Re: Quantifying Value: response to Murray Smith

Alan Freeman (
Sat, 17 Jan 1998 15:37:29 +0000

> Comments:
> ...But in
> the passage that Alan cites from TSV Marx seems to be referring to
> unproductive labour of a certain type, and this passage by no means
> suggests that all unproductive labour is exchanged with revenue.

I'm not convinced that the passage concerned deals with labour of a certain
type: rather I think it's the other way around, and it is when Marx talks
about the labour of circulation that he begins to make distinctions between
types of labour. In the section I cited he is dealing with activites that
go in your category (1). Here, Marx seems at pains to point out that the
form of the product doesn't matter. He critiques Smith's 'Physiocratic'
attempt to distinguish labour that is 'fixed in a vendible commodity' from
labour of services, demonstrating that there are instances of both
material production and service production to be found in productive
and unproductive activities, so that the material-service distinction
(a distinction between types of labour) does not discriminate between
productive and unproductive labour in this instance.

Thus a schoolmaster who works for a private school is productive
(supporting Mike and myself that services are productive) but a seamstress
in the retinue of a lord who makes shirts for him is unproductive. This is
because the schoolmaster is employed by capital and the seamstress by

So I don't read it as 'labour of a certain type' unless by 'type' you mean
'in a certain social relation', and in this section at least I think he
asserts pretty categorically that unproductive labour is exchanged with

As I see it, the question is whether this enters into contradiction with
the second major part where Marx discusses productive and unproductive
labour: the labour of circulation. I suspect that this is what you are
getting at.

Your view seems to be that the labour of circulation (banking, commerce)
and perhaps supervision (bodyguards, security etc) is exchanged directly
against capital, contradicting the assertion in the section that I cited,
where Marx asserts that it is not exchanged directly against capital. In
that case, the citation I gave is not directly relevant since Marx (in
TSV) assesses the second type of labour in the addenda to TSV, in a
distinct discussion.

> "Having established what, for Marx, generally constitutes productive
> labor, we are now in a position to distinguish between the main forms of un
> productive labor. Two principal forms are already indicated: 1) labor that
> belongs to non-capitalist socioeconomic activities and therefore cannot
> serve to augment social surplus value (the labor of self-employed
> commodity producers is a major instance of this form within capitalist
> societies); and 2) labor that is in the emply of unproductive capital and
> therefore 'facilitates' the valorization process but does not contribute
> directly to the augmentation of social surplus value. To these should be
> added: 3) labor that is supported by the capitalist state and likewise
> contributes 'indirectly' to valiorization and accumulation through its
> role in maintaining the institutional framework of capitalist society; and
> 4) labor in the employ of productive capital that nevertheless contributes
> only indirectly to valorization and accumulation (supervisors, security
> guards, etc.)

(1) I think I agree with this. I'd observe that your categories fall into
two broad classes as you later make clear: on the one hand your category
(1) and on the other hand, all the others.

I tried to simplify this with the notion that the activities in category
(1) are unproductive because, although they produce a use-value they
produce no exchange-value, whereas the activities in category (2) produce
no use-value. This was an attempt to summarise in its bare essentials the
results of the more complex analysis that we've all been engaged in for the
last fifteen years, to varying degrees.

(2) "Labour that belongs to non-capitalist socioeconomic activities" has, I
think, to be precised. I would precise it as 'labour that does not exchange
against capital' (and hence must exchange against revenue) since in my
view, the very words 'exchanged against capital' means 'purchased by a
capitalist for the purpose of creating value', that is, functioning in the
labour-process, the unity of the production of use- and exchange-value. If
one took the view that certain types of labour can be functioning in the
labour process without creating value, then I think that one would have to
conceive of them as functioning a bit like a beast of burden, a special
kind of constant capital endowed with life. I'm not criticising this
necessarily, I am just trying to work out how to conceive of it.

The only other way that I can see to make this distinction is to separate
paid from unpaid labour. In that case we would say that the labour of
servants (paid in money) is capitalist but the labour of serfs (paid in
product) is not. I don't think that's what you mean but I'd like to check.
So the precise meaning of 'non-capitalist' does not reside in whether the
labour is paid, but in where the money comes from. So it seems to me that
'non-capitalist socioeconomic activities' must mean 'activities that do not
create value for the capitalists'. Since these may nevertheless be paid
activity, the important distinction is then where the money comes from.

I would then conceive of 'non-capitalist' activities as 'activities that
are directly exchanged against revenue', i.e. without the intermediation of
a third party who sells the results of the activity.

(3) self-employed is a big one. I have always tended to treat it using
Marx's notion of the petty commodity producer as 'divided in two' so that a
part of the revenue of the small independent peasant or artisan consists of
wages and a part consists of profit. When one meets real small commodity
producers and looks at their accounts and the way they get their income, in
practice it's IMO usually pretty clear what they are doing. Many very
definitely work as businesses (including bringing in extra labour, and
paying for it) so that they are clearly capitalists with, as it were, 1.2
workers. Others very definitely work as workers. They are really paid for a
definite activity rather than a definite product, and the body which hires
the activity does not hire it for personal use, but to sell the results to
someone else. In that case their income is a disguised wage.

It can be misleading to put all self-employed in the category of capitalist
because there has been a huge expansion of so-called 'self-employed' who
are really just workers with no employment guarantees. This is very visible
in the figures. It also means that the reserve army of labour gets
disguised as a vast army of mendicants, particularly in the third world.
But on the other hand there is a shading-off, a gradation from pure workers
to quite substantial businesses, like the well-off independent
professionals or the kulaks. So we can't call them all workers either. My
view is that the actual categorisation of the self-employed is (like many
so-called difficult tasks) just a very concrete task. In each country, one
has to look at the actual structure of the people who are called
self-employed and see, in great detail, what they actually do.

