the vexed issue of productive labour in a modern setting
j.bendien at SPAMwolmail.nl
Thu Jun 7 22:08:00 MDT 2001
Replying to Greg, Carol and Julio:
When the classical political economists were concerned with the issue
productive labour, they wanted to identify which types of labour increased
(material) wealth, and which did not. Adam Smith for example regards
"services" as unproductive, but he has in mind mainly personal services.
To some extent this classical problematic survives in Marx's thought as
well, in the following way: in capitalist society (material) wealth takes
the form of a mass of commodities, and productive labour in capitalist
society is commodity-producing labour, labour that adds to the total stock
However "wealth" and "value" are not identical concepts in Marx. Marx also
recognised that labour, although it is the only source of new value, is not
the only source of wealth; land is also a source of wealth for example. He
distinguishes economic (material) wealth from human wealth saying that
human wealth is wealth in human relations. Part of his critique of
capitalism is that it periodically destroys material wealth, and often
substitutes alienation for human wealth. A man may "gain the world but lose
So anyway Marx takes over the concept of productive labour from the
classical economists, but he criticises it, he tries to critically
reformulate and refine the concept. He tries to define it from the
standpoint of what is "productive" from the standpoint of the individual
capitalist and from the standpoint of the capitalist system as a whole,
recognising that in a class society there cannot really be any "neutral"
definition of this concept. On the one hand, he rejects a narrow definition
which limits this concept to labour which has a specific kind of
"productive" output. On the other hand, he rejects the notion that any
labour is productive. But he doesn't really get beyond saying that
productive labour is labour that produces surplus-value, he is interested
in that social relation especially. The general economic concept of
"productive labour in class society" that is implicit in Marx, is labour
which can produce an output in excess of its own reproduction cost
(measured in labour time).
So then it doesn't matter what the activity is, as long as it produces
surplus-value for an employer then it's productive labour.
Macro-economically we might then say that labour is productive, if it adds
to the total mass of surplus-value. From this then follows straightaway the
"law of motion" that capitalist development will, ceteris paribus, convert
more and more labour into productive labour over time. In other words, the
overall division of labour will gradually change so that the proportion of
productive labour is increased at the expense of other sorts of labour.
But really we are no further ahead at this point, because for the purpose
of measurement we have to identify the activities which create
surplus-value and the ones that don't. This has to be resolved in a
non-arbitrary, coherent way.
Why is important to know this anyway, you might ask ? Well (1) because it
affects our understanding and measurement of the Marxian rate and the mass
of surplus-value (roughly, aggregate profits). In turn, this (2) affects
our understanding of the Marxian rate of profit, of the tendency of the
rate of profit to decline, of the social reproduction process and of major
crises of the capitalist system.
For instance, in the Grossmann-Mattick-Shaikh theory of crisis, a
disproportion develops in capitalist development such that, in the course
of a decline in the rate of profit, at a certain point the total volume of
surplus-value ceases to grow, resulting in a major crisis. The proportion
of labour which is productively employed has quite a lot to do with that.
It is rather trivial if unproductive labour amounts to only 5 percent of
the labour force, but it is a different story if it is 50 percent or more.
(3) Marx does need a concept of productive labour because he wants to say
that in the capitalist mode of production (perhaps as distinct from
merchant capitalist activity, in proto-capitalist society or precapitalist
society), no new value is created or added in exchange. From the standpoint
of an individual capitalist, productive labour is simply labour which
generates profits, but from a macroeconomic point of view this cannot be
true in Marx's theory. Labour which is only involved in enacting the
exchange process (the sphere of circulation) does not create surplus-value
and hence is not productive. Nevertheless banks, administrators, security
people and traders etc. make profits. Where do they come from ? Well Marx
seems to argue they are a kind of "transfer payment" from the productive
sector, a transfer of surplus-value. The value product is already there
prior to its distribution, and banks etc. appropriate a part of the
surplus-value it contains in the form of interest, rents, mark-ups etc.
This appropriation is a kind of "impost" on the productive sector, a sort
of "faux frais" of production.
A similar argument is made by Paul Cockshott about the arms industry: he
argues the surplus-value obtained there, is a transfer payment from the
productive sector. But, you might argue, bullets and missiles are
commodities like any other, so why isn't the labour producing them
productive ? He argues, because these commodities do not enter into the
process of economic reproduction. It seems more like hot air to me but OK
that is what he argues.
