the prices of production. My point is that
theoretically one can impose the condition on the system that the total prices of
production in the system must be equal to the total labor-value of the net product.
This gives us a theoretical m, which could be used to solve for the prices of
production of n commodities and the general rate of profits. But would the
properties of the results satisfy what Marx was looking for? Taking the market
prices of the net output to derive m for the macro empirical exercises leaves the
question of the determination of n prices of production and the rate of profits
unresolved, and that's why i think it is not a solution to the transformation

> Ajit:
> >  Moreover, given this "m" when we analyze the property of Marx's
> >system, we find that many conclusion derived from taking this "m" do not sit
> >well with Marx's analysis of his system. Same happens with other such
> >candidates such as taking total prices of gross output equal to total gross
> >value on the ground that prices must be bound by the total value substance.
> Duncan:
> At the risk of repeating myself, let me repeat myself. The problem
> with gross output is that it is not well-defined except in special
> abstract models like the circulating capital model of production, and
> is, in particular, not unambiguously defined for real capitalist
> production.


That's true. For empirical analysis it will have more problem than your m. My point
was that theoretically several m's can stand on the same footing, in the sense that
all of them have some advantages and disadvantages.

> Ajit:
> >  It
> >seems to me that Sraffa's standard commodity comes closest to what Marx was
> >looking for, but still is not completely satisfactory in my opinion. Thus i
> >think the problem as posed by Marx is still unresolved, and probably
> >insolvable.
> Duncan:
> In my opinion the unresolved and probably unresolvable problem is
> Ricardo's problem of the invariable standard of value. I still don't
> see why you link the standard commodity to this issue, since Sraffa
> himself rejected that interpretation, and later loyal Sraffians like
> Heinz Kurz have taken great pains to show that while the standard
> commodity  simplifies the real wage-profit rate relation by
> linearizing it (assuming that wages are paid at the end of the
> production period), it does not constitute a standard of value which
> is invariable with a change in the real wage. (Unless the real wage
> happens to coincide with the standard commodity, which there is no
> theoretical reason for supposing to be the general case).


Sraffa's standard commodity is a solution to Ricardo's problem of invariance of
money commodity in the face of changes in wages given the technology. Ricardo's
other problem of finding an invariant measure of value (prices) in the face of
technological change, of course, has no logical solution. That's why i say a
dynamic theory of prices is a pipe dream, which Ricardo had dreamt long time ago.
And had many on ope-l read this literature, they could have saved plenty of theirs
and other peoples time and energy. However, Ricardo's two problems are logically
separate. The first problem has been solved by Sraffa and the second problem has no
solution. As I have mentioned in my recent RRPE paper that Sraffa did not consider
the standard commodity to be a solution to the transformation problem. But I still
think that Eatwell has a point. In your kind of framework the standard commodity
gives a better m, theoretically speaking, than the ratio of total prices of
production of the net output to the total live labor-time. I think both Kurz and
Salvadori underplay the role and significance of the standard commodity in Sraffa.
My sense is that workers do not need to consume the standard commodity for Sraffa's
result to go through. All you need is to measure the wages in terms of the standard

> Ajit:
> >
> >Now in your above statement, "...the basic macroeconomic determinants of
> >profitability are the ratio
> >of the value of output to the value of capital (what Tom Michl and I
> >call the "productivity" of capital) and the ratio of profit to total
> >value added (which is a transformation of the rate of exploitation).
> >I take the essence of Marx's theory of value to be the observation
> >that you can't change the rate of profit in the macroeconomy without
> >changing one or the other or both of these variables."
> >I'm not sure what you are trying to get at. In monetary terms your ratio of
> >"value of output [gross] to value of capital" minus one will give you the rate
> >of profit, i presume.
> Duncan:
> Not quite, since you have to deduct wages to get the profit rate. The
> productivity of capital minus one gives you the "maximal" profit rate
> corresponding to a zero real wage.


Well, I thought, in Marxian terminology, the gross output would be given by C+V+S,
and the capital will be given by C+V (assuming no constant capital). Thus the ratio
of gross output to capital minus one will give you S/(C+V), which is Marx's rate of
profit. But it seems you are taking the value of net output and wages are not
included in capital.

> Ajit:
> >Now this rate of profit is determined and will not change
> >unless something changes either the value of output or the value of capital.
> >What is the significance of the second ratio here? In anycase, I do not have
> >any quibble over these macro level definitional identities. My problem is, how
> >these macro level definitional identities help us in solving Marx's
> >transformation problem?
> Duncan:
> The "other factor" is the division of net output between wages and
> surplus value, expressed as the ratio of profits to the value of net
> output.
> I think "Marx's transformation problem" boils down to the question of
> whether it is possible to reconcile the appearance that profit is
> proportional to capital (at the level of abstraction of Smith or
> Ricardo's "natural prices") with Marx's view that the essence is that
> surplus value is unpaid labor time. Marx argues that it is possible
> to reconcile these points of view by regarding competition as a
> process of redistributing a given surplus value among the
> capitalists. This seems to be a reasonable and persuasive approach.


I'm inclined to agree with you on this.

> Duncan:
> In the notes that Engels published as ch 9 of Volume III of Capital,
> Marx makes a first pass at a mathematical (or at least arithmetic)
> demonstration of this idea, but falls into the fallacy of claiming
> that not only the aggregate surplus value, but also the aggregate
> profit rate, will be invariant to the redistribution of surplus
> value. In my reading, at least, Marx is aware that things are a bit
> more complicated than that because of the possible repricing of the
> elements of constant capital, but he doesn't pursue the matter, and
> doesn't seem ever to have returned to it systematically.


I'm inclined to agree with you on this too.

> The point I was trying to make was that perhaps Marx was more
> interested in the macroeconomic determinants of the profit rate and
> surplus value than he was in the determination of relative prices.


However, the question of the determination of relative prices are linked to the
determination of the general rate of profits, which you need to correctly determine
to establish Marx's relationship with total profit and total surplus value or even
the rate of surplus value and the profit-wage ratio. Cheers, ajit sinha

> Cheers,
> Duncan
> --
> Duncan K. Foley
> Leo Model Professor
> Department of Economics
> Graduate Faculty
> New School University
> 65 Fifth Avenue
> New York, NY 10003
> (212)-229-5906
> messages: (212)-229-5717
> fax: (212)-229-5724
> e-mail: foleyd@cepa.newschool.edu
> alternate: foleyd@newschool.edu
> webpage: http://cepa.newschool.edu/~foleyd

This archive was generated by hypermail 2b29 : Tue Oct 31 2000 - 00:00:09 EST