[OPE-L:3778] Re: Revisiting the Unimportance of the Transformation Problem

From: Allin Cottrell (cottrell@wfu.edu)
Date: Thu Sep 07 2000 - 08:42:53 EDT

[ show plain text ]

On Wed, 6 Sep 2000, Paul Zarembka wrote:

> There is a very interesting, new paper by Alejandro Ramos,
> "Value and Price of Production: New Evidence on Marx's
> Transformation Procedure", *Int. Journal of Political
> Economy*, Winter 98-99 but only in print very recently.
> The basic message is that Engels' edition of Volume 3 left
> out portions of text in which Marx clearly puts forward a
> second illustration of the transformation problem.
> Furthermore, Bortkiewicz even distorted what Engels did
> publish, distorted in a manner to support his own two-system
> interpretation of Marx's attempted solution....

I agree that Alejandro's paper is interesting. He was kind
enough to show me it before publication and I wrote a
response. In case it's of interest I append a slightly edited
version below.

Dear Alejandro,

Thank you for letting me see your article "Value and Price of
Production..." I am impressed by your scholarship...

On the other hand, you will not be surprised to hear that I
disagree with the main conclusions you draw. You contribute to
the discussion of both theory and the history of thought. I too
am interested in both of these. I shall confine my remarks to
two passages where the differences between us, on both counts,
are particularly apparent.

You say (I'm splitting up your text for ease of reference):

(1) 'In [the Bortkiewicz type of] interpretation, the value and
price of a commodity differ for two reasons: on the one hand
because surplus value ... is not equal to average profit ...
_and_, on the other, because cost-price "in value" ... diverges
from cost-price "in price" ...'


The problem with the contrast you want to draw in (1) - i.e.
Bortkiewicz and co. _versus_ the real historical Marx - is that
Marx himself says exactly this (that value and price differ for
the two reasons mentioned) on more than one occasion. For
instance Marx says in TSV:

"The difference between the cost-price and the value of the
commodity is thus brought about in two ways: by the difference
between the cost-price and the values of commodities which
constitute the pre-conditions of the process of production of
the new commodity; by the difference between the surplus-value
which is really added to the conditions of production and the
profit..." (related to the discussion of Bailey; I can find the
page if needed).

Marx also says In Capital III:

"We have seen how a deviation in prices of production from
values arises from:
  1) adding the average profit instead of the surplus-value
contained in a commodity to its cost-price;
  2) the price of production, which so deviates from the value
of a commodity, entering into the cost-price of other
commodities as one of its elements, so that the cost-price of a
commodity may contain a deviation from value in those means of
production consumed by it, quite aside from a deviation of its
own which may arise through a difference between the average
profit and the surplus-value." (Moscow, Progress Publishers,
1971, pp. 206-7)

This seems to me to be precisely the view you attribute to
Bortkiewicz, as an erroneous and unsupported interpretation of

This would seem to be inconsistent with the equations you
reproduce from the Main Manuscript, according to which "Werth =
K + m" and "Productionspreiss = K + p'" (with only one
"cost-price" term involved, namely "K", and hence only one - and
not two - sources of deviation of price of production from
value). The interpretation which I favour, and which saves
Marx's consistency, is that when Marx writes the sort of thing
you quote he is still assuming that value = price for the
commodities entering into cost-price (although he will write
later that there is "always the possibility of error" in making
this assumption). Otherwise it is a flat inconsistency on
Marx's part, and it is then hard to know what to make of his

(2) 'As a result, values and production prices are defined in a
_dualistic_ fashion as two completely separate "systems" or
"worlds", thus severing the real and conceptual _unity_ between
them: it is the case neither that value is expressed effectively
(although contradictorily) by price nor that price is a
manifestation of value. Both magnitudes are conceived, rather,
as parallel, _non-related_, ideal rules of price formation.'


The "dualist" insists on a conceptual distinction between the
system of prices of production and the system of values. Does
it follow that he is ruling out the possibility that price
"expresses" value or is a "manifestation of" value? Is he
breaking a "real and conceptual unity"?

Well, we "dualists" are nonetheless talking about one world.
It is a world in which (a) workers produce commodities via the
expenditure of labour time, and in which almost all commodities
are reproducible given the expenditure of sufficient labour.
It is also a world in which (b) the means of production are
owned, and labour-power is hired, by capitalists, and in which
the product belongs to the capitalist in the first instance, and
where (c) capitalists attempt to move their capital between
sectors of the economy (to some extent and over time) in search
of the best profit rate - thus potentially producing some
tendency towards convergence of sectoral profit rates (or at
least placing some limit on the degree of dispersion of sectoral
profit rates).

In relation to this world, the concept of "value" is not an
"ideal rule of price formation" (or at least not in the first
instance). It is an conceptualization of the total direct plus
indirect labour time required to produce a commodity - something
which is not directly or easily measurable due to the manifold
interdependencies of the capitalist economy but which, once it
is set out clearly, can be seen to express a very important
property of commodities in view of (a) above. Such an important
property - such an intuitive measure of the real "cost" of
commodities - that one might, as a first approximation, expect
to find market prices "gravitating" around values.

