[OPE-L:3919] Re: The Transformation Problem 2

From: Ajit Sinha (ajitsinha@lbsnaa.ernet.in)
Date: Sun Oct 01 2000 - 07:16:44 EDT

Though Bortkiewicz's  two  papers  were  published  in  Polish  in
1906-7, it  came  to  international  attention  only  when  Sweezy
presented Bortkiewicz's arguments  in  his  famous  book  (Sweezy,
1942) and later published an English translation of his 1907 paper
(in Sweezy ed. 1949). After this the first round  of  debate  took
over which  invariance  postulate  between  value  and  prices  of
production should be added to the  system  (see  Laibman  1973-74,
1992 for a detailed discussion on various invariance  postulates).
Furthermore, Bortkiewicz had conducted the  transformation  debate
within the  system  of  simple  reproduction  schema  (see  simple
reproduction). Winternitz (1948) argued  that  the  transformation
problem does not require the assumption of balanced  relationship.
And later Seton (1957)  further  generalised  the  formulation  to
n-commodity system where there is no reason to assume that  "every
physical commodity [is] not merely unequivocally  identifiable  as
the product of one or other  of  these  [sectors],  but  that  its
ultimate  use  in  the  economy  [is]   equally   invariable   and
predetermined by its department of origin." (p. 150).  Soon  after
this, in 1960, Sraffa, not intending to solve  the  transformation
problem,   presented   a   physical   input-output   system   that
simultaneously solved for the relative prices and the average rate
of profit, given the real wage rate from outside the system.  From
Sraffa's physical input-output system, one could  directly  derive
the commodity values as well as the solution for  prices  and  the
average rate of profit. This gave rise  to  the  charge  that  the
labour-value accounting is redundant for a theory  of  prices  and
distribution in a surplus approach framework (Steedman, 1976;  see
'surplus approach economics'). Furthermore,  it  had  been  argued
that the transformation of values to prices of  production  cannot
be relied  upon  because  there  is  no  objective  basis  to  the
selection of one invariance postulate from many  good  candidates;
as Seton had proclaimed "... and to that extent the transformation
problem may be said to fall short of complete  determinancy."  (p.
153). Recently this charge has  been  put  in  a  much  friendlier
terms. Garegnani (1991) has argued that  Marx  did  not  have  the
simultaneous equation method available to him, so in  the  absence
of this tool Marx's method of deriving the average rate of  profit
was the best that could have  been  done.  Much  earlier,  Eatwell
(1974, 1975), relying on Garegnani's reading of Marx,  had  argued
that  Sraffa's  discovery  of  the  'standard  commodity'(see  the
'standard commodity'), which was  apparently  developed  to  solve
Ricardo's problem of the 'invariable measure of value', is also  a
solution to Marx's transformation problem. He argued that with the
help of Sraffa's system and the  'standard  commodity'  one  could
derive the average rate of profit  directly  from  the  production
conditions and the given rate of surplus value in the  system,  as
Marx had attempted.

Though the  Sraffians  have  solved  the  problem  of  prices  and
distribution  for  the   surplus   approach   economics,   thereby
vindicating Marx's basic theoretical framework, this has  come  at
some price to Marxists. In the Sraffian approach  the  concept  of
labour-value of commodities becomes redundant. However,  for  Marx
the concept of labour-value was essential; because on  this  basis
he was able to divide the live labour-time spent in the production
process between necessary and surplus  labour-time,  which  showed
that the source of all non-wage income lies in making the  workers
work for longer time than  what  was  needed  to  reproduce  their
wages--  thereby  revealing  the  exploitative   nature   of   the
capitalist system. Sraffians argue that Marx's notion  of  surplus
value and exploitation can be upheld from the distributional point
of view without having to use labour-time as the unit  of  measure
for economic variables. They argue on the basis  of  the  Lockeian
property rights that since capitalists  have  no  active  role  in
production they have no right to a share in the net  output.  Thus
the capitalists' share in the net output  can  be  interpreted  as
exploitation of labour, and the  degree  of  exploitation  can  be
measured by the money value of the surplus product divided by  the
money value of the net product (see Hodgson  1982).  Sinha  (1991)
has criticised the Sraffian notion of exploitation as non  Marxist
on the ground that it leaves the question of control of the  means
of production and the labour process out of sight. The net  output
is derived by deducting the used up means of production  from  the
gross output. This deduction is not just a mathematical  procedure
but requires the physical replacement of the means of  production,
which raises the question: who controls the means  of  production?
For whoever controls the means of production must also control the
gross output for the required deduction to take  place.  Thus  the
notion of capitalist relation of production, which gives  rise  to
capital's control over workers' time  and  hence  exploitation  as
appropriation of labour-time in the process of production is  more
fundamental and prior a category than the notion of the  right  to
appropriate  a  share  in  the  net  product.  Marx  himself   had
vehemently criticised the notion of exploitation based  on  income
distribution in the 'Critique of the Gotha Program'.

Many scholars have argued that the divergence of total profit from
total surplus value or the total prices  of  production  from  the
total value is not all that damaging for Marx's basic  proposition
about exploitation, since it can be proven that positive profit is
possible if and only if  there  is  positive  surplus  value  (see
Wolfstetter, 1973;  Morishima,  1973,  Morishima  and  Catephores,
1978). Recently Sinha (1991, 1996) has argued in favour  of  using
the condition that  total  value  is  equal  to  total  prices  of
production as an outside constraint  on  the  system,  given  that
values are substance and it is neither created or destroyed in the
process of exchange.  Moreover,  the  system  must  be  put  in  a
balanced state since only  in  a  balanced  state  the  prices  of
production could actualise. In this  case,  Morishima  (1973)  has
shown that Marx's average rate of profit will come out to  be  the
correct solution if there is zero consumption by the  capitalists,
ie. all the surplus value is reinvested or  accumulated.  However,
this, in general, will not be true  in  the  case  of  capitalists
consuming a part of the surplus value. Shaikh (1984)  argued  that
this happens because capitalists' consumption becomes part of  the
revenue and falls out of the circuit  of  capital.  Since  we  can
explain the divergence of prices of production from values as well
as the divergence of total profit from total surplus value on  the
basis of the value analysis  itself,  the  transformation  problem
should be considered solved.

Note: the reader should know that I'm not convinced with Shaikh's argument. The
point has been left uncriticized there. Ajit Sinha

> Ajit in #3900:
> "First of all, the determination of what you call the "m" is the central
> problem with Marx's transformation problem."
> What is your definition of "Marx's transformation problem"? It seems to me
> that there is not an universal agreement about what is understood by this.
> Why is the determination of "m" the "central problem" of this... "problem"?
> "That's why assuming that the value of m is "given" can never be taken
> as a solution to the problem, and particularly when it is added that not only
> the real m is unknown but is unknowable."
> So, this problem would not belong to the realm of science? What is this,
> then??
> "This simply confirms the Steedman critique."
> Do you mean the "redundancy" issue here?
> Alejandro
> P.S. If you're very busy, you can perhaps post any of your writings (cited
> in another post) regarding this matter. It's obvious that for people like
> me it's really impossible to get a copy in another way.

This archive was generated by hypermail 2b29 : Tue Oct 31 2000 - 00:00:07 EST