re. productive and unproductive labour

Allin Cottrell (
Mon, 2 Feb 1998 17:45:05 -0500 (EST)

Some thoughts in light of the recent discussions.

On what basis does Marx insist (most notably in Capital, II)
that the work involved in the capitalist circulation process is
unproductive (of value, and of surplus value)? Some
contributors have suggested that Marx thought that in this case
the use-value produced is somehow "not of the right sort" for
the labour to count as productive. If this is Marx's position,
it seems vulnerable, since elsewhere he clearly insists that the
specific character of the use-value produced (essential versus
frivolous, durable commodity versus evanescent service) has
nothing to do with the productive/unproductive distinction:
what matters is the set of social relations within which the
labour is carried out.

I think Marx's position is stronger than that. He's not saying
that the labour of circulation doesn't produce the right sort of
use-value, but rather that it doesn't produce _any_ use-value
(coupled with the idea that the production of use-value is a
necessary condition for the production of value, regardless of
the social relations in force). Now there's no doubt that the
labour of circulation is "useful" to somebody: it's "useful"
for a capitalist to get his commodities sold. We infer that
Marx means something more specific by the production of
use-value. There's a clue in the analogy he uses. He says that
the work involved in selling goods is like that involved in a
law suit over the ownership of an asset. Whatever it
accomplishes (however "useful" it is for the party who comes
away with title to ownership of the asset), it does not augment
the sum of use-values in existence. It's a zero-sum phenomenon,
with regard to use-value. I think this is a clear and
defensible criterion.

Also some people (including me) have suggested that what matters
for Marx's distinction between productive and unproductive
labour is whether or not the work in question is necessitated
solely by the property relations characteristic of commodity
production (i.e. beans must be grown, harvested, canned and
transported to the point of consumption in any social order, but
only if the canned beans take the form of a commodity is the
work of the sales clerks required, ergo, the latter is

On re-reading, I'm not sure that's accurate as a representation
of Marx's thinking. In Capital II Marx notes that the work of
book-keeping (in the broad sense of keeping track of the
movement and disposition of goods) is _not_ something specific
to commodity production -- indeed it will become more important
in a planned economy. Yet he still clearly says such work is
unproductive. It's an "incidental expense", regardless of the
mode of production. This time the criterion is not quite the
same as in the zero-sum lawsuit example. Actually, it seems
just plain commonsensical: book-keeping (even in the extended
sense) is as different from actual production, says Marx, as the
bill of lading is from the cargo in the hold of the ship. It's
a "real" distinction.

Allin Cottrell
Department of Economics
Wake Forest University, NC