[OPE-L:5669] prouctive and unproductive labor

Fred B. Moseley (fmoseley@mtholyoke.edu)
Mon, 3 Nov 1997 13:02:07 -0500 (EST)

[ show plain text ]

This is a response to Mike W.'s recent criticism of Marx¹s
concepts of productive labor and unproductive labor. I
agree with a number of the points that Mike makes,
although I disagree with the conclusions he draws, for
reasons to discussed below.

1. First of all the agreements: I agree - with a
qualification - that productive labor in capitalism refers
to labor that produces *value*. The qualification is that
productive labor is more precisely labor that produces
*surplus* value for capital. The production of value is a
necessary but not sufficient condition for productive
labor. For example, self-employed commodity producers
produce value, but not surplus-value for capital; hence
their labor is not productive labor.

I also agree that *service* sectors may also produce
commodities and surplus-value for capital, and hence
labor in these service sectors may be productive labor.
However, one must be careful how one defines ³service²
in this regard. Services such as hotels and restuarants,
recreational services, repair services, health services,
education services, etc. that are produced by capitalist
enterprises involve productive labor. But a broader
definition of ³services² which perform functions of
circulation (such as trade and finance) do not involve
productive labor, for reasons which I will discuss below.

I also agree that the social usefulness or lack thereof of
the commodities produced has nothing to do with the
distinction between productive labor and unproductive
labor. Labor that produces nuclear warheads within
capitalist enterprises is productive labor. So I agree with
Mike¹s critique of Baran and Sweezy's "socially useful or
wasteful" criterion of productive and unproductive labor
(I have made similar critiques in the past).

However, and here is where our disagreements begin,
Marx¹s reason for considering circulation labor as
unproductive labor is not that it is "socially wasteful in
some trans-historical sense", but rather the fundamental
premise emphasized throughout Capital (beginning in
Chapter 1 and especially in Chapter 5 of Vol. 1) that *the
exchange of commodities is essentially an exchange of
equivalent values.* No surplus-value is produced in such
an exchange of equivalent values. As Marx put it in
Chapter 5, "where there is equality, there is no gain."
According to Marx's theory, commodities enter the
process of circulation with their values already
determined (by the labor-time required to produce
them). The function of circulation labor is to transform
this predetermined amount of value from the form of the
price of commodities to the form of money money, or vice
versa. No additional value is produced in this change of
form of a predetermined amount of value. Please see the
long paragraph on p. 260 of the Penguin edition of Vol. 1
(the sixth paragraph from the beginning of Chapter 5).
Therefore, Marx's assumption that circulation labor is
unproductive labor is not based on a comparison with a
hypothetical socialist society, but rather on the function
performed by this labor within capitalist society and the
assumption that this function is not productive of
additional value.

With regard to Mike's "two broad ways" to approach the
problem of productive and unproductive labor, I myself
try to follow both paths - both to try to figure out exactly
what Marx meant by this distinction, his basis for the
distinction and the role it plays in his theory, and also to
try to apply this distinction to an analysis of
contemporary capitalism (as in my book on the falling
rate of profit). I think that the distinction can be
operationalized in a rough sort of way, as I and Shaikh
and others have done - including listmembers Alan F. and
Paul C. and Allin C. For the UK economy. (I and colleages
- including listmember Abelardo Marina - are currently
working on similar estimates for the postwar Mexican

Most importantly, this distinction makes a huge
difference in how we view contemporary capitalism and
in the trends of the key Marxian variables during the
postwar period, at least for the US economy, (and I think
most other economies as well). If one ignores this
distinction, as Weisskopf and many others have done in
their empirical work, then it appears that the rate of
profit declined in the postwar period because the "rate of
surplus-value" declined, not because the composition of
capital increased (i.e. contrary to Marx¹s theory). It also
appears as if the rate of surplus-value in the US economy
is now only about 0.2, after more than a century of
capitalist development, which implies that about 850f
the average working day is necessary labor and only 15%
is surplus labor.

On the other hand, if one takes this distinction into
account, as I argue that one should in a rigorous
application of Marx¹s theory, then one reaches very
different conclusions. The rate of surplus-value
increased about 20 0n the postwar period, rather than
declined, and the rate of profict declined because the
composition of capital increased faster than the rate of
surplus-value, as Marx concluded. Furthermore, the rate
of surplus-value based on this distinction is about 2.3,
which implies that more than 1/3 of the average working
day is necessary labor and the remaining 2/3+ is
surplus labor, as one would expect after a century of
capitalist development in which the productivity of labor
has generally increased faster than the average real

Furthermore, I think that this distinction provides a
superior explanation of the decline in the "conventional"
rate of profit in the postwar US economy (the
"conventional" rate of profit is the rate of profit on the
total capital; the above discussion has to do with the
"Marxian" rate of profit on productive capital only).
According to this explanation, the main cause of the
decline of the conventional rate of profit (which has
declined a lot more than the Marxian rate of profit) is a
very significant increase in the ratio of unproductive
labor to productive labor during the postwar period. I
would be happy to review some of the details of this
explanatioin if anyone is interested (please see Chapter 4
of my book). But the point that I wish to emphasize now
is that this theory is able to explain not only why the rate
of profit declined, but also why the recovery of the rate
of profit has been so slow since the mid-1970s, in spite of
declining real wages and continuing productivity growth.
The reason is that the ratio of unproductive labor to
productive labor has continued to increase and has
partially offset the very significant increase in the rate of
surplus-value during this period. I would argue that in
this important sense, the "unproductive labor" theory is
superior to the ³profit squeeze² theory - the main
alternative radical theory of the decline in the profit rate
presented by Weisskopf and others. It seems to me that
the "profit squeeze" theory cannot adequately explain
why the rate of profit has recovered so slowly. If the
decline of the rate of profit was indeed caused by lower
rates of unemployment that increased the bargaining
power and wages of workers, then surely two decades of
higher rates of unemployment have significantly reduced
the power and wages of workers, so this problem should
have been solved and the rate of profit should have been
more or less fully restored. But this full recovery has not
occurred. Less than half of the prior decline has been
recovered (this is discussed in my forthcoming paper
"The Rate of Profit and the Future of Capitalism" in the
next issue of the RRPE). It seems to me that this is strong
empirical evidence in support of Marx's distinction
between productive and unproductive labor.

In sum, I think that Marx's distinction between productive and
unproductive labor is both logically coherent and also has
significant explanatory power in understanding
contemporary capitalism. I look forward to further