Treating them as unproductive might be a solution but I think it would
introduce some odd biasses. It would, for example, cut the national product
of India in half.

> "The last three forms of unproductive labor -- what might be termed
> 'circulation labor' (form 2) and 'social maintenance labor" (forms 3 and
> 4) -- have expanded considerably under advanced capitalism, while the
> first form has declined precipitously.

Except for a very large category: the labour of the state. I think this
part of Marx's analysis is the most appropriate for dealing with the labour
of the state, which functions under capitalism like a collectivised army of
servants (and has its historical origin in exactly that).

> "In Capital II and Capital III, Marx unequivocally affirms the existence
> of forms of unproductive labor that are not exchanged withe revenue, but
> rather with capital. In these later treatments, Marx is preoccupied with
> the specifically capitalist incarnations of unproductive labor: the
> socially necessary forms of unproductive labor that serve the reproduction
> of the capitalist socioeconomic order. To be sure, in Capital II Marx
> still refers to unproductive workers 'who receive for their services a
> part of the luxury expenditure of the capitalists' and 'who are themselves
> to this extent a luxury item' (Vintage edition, 1981, 486). But the main
> focus is no longer on 'form 1' unproductive labor of the luxury type.
> rather it has shifted to unproductive labor in the sphere of circulation,
> and to unproductive labor as a cost of circulation. The activities of
> buying , selling and bookkeeping are singled out by Marx as types of
> unproductive labor with a very different import than the unproductive
> labor of butlers, mails and stable boys:
> '...a part of the variable capital must be deployed in
> acquiring...labour-powers that function only in circulation. This capital
> advance creates neither products nor value. It proportionately reduces the
> scale on which capital advanced functions productively. (ibid. 210-211)'
> "Clearly, the unproductive labor with which Marx is here concerned i
> socially necessary to capital in so far as it contributes not to the
> 'creation' of surplus value but to its realization in the sphere of
> circulation. Moreover, Marx is clear that this form of unproductive labor
> is exchanged with capital. Yet Marx also refers to the capital advance
> made to acquire these unproductive labour-powers as a 'part of the
> variable capital' -- a problematic suggestion in view of his definition of
> variable capotal as capital exchanged for labor-power of the productive
> type (Marx, Capital I, Vintage edition, 1977, p. 317).
> "This apparent contradiction in Marx's thinking on the problem of
> circulation labor reflects a contradiction that exists in reality: the
> circumstance that labor involved in the circulation of capital is
> 'productive' from the standpoint of individual (commercial or financial)
> capitals, even though it is unproductive from the point of view of the
> social capital. What we are dealing with then is a 'variable capital sui
> generis': a portion of the social capital exhibiting some but by no means
> all of the characteristics associated with variable capital....

Thus, what is at stake seems to be whether there is a contradiction between
my category (1) (produces no exchange-value) and my category (2) (produces
no use-value). The question seems to reduce to this: does labour that
creates no use-value, in any case still produce exchange-value?

My feeling is the following: what you seem to be saying is that the labour
of circulation is really productive, because it is necessary to capital.

Let's begin from the case where instead of hiring its own guards,
accountants, treasury managers etc, the capitalist firm engages the
services of a bank, firm of security guards or independent firm of

There is a point of view (that of the US NIPA accountants but not that of
the UK NIPA accountants) which says that a bank is really selling a
like a hairdresser. Therefore its product is 'exchanged against capital',
that is, it is a cost of production like any other, an 'input'. What I'd
like to know is, how does this differ from any other input? In other words,
what practical meaning does it have to designate it as unproductive? In
this case, therefore, it would seem that the outcome of the approach is
that when purchased as a product, the activities of circulation are a
service like any other, a product like any other, that is, simply a part of
constant capital.

Now, one might say that if one engages the services of a distinct agency
for the purpose of doing accounts, guarding the premises, running the
treasury department, etc then one is buying a product, and this is
different from the circumstances where one directly hires the accountants,
the guards or the treasurers. I think that what Marx was trying to argue is
that it was no different, and that's always been the way I have tried to
look at the matter. If you agree with me, that when one buys - let us say -
the output of a firm of accountants, then this is an input like any other,
then I would ask, why is it different when you hire their workers directly?
This would imply that when a company outsources its accounts department,
perhaps setting up the accounts department as an independent commercial
enterprise in its own right and 'buying' its services, the workers were
transformed into productive workers by the act of outsourcing. I don't
necessarily say this is wrong, I just want to be clear what the concept is.
I'd like to know, what difference it makes to say that an accountant is
unproductive, in your way of conceptualising it. Does it make any
difference at all?

Now finally, as regards Marx's own categories, there is the following
question: are they consistent? I think they are. It is consistent (though
it might be arguable) to reason that labour which produces no new
use-values, also produces no new exchange-value. It is consistent to reason
that, if it produces no new exchange-value, it functions for capital
exactly like the other kind of labour which produces no new exchange value,
namely that labour which they hire directly for their enjoyments
(servants). It is consistent to treat both categories as paid for out of

I think, though I'm not sure, that you are not really saying that a special
portion of unproductive labour is exchanged with capital (that is, is a
part of constant capital). You seem to be saying that this labour is
labelled as unproductive. Which is fine, but it isn't the same as saying
that Marx was inconsistent; it's just an alternative point of view to