Logically the implication of Marx's transfer idea that the wage bill of
non-productive labour cannot be a component of variable capital, because it
adds no new value to the social product, even though it is part of its
cost-structure. What is the source of this wage income then ? Marx himself
does not solve this problem. One school of Marxists (e.g. Moseley, Shaikh)
says it represents a deduction from the current flow of surplus-value.
Another school (e.g. Murray) says this cannot be true, because this would
mean these workers directly appropriate surplus-value, i.e. they are really
capitalists; also perhaps because capital must be laid out on wages for
unproductive labour BEFORE a fraction of surplus-value can be claimed, or
perhaps because this capital really should be seen as a component of
circulating constant capital (Mandel resolves the issue by saying
unproductive labour is paid out of "social capital", presumably a fund of
circulating constant capital which already exists). Again, this problem is
not so important if unproductive labour represents 5 percent of total
labour, it is a different story if it is 50 percent or more.
But leaving this aside, another problem is the definition of "commodity"
itself. It is through the production of commodities that surplus-value is
created, so then we need to be able to specify what a commodity is and what
commodity production is. For instance, if I buy a tourist package or a
financial package, is that a commodity ? Are services (where production and
consumption coincide) commodities ? Is information a commodity, and under
what circumstances ? Some of this is discussed on the OPE-L list.
A commodity, says Marx, has both a use-value and an exchange-value. Marx
explicitly defines the use-value of a commodity in terms of the physical
attributes of a tradeable object which is capable of satisfying human needs
or wants. He wants to say this object is not an object in thought, but in
tangible material reality. If we were to say that a use-value is the
"useful effect" of a tradeable object, then it would seem that at some
point we must leave the materialist terrain and enter the terrain of
subjective (marginal utility) theory. Because whether or not a commodity
has a "useful effect" can be a purely subjective matter.
Can information be a commodity ? Of course, provided it is in a form so
that it is a tradeable object, preferably a form in which you can
mass-produce it for a mass of consumers. You have to be able to separate it
from the producer, and attach private property rights to it (be able to
monopolise it). For this, it may have to have some material substratum. The
political economy of information (the mode of exploitation of information)
hasn't really been well analysed by Marxists, which is one reason Marxism
is no longer so popular in educated circles. One of the first people who
explicitly analysed information as a commodity was Kenneth Boulding, but he
wasn't a Marxist.
Potentially information exchange provides a boundless source of
exploitation and profits, within all sorts of subtle exchange transactions,
but this potential may not be realised, because of the nature of
information itself, and the nature of human beings, which may be difficult
to adjust to this fully. Some may adjust to it because they have to, others
may not. It is an open question, information economics as a specialised
science is only just starting to get going, it's a relatively new field of
The reason why I think information can be an "unstable commodity" is (1)
because it may be difficult to monopolise it (e.g. Napster), (2) it may
be difficult to extract it from the producer, within the framework of
conventional bourgeois law (personal rights, privacy rights, intellectual
property rights etc.), and (3) its use-value may vary to extremes or reduce
to zero, in the space of less than minute, so that its exchange value is
lost or suddenly jumps to a fantastically high level (e.g. the stock
exchange). As soon as somebody "lets out the secret" as it were, the
information ceases to be monopolised, it ceases of be the private property
of an owner, and what is worse it may lose its use-value altogether.
The dehumanising (alienating) aspect of information as a commodity is a
big subject. It resides partly in the fact that communications become
reshaped as transactions, that is the "natural" conversation or interaction
between people who give, get and receive out of mutual appreciation or love
becomes reduced to an exchange transaction. Interpersonal communications
become governed by rules such as "I give only in order to get or to
receive", "I prefer to get or receive rather than to give", "I receive only
in order to get", and so on. But there is a lot more to be said about it,
for instance in the Lukacs/Meszaros/Ollman type of analysis of reification
As regards Mandel's somewhat "Smithian" concept (cf. Alvater's criticism)
of productive labour, this is set out in Marxist Economic Theory, Late
Capitalism and his Introduction to Capital Vol. 2.
Personally I do not think all of Marx's ideas of productive labour in
capitalist society can be sustained at the same time, they are not
consistent. I think the only acceptable solution is to say that all labour
is "productive" from a capitalist point of view simply if it generates
profits for the employer engaged in the exploitation of that labour,
regardless of whether transfers of surplus-value occur or not, and
regardless of the form that profit takes (rent, interest or whatever). Such
is the economic valuation of labour from a capitalist point of view. From a
socialist point of view the definition may be quite different, of course,
because a different scheme of valuation is applied. A socialist might
question, does this activity really contribute to material or human wealth,
or reduce it, and so on.
More information about the Marxism