[OPE-L:4307] Re: Re: SV, labour and machines

From: Steve Keen (s.keen@uws.edu.au)
Date: Fri Oct 27 2000 - 09:00:28 EDT


The distinction *is* significant, but it is not, in my opinion, the reason
why labour creates surplus value.

It is why Marx first believed that it did--and in at least having a reason,
he was well ahead of Ricardo and Smith, who as he notes, could only say
that there was a difference between labour embodied and labour commanded,
but could not say why. Both Smith and Ricardo tended to believe also that
the contribution machines made to production was identical to their
cost--but again they couldn't say why.

Smith was less consistent than Ricardo on this front--he frequently argued
that machines returned much more than their cost of production--but again,
he couldn't say why.

However, when Marx developed the use-value/exchange-value analysis, he came
up with a second reason as to why labour was the source of value, but this
reason was in one sense diametrically opposed to his old reason: rather
than focussing on what is unique about labour--the labour/labour-power
distinction--he now focussed on what labour-power had *in common* with all
other commodities--including machinery.

In this sense, it would have been remarkable if this new logico-historical
reason had reached the same result as his previous logico-historical
reason, given that the former was a positive method, whereas the latter was
a negative one. Marx convinced himself that they were consistent, but I
believe that this was an instance of what he had previously said that Hegel
had done--let an incomplete appreciation of his own principles lead him
into error.

The type of reasoning you undertake in your second paragraph is typical of
the reasoning Marx used prior to 1857 to characterise why labour produced
value and machines did not. It is also typical of the reasoning he used
*after* 1857, with only one very important exception, and another important

The exception occurred in the Grundrisse, where--soon after having
developed the dialectic, and in the context of discussing technical
change--he pondered that "its devaluation in the service of production is
not proportional to its increasing effect on production." (p. 383)

This indicates that the positive methodology was capable of reaching a
different conclusion than its negative predecessor--that the general rule
of incommensurability between use-value and exchange-value for all
commodities led to the position that, when that primary use-value (the
use-value to the purchaser, a capitalist in the M-C-M+ circuit) was
quantitative, there would be a difference between use-value and
exchange-value which would be a source of surplus-value to the capitalist.

However, Marx retreated from this. I could find no subsequent time where
Marx reached this conclusion--though I haven't read all his correspondence,
so maybe it's sitting in there somewhere--but there is a readily available
reference where you see Marx tortuously trying to avoid this conclusion, by
reasoning so that, in the case of machinery, the use-value and
exchange-value are identical--and this is the qualification.

The reference is Capital, pp 193-199. The punchline to his proof, though he
didn't explicitly put it in this form, was that the exchange-value of a
machine was also its use-value.

Check that section of Capital, and you'll see that Marx was trying to
torture his positive methodology--in seven pages of dense reasoning--to
make it reach the same conclusion as his negative methodology. In the end,
all he managed to do was to completely obscure his positive methodology, so
that it was effectively lost to Marxism.

Since then, and until Rosdolsky's work (then Groll, then mine), Marxists
were generally aware only of Marx's negative methodology--and there was a
discussion on OPE recently where people concurred that Marx's proof of why
labour alone produces value was a negative one.

But why did it take him seven pages to prove that machinery didn't produce
value, if the basis of the proof was simply that there is no distinction
between commodity and commodity-power in the case of machinery, whereas
there is in the case of labour?

In my analysis, his positive methodology allowed him to do something no-one
else had ever been able to do--to identify why all inputs to production can
generate a surplus for the capitalist.

Prior to Marx, the Physiocrats had argued that land was the only source of
surplus--on the basis of a reductio-ab-initio argument that land preceded
man, and therefore land alone was the source of value. The reason they
believed this was that, in agriculture, it is obvious that the agricultural
outputs are very much greater than the agricultural inputs--one bag of
wheat in, 1,000 bags of wheat out. The physical surplus is apparent.

Then we had Smith and Ricardo, pointing out a similar sort of physical
difference: the labour it took to make something (labour embodied) was
different to the labour something could purchase (labour commanded). Marx
codified this into the distinction between labour and labour-power. Its
basis, as with the Physiocrats, was the somewhat less observable phenomenon
that the labour performed by workers exceeded the labour necessary to
produce their means of production.

