[OPE-L:4592] Dobb on the transformation problem.

From: Rakesh Narpat Bhandari (rakeshb@Stanford.EDU)
Date: Sat Dec 02 2000 - 20:11:31 EST

Here's Dobb in his Theories of Value and Distribution, p.159 :

"It is quite true that Marx never satisfactorily demonstrated how 
these Prices of Production were related to or derived from Values; 
and without such a demonstration there was no logical ground for 
declaring that the former were determined by the conditions of 
production and the socal relations of production that he had dealt 
with in Vol 1. Indeed the arithematical examples that he uses to show 
the connection are unsatisfactory, since the transformation into 
Prices of Production is only applied to the outputs, and not to the 
inputs (whence it follows that he uses the same rate of profit as in 
the value situation). Although Marx shows awareness of the 
incompleteness of his own solution and hints at the real nature of 
the problem, he nowhere improves on these examples--possibly because 
of the incompleteness of the manuscript of Vol III, which he never 
succeeded in finishing, let alone rewriting. Moreover, he seems to 
maintain *both* that total prices will equal total values *and* that 
total profit will equal total surplus value: two conditions that are 
incompatible save in exceptional circumstances...But since both 
inputs, including labour power, and outputs have to be transformed 
into price terms, and hence in all probability the rate of profits 
will be affected, these have all to be determined simultaneously and 
interdependently, by solving a set of simultaneous equations."

1. Dobb is correct that the transformation procedure is incomplete. 
Here I disagree with Fred.

2. I agree with Fred however that Marx's original, unmodifed 
transformation tableau (Capital 3, p. 256) is not in values, as Dobb 
suggests.  The cost price (c + v) is the money sum which has been 
laid out as constant and variable capital. However  the prices at 
which the input means of production and means of subsistence were 
sold has been determined on the controlling assumption that all 
commodities exchange at prices *proportional*  to value (that is, all 
commodities are required to exchange at their full labor value * the 
monetary expression of labor value which has remained constant 
throughout the demonstration). So the inputs are not in values; they 
are in prices proportional to the values of the input means of 
production and means of subsistence. There is indeed no value 
table--you, Alejandro, Andrew and others are correct.  The original 
table is denominated however in direct or simple prices, and the 
inputs are not already in the form of prices of production.

3. While Marx holds total price (422), total cost price (312) and 
total surplus value (110) invariant in his own *incomplete* 
transformation (Capital 3, p. 256), it follows from Marxian theory 
that given an output of fixed value and price,  total surplus value 
would have to be modified in inverse direction to the modification of 
cost price consequent upon the transformation of the inputs while 
this modified sum of surplus value should  equal the sum of profits 
in the final transformed scheme. In short, the second equality (sum 
of surplus value = sum of profits) simply cannot be an invariance 
condition as well as an equality in a complete transformation 
procedure.  Dobb is wrong about this. There is only one invariance 
condition in the complete transformation.

(Both Andrew and Allin have objected that if surplus value is defined 
as total price minus modified cost price, it is possible that after 
the complete transformation, the total cost price of the commodities 
could be less than the value of the commodities consumed in the 
output and surplus value thereby increased; in other words, unpaid 
*direct* labor would no longer be the sole source of surplus value. 
It seems to me impossible however to come up with an example with the 
remotest relation to reality in which such a result would obtain. At 
any rate, Marx did not argue that only unpaid direct labor is the 
source of surplus value; he argued that in his gedankenexperiment in 
which he has postulated  that all commodities exchange at prices 
proportional to their values, the consumption of labor power was the 
sole source of surplus value. In the real history of capitalism--as 
opposed to Marx's thought experiment--the consumption of freely sold 
labor power, i.e., unpaid direct labor,  has indeed *not* been the 
exclusive source of surplus value, e.g., plantation slavery, 
colonialist extraction of raw materials, various forms of unfree 

4. It is grossly antithetical to Marxian theory to transform the 
inputs and outputs via the use of simultaneous equations into the 
identical unit prices of production, as recommended by Dobb.  This 
has the effect of excising dynamic changes in unit values from the 
formalisation of Marx; Marx simply cannot be formalized in terms of 
simultaneous equations--Carchedi and Alan F are right!   Ricardo 
himself recognized: "alterations in the quantity of labour necessary 
to produce commodities are a DAILY occurence. Every improvement in 
machinery, in tools, in builidngs, in raising the raw material saves 
labour, and enables us to produce the commodity to which the 
improvement is applied with more facility, and consequently its value 
alters." (principles, p. 36, Sraffa's ed; my emphasis).

It does not follow from the recognition that the inputs  have to be 
put in the form of prices of production that their magnitude should 
be identical to the output unit prices of production. In fact we know 
that the technical conditions in the period t - 1 to t could not have 
been the same for the data in Marx's one period tableau of t to t + 
1. We thus cannot determine the unit prices of production for the 
inputs at (t) from the one period data at hand (t to t + 1). Marx was 
correct to leave the inputs as they were and only make a mental note 
that the cost price should not be equated with the value of the used 
up commodities.

One does not need to know what the unit input prices of production 
and the cost prices were in order to understand the *logic* of Marx's 
transfomration procedure which demonstrates how the average rate of 
profit itself can become the form in which the law of value asserts 
itself. A modification of cost price would have minor effects on the 
magnitudes of the average rate of profit and the prices of 
production; it would not change the logic in terms of which they are 

Yours, Rakesh

This archive was generated by hypermail 2b29 : Sun Dec 31 2000 - 00:00:03 EST