Price of production, on the other hand, _can_ be described as an
"ideal rule of price formation" - "ideal" in the sense that it
gives all capitalists their "fair" share of the aggregate
proceeds of the exploitation of labour (something which they
obviously never receive in reality). It is the never-achieved
endpoint of the inter-sectoral movement of capital in search of
the highest available profit, a concept formed in abstraction
from things like technological change, active marketing
strategies, and changes in consumer demand, which are
continually working to create _divergence_ of profit rates.

Thus, so far: Value is not "ideal". It expresses something
quite real, if not directly measurable, namely the total labour
required to produce a commodity. Price of production is more
"ideal", if you wish. It's the answer to a hypothetical: what
prices would have to rule if (impossibly) all capitalists were
to achieve the same rate of profit?

At the time of Marx's writing, the most astute political
economists (Smith, Ricardo) had claimed _both_ that prices would
tend to be proportional to values (total labour requirements)
_and_ that profit rates would tend to be equalized. Ricardo
realized that these two statements did not exactly jive. Marx
set out to produce, among other things, a more accurate and
precise account of the relationship between values (labour
contents) and prices of production (equalized-profit prices).

Are the dualists "severing a real and conceptual unity"?
That's not the way I see it. We have three distinct concepts:

A. the total amount of labour time required to produce a commodity.
B. the idealized "price of production".
C. the prices at which commodities are actually bought and sold.

No pair of concepts from the threesome forms a "unity". But
there may be close relationships among them. Take A and C.
Actual market prices (C) represent the terms on which
alternatives are offered on the market, i.e. the cost to
individual agents of acquiring particular commodities.
Labour-requirements (A) can be seen as representing the cost "to
society" of producing particular commodities (given that almost
all commodities are producible with enough labour). This is an
interesting hypothesis: labour-contents will tend to "shine
through" the mirk of market competition, or in other words
market prices will tend to be roughly proportional to values, so
that the cost to an individual of acquiring a given commodity
bears a non-arbitrary relationship to the cost to society of
producing it. This was Ricardo's thought; I believe it was also
(basically) Marx's.

B and C: _If_ there is a strong tendency for the profit rate to
be equalized, the terms on which individuals actually buy and
sell commodities cannot diverge too far from the terms on which
all capitalists would make the same rate.

A and B: Marx helped to clarify this relationship: it does not
seem that A and B _have_ to be close, but they will be closer
(a) the less the dispersion of organic composition of capital
across sectors and (b) the lower the overall rate of profit.
If the profit rate is high and organic composition is very
widely dispersed, than A and B will be widely separated.

We may say, with definite empirical content, that prices
"express" or "manifest" values - _if_ it turns out that A and C
are "reasonably close" (according to some suitable metric).
If, on the other hand, it turned out that the ratios under B and
C were very close, but quite different from A (e.g. if organic
compositions were widely dispersed and the profit rate both high
and fairly well equalized), I for one would _not_ want to say
that prices "expressed" values. I think this would be quite
misleading. I am suspicious of any argument that purports to
prove that price "expresses" or "manifests" value _regardless of
the empirical relationship between A and C_. (I doubt whether it
was Marx's intention to produce such an argument. Were this his
intention then I for one would regard his theory as a piece of
obfuscation and a regress from Ricardo.)

(3) 'Dualism can be traced back to Engels' "historical"
conception of the transformation, in which "value" would belong
to a pre-capitalist stage ... while "production price" is
particular to capitalism.'

I don't find this plausible. The _historical_ separation that
Engels spoke of (values under simple commodity production,
prices of production under capitalism) is quite different from
the _logical_ separation of value and price of production in
Bortkiewicz and his followers. On the latter view (where the
equalized rate of profit is a maintained hypothesis) we have
prices "equal to" values not before capitalism, but in a
hypothetical capitalism where organic composition is everywhere
the same (or the rate of profit close to zero). There's nothing
"historical" about Bortkiewicz's prices of production or values:
they're both solutions to sets of simultaneous equations, but to
different sets.

Besides, it is wrong to describe the "historical" conception
(whatever its merits) as an invention of Engels. In this
respect Engels was following Marx. See for instance the lengthy
discussion on pp. 175-7 of Capital III (Moscow, 1971), which
ends with these words:

"The exchange of commodities at their values, or approximately
at their values, thus requires a much lower stage than their
exchange at their prices of production, which requires a
definite stage of capitalist development. ... Apart from the
domination of prices and price movements by value, it is quite
appropriate to regard the values of commodities as not only
theoretically but also historically _prius_ to the prices of
production. This applies to conditions in which the labourer
owns his means of production, and this is the condition of the
land-owning farmer living off his own labour and the craftsman,
in the ancient as well as in the modern world."

I have not checked this against the MS, but the passage is
certainly not flagged as an interpolation by Engels.

Allin Cottrell
Department of Economics
Wake Forest University, NC

This archive was generated by hypermail 2b29 : Sat Sep 30 2000 - 00:00:04 EDT