I am taking up an issue raised by Gil, Alan and (briefly) Patrick, namely the
status of Marx's argument in chapter 5 of vol. I. I have a somewhat
different perspective, though it overlaps in some ways with points made by
others.
Marx admits that use-value can be augmented via exchange (if commodities pass
from those who do not have a need for them into the hands of those who do).
But he insists that exchange-value cannot be augmented in this way. Now if
we assume that prices and values (in the sense of embodied labour-time) are
equivalent, the latter point is made very easily. It is obvious that
exchange per se cannot augment the sum of the socially necessary labour-time
embodied in the commodities exchanged, and so the non-augmentation of
exchange-value follows as a direct corollary. If, on the other hand, we
assume nothing regarding value-price equivalence, the proposition that
exchange-value cannot be augmented via exchange requires a specific
supporting argument in its own right. Marx claims to supply such an argument
-- and he is obliged to do so if he wants to prove his point in full
generality. Following this argument leads to some quite interesting
conclusions.
Before going further, let's back up a little: What exactly is to be
explained? Look back at chapter 4. It is the phenomenon whereby the
capitalist can lay out $100 on commodities, then turn around and sell
commodities to the value of $110, so that "more money is finally withdrawn
from circulation than was thrown into it at the beginning". To make a
puzzle of this, certain stipulations are necessary. For instance if the
capitalist lays out $100 on cotton at t0, then at t1 sells that cotton plus
some more from his inventory for $110, there is no mystery. There must be
some definite lineage traceable from the original commodity-purchase to the
final commodity-sale. On the other hand, we don't want to stipulate that the
final commodity-bundle is the _same_ as the original bundle, since that would
restrict the analysis to the case of merchant capital. The stipulation has
to be that the final commodities were produced using the originally purchased
commodities alone (where the case of identity may be counted as a special
case of production).
So, what makes this possible? On the face of it, one possibility is that our
capitalist ends up with an extra $10 while somebody else ends up down $10, on
the set of transactions as a whole. But if that "somebody" is another
capitalist there is no net gain to the capitalist class, and Marx's problem
is to explain how the phenomenon of surplus-value can be generalized for the
capitalist class as a whole.
Is this another possibility: there exists a class of _non-capitalists_ who
supply certain inputs to the capitalist class for less than their "true
value"? In other words, the capitalists are selling the $110 of goods at
"what they are worth", but the trick is that they are obtaining $110 worth of
inputs for $100. A conceptual issue arises here: According to what metric is
the output sold "at its value", and the inputs sold to the capitalist "below
their value"? An associated question which I'll tackle first: In what
precise sense are the non-capitalists "down $10" on the transactions?
The capitalists' gain is tangible: there they are with $110 in the coffers,
while at the beginning of the process there was only $100. What about the
non-capitalist input-suppliers? Suppose they started out with zero in their
coffers (their "first move" is a sale, not a purchase, and they don't "start"
with money on hand). Let's get a bit more concrete, and make the
non-capitalists workers, supplying labour. And let's have the capitalists'
initial purchases divided into $50 on produced means of production
(intra-capitalist transactions) and $50 on labour. By assumption, the first
$50 buys goods "worth" $50. So -- on this line of reasoning -- the second
$50 must buy labour that is "really worth" $60. Let's have the workers spend
all of their income on consumption, so at the end of the period -- i.e. after
the sale of the final goods -- the workers again have zero in their coffers,
having received and spent $50 in the meantime. Their "loss", in that
case, is not at par with the capitalists' gain. While the latter is
tangible, the former is, if you like, "virtual": they have sold their
commodity for, in some sense, "less than it is worth", but they do not have
less money than they began with.
So where has the extra $10 come from? Follow it one more time: The $50 in
wages was paid out by the capitalists, and flowed back to them. The
capitalists sold goods to the (monetary) value of $110. There are only these
two classes. Conclusion: the remaining $60-worth of final goods must have
been purchased by the capitalist class. If we assume that all goods are sold
"cash on delivery", it follows that the capitalists must, somehow, have had
the extra $10 in their possession _before_ the final sales were consummated,
else those sales could not have gone through. Yet by assumption, at the
beginning of the period their coffers contained only the $100 they spent on
inputs. Ouch! There would seem to be only one resolution: extra money to
the value of $10 has been created somewhere along the line, and it has been
created to the account of the capitalists, enabling them to buy the $60 worth
of final product. In relation to the modern setup, this is not so strange:
the capitalists have borrowed $10 from the banks, who provided this money via
deposit-creation. The entry that balances the extra $10 in the industrial
capitalists' coffers is a debt of $10 to the banking system. The debt has
financed either capitalist consumption, or the net acquisition of means of
production (accumulation).
The hyothesis that the workers sell labour "worth" $60 to the capitalists for
$50 is, in effect, Rodbertus's theory of exploitation. As we all know, Marx
favoured the formulation that the workers sell labour-power "worth" $50, but
perform labour that creates a value of $60. One might say that there are
conceptual problems with Rodbertus's version, but Marx's version is not
without difficulties of its own (very briefly, it is problematic to talk of
"the value of labour-power" when labour-power is not a commodity produced
under capitalist conditions of production, has a "historical and moral
element" and so on). But anyway, which of these formulations one favours
does not actually make any difference to the monetary aspect of the
discussion above. Whichever way one looks at it, if one pushes Marx's line
of argument hard enough one ends up with Kalecki (for it is he who points out
that the source of money profits is debt-financed capitalist expenditure).
Allin Cottrell
Department of Economics
Wake Forest University