But in all this, it wasn't possible for either group of economists to
perceive the same possibility with respect to machinery and other commodity
inputs to production. Some of these were completely used up in
production--circulating capital--how could these add more value than they
contained? Others, machines, took a long time to wear down, but the very
heterogeneity of machinery means that there is no observable difference
between their cost--depreciation--and their contribution to production.

Marx's positive methodology provided a means by which this could be
seen--the distinction between their exchange-value and their use-value.
Hence Marx's statement in the Grundrisse, which in full was that

"It also has to be postulated (which was not done above) that the use-value
of the machine significantly (sic) greater than its value; i.e. that its
devaluation in the service of production is not proportional to its
increasing effect on production."

This insight reveals that there is just as much reason to see machines as a
source of surplus value as there is to see labour as the source--in other
words, no input to production is specially blessed when it comes to value
creation. Labour is of course different on a whole host of fronts--but
these pertain to issues other than the creation of value and hence of
potential profit for the capitalist.

In the specific case of the distinction between the labour and
labour-power, this was a basis to argue that, whereas commodities
exchange-values equal their values, labour's exchange-value--the
wage--should normally exceed its value. Hence Marx's description of the
value of labour-power as the *minimum* wage.

Now the use-value/exchange-value logic can also be used to argue that the
exchange-value of machinery will sometimes exceed its value--but unlike
labour, it also indicates that on occasion its exchange-value will fall
well below its value. So it makes the analysis of prices much more
complicated than Marx's previous negative methodology did; but it also
eliminates the transformation problem--since surplus can be garnered from
machines as much as from labour, the link between s and v is broken, and
hence different organic compositions have no necessary impact on the rate
of profit.

By ignoring this positive methodology, and continuing with Marx's old
negative reasoning, in my opinion Marxists have not simply thrown the baby
out with the bathwater, they've kept the bathwater and thrown out the baby.
So adherents to the labour theory of value are stuck with the same old
dilemma that hobbled Marx, because they have not used the Master's deeper
logic to transcend his own inadequate appreciation of it. That's why I
think Marxists have done Marx such a disservice.


At 03:01 PM 10/26/2000 +0100, you wrote:
>What about my (Marx's?) argument that it is the distinctive nature 
>of the use value of labour power that is significant?
>Labour is universally and creatively transformative, unlike 
>machines. This means that the contribution of labour to production 
>is different to (bears no necessary relation to) the labour power 
>hired. They are two different things so can have two different values 
>so SV is possible. For machines, on the other hand, this 
>contribution is already 'contained' in the machine purchased. On 
>being 'realised' in production, the machine contributes something 
>particular, preordained and fixed - its contribution to production is 
>merely the realised form what was purchased to begin with. There 
>are not two different things in this latter case, and so the (present) 
>value of the sum of contributions of the machine is equal to the 
>price of the machine. No SV here.
>On a different point, any thoughts on my effort to interpret the 
>exchange value / use value incommensurability?
>Many thanks,
>On 27 Oct 2000, at 0:31, Steve Keen wrote:
>> Hi Andy,
>> I'm quite willing to accept that the adjective "pure" is too extreme. Marx
>> of course allowed that the use-value of workers had a qualitative
>> aspect--that it was work of a particular type, whether weaving,
>> ironworking, etc.--as well as having a quantitative use-value to its
>> capitalist purchaser. Of course the same subtlety applies to machines,
>> which have a qualitative aspect as well--that a machine used to produce
>> steel cannot be used to shear sheep, for instance--as well as having a
>> quantitative use-value to its capitalist purchaser.
>> My proposition is that the latter, quantitative use-value, is the reason
>> why capitalists both hire labour and purchase machinery, and that on this
>> logic, both are sources of surplus value.
>> Cheers,
>> Steve
Dr. Steve Keen
Senior Lecturer
Economics & Finance
University of Western Sydney Macarthur
Building 11 Room 30,
Goldsmith Avenue, Campbelltown
PO Box 555 Campbelltown NSW 2560
s.keen@uws.edu.au 61 2 4620-3016 Fax 61 2 4626-6683
Home 02 9558-8018 Mobile 0409 716 088
Home Page: http://bus.macarthur.uws.edu.au/steve-keen/

This archive was generated by hypermail 2b29 : Tue Oct 31 2000 - 00:00:12 EST