[OPE-L:1007] Allin's second question

Alan Freeman (100042.617@compuserve.com)
Thu, 8 Feb 1996 07:49:47 -0800

[ show plain text ]

(On point 2 of Allin's OPE-L 985 of 7/2/96; continues
from response to Allin's point 1)

Are values defined independent of prices?

You say

"On the approach expressed by the equation v = pA + L values
are *not* defined independently of prices"

I think this illustrates the confusion sowed by purely nondualist
treatments which are not also sequentialist. I should have made it
clearer that time subscripts, if not explicitly inserted, are implicitly

This makes it clear that there are *two* sets of prices involved;
the price of inputs, and the price of sales. If one writes the equation

v(t+1) = p(t) + L(t)

one sees that values are indeed independent of the prices *for
which the goods are sold*. That is, if the product is linen and
the inputs are cotton, the value of linen is independent of the
price of linen. Moreover the "price" of linen is not uniquely
defined, since the price of linen at time t+1 many not equal the
price at time t.

The value of linen at time t+1 *does* depend on the price of cotton
at time t. But this does not at all violate the conditions of Marx's
Chapters 1-5 value definitions, because these chapters discuss only
values and prices at a given moment in time, or at most discuss
simple circulation, which involves no production, so the prices
and values under discussion are always the prices of the same

'Independence from prices' means that the value of a product is
independent of the price of that product, in the current
period. In the case above, let's put some numbers to it.
Suppose the cotton costs $15 and the labourers add $10. Then
the value of the linen is $25. But this $25 worth of linen
might actually sell for $30, $50 or $10,000,000. It would have
no effect on its value. Moreover the same amount of linen
purchased last year, at the same time as the cotton, might have
cost $20, $30 or anything. This too would not affect the value
of the linen we just produced.

To forestall as possible objection, are we defining surplus-
value as simply identical to profit, and value as simply
identical to price? No: to see this I'll ask another question
that amounts to the same thing:

How do we know that the $10 added by the labourers is really
the surplus-value and doesn't contain any transfers

Because it is not given by the circumstances of the individual
producer but of society as a whole. To calculate surplus value
correctly, strictly we should ascertain the average value
added by one worker; we do this by taking the total value added
by all workers across the whole economy (in money). Dividing this
by the number of workers yields the value added per worker, in
money terms, this year. We would require more calculation to
reduce this money measure to labour-hours, but as I am sure you
would agree this is in principle do-able.

Surplus value is then this magnitude multiplied by the
number of person-hours used in making the linen.

I'm not asking you to agree with this, because you have a
different empirical methodology, but I would try to convince
you that it is a perfectly practical and rational procedure
which defines the value of linen independently of its price.
And it is highly compatible with Marx.

It also makes it clearer, in my view, what Marx means when
he speaks of the 'pure' case. I think by the 'pure case'
he simply means that the linen actually sells at its value.
But this does *not* depend on the assumption that the
cotton is *purchased* at its value. In point of fact Marx's
treatment is indifferent to price-value differences in inputs,
because the value of inputs is represented by a sum of money,
not a quantity of goods.

I think the underlying simplicity of this construction is
disguised by the simultaneist presentation. If we regard
"prices" or "values" as the solution to a simultaneous equation
then there is only one set of "prices" and one set of "values".
The simultaneous equation becomes part of the definition of
price, and the 'problem' becomes to reconcile two
irreconcilable definitions.

For us, price is not 'defined' in the sense of 'algebraically
determined'. It is a given. It is data. Therefore, when you say

"given v = vA+L, the equation v=pA+L holds if and only if p=v"

I would say no, there are two sets of prices and two sets of
values here. First of all given

v(t+1) = p(t)A(t) + L(t)

then the following equation

v(t+1) = v(t)A(t) + L(t)

will hold if and only if prices and values were empirically
equal at time t. But even if this is the case, there is no
reason that prices and values should be equal at time t+1; and
in fact the goods whose value is v(t+1) can sell for anything
you like; we cannot know until we have observed what happens.

For the simultaneist, however, if prices and values are equal
at one time, they must be equal at all times, and this is
probably where the confusion arises.

Socially necessary labour time

Now, what is the 'closeness of fit' between this and socially
necessary labour time? What happens when, in Marx's words, the
capitalist 'lays out a sum of money c upon means of

As discussed in OPE-L:973 of 7/2/96, any sum of money is also a
sum of value. Indeed it can even be converted from money to
hours, if desired. If the monetary expression of value is $1
per hour then the capitalist's $15 represents 15 hours.

True, this 15 hours is different from the socially necessary
labour time required to produce the cotton, which was 10 hours.
But it has a very clear connection with the total labour of
society, as follows: if we consider the total value of the
outputs of society in the previous period, then this total
value has a monetary expression, let us say $1500 (small

What does the $15 represent? One-hundredth of this social
labour. If this labour was 1500 hours, then they represent 15

Why is this different from $10? Because in the course of
circulation, the capitalists purchasing the cotton have been
obliged by the high price of cotton to use a larger than normal
share of the labour of society, in order to produce their
linen. On the other hand, capitalists somewhere else will draw
the benefits of this, because they will pay $5 less for their
inputs and so use a smaller than normal share of the total past
labour of society to produce whisky, or corn, or whatever.

It is a common but I think utterly pernicious idea that for
Marx, money was a veil. The standard interpretation requires
that by some means that has never been explained and indeed is
supposed to be part of the Great Mystery which only highly
trained performing Hegelians can understand, the producers of
linen can act as if the cotton really did cost them $10 instead
of the $15 they paid.

Not so.

They *cannot* act as if the cotton really did cost them $10
without abolishing capitalism, or to be precise without
abolishing money. $15 - that is, 15 hours - *is* its social
cost to them. For them, as long as society fixes the price of
cotton at 15 hours, then 15 hours *is* the labour time socially
necessary for them to acquire their inputs. If, for example,
they were free peasant labourers purchasing this cotton on
the market but producing directly with their own labour, they
would have to labour for 15 hours, not for 10.

Money is *not* a veil. Money allocates and reallocates,
distributes and redistributes, ceaselessly and remorselessly,
the past labour of society. Money is a *social relation*, as
Marx himself emphatically says. In a society in which you can
only obtain goods with money, the labour time socially
necessary to acquire these goods *is* the value of this
money with which you acquire them.

This could only be falsified if by some mysterious means, what
was true for the individual ceased to be true for society. The
'standard construction' betrays its technocratic origin by
speaking as if society really were communist, and really had
constructed a hidden organised division of labour which exactly
corresponded to shadow prices secretly equal to values.

But the real division of labour in society is not that which
corresponds to sale at values. If the cost of acquiring, say,
diamonds, is exceptionally unfavourable because of a monopoly
or because of rent on mineral production, then the number of
labourers using diamonds as grinding tools is correspondingly
reduced. And the number of bourgeois who bedeck their partners
with the output of this industry is correspondingly increased.
For 'society' as it actually is, for 'actually existing
capitalism' to coin a phrase, the consumption of diamonds cost
*more* labour than the theoretical communist optimum.

Thus, I would argue, the nondualist construction is a more
accurate and faithful representation of 'socially necessary'
than the standard solution.