[OPE-L:6436] More on prices, technical choices, and values

From: Gil Skillman (gskillman@MAIL.WESLEYAN.EDU)
Date: Wed Jan 23 2002 - 15:50:11 EST

Paul C. writes in 6247
>This is an initial reply without having worked out the maths in detail but
>it strikes me that your example may not hold if one carries out a full i/o
>analysis. In those circumstances there will be only one choice ot technique
>which maximises the overall output of the economy. One or other of your
>two techniques will be dominated. The technique which is the dominant
>one will be the technique which uses the minimum labour overall, 
>taking intersectoral feedback into account and as
>such, defines the amount of  socially necessary labour to produce the
>product, and is thus the defining technology for determining value.

You could well be right, Paul.  Obviously I didn't work out the full
mathematical details here either.  But I note the following comments from
John Roemer in his _General Theory of Exploitation and Class_, where he
obviously *has* worked out the mathematical input-output details.
Commenting on his analysis of a subsistence economy with alternative
production techniques and endogenous prices, he gives the following
prospects for reswitching:

"Clearly, reswitching can only occur when neither of the [class-specific
endowment] vectors dominates the other component-wise....Perhaps the most
prevalent example of real exploitation is where some producers have a lot
of everything, and others very little of anything, and in this case the
reswitching phenomenon cannot occur.  Second, we see that reswitching here
is a phenomenon of multiple equilibria in a general equilibrium model.  We
are, as it were, comparing the situation on two identical islands which are
supporting different prices; if we heed Joan Robinson's warnings about the
inadmissability of such interisland comparisons, then must view this
reswitching as a curiosus." [p. 44]

In sum, he argues for the generality of the case I've suggested, in which
the choice of technique, and thus the vector of commodity values, depends
on relative prices (which depend in turn on endowments), even though
reswitching does not typically occur.  

>It is quite possible, and Marx makes this clear at a number of places,
>that where the rate of exploitation is high, capitalist production will use
>more labour than is socially necessary, chosing to waste labour rather
>than use machinery. But this is not relevant for determining whether
>this amount of labour is socially necessary, it only indicates that
>capitalist production is inefficient by the criterion of socially necessary

First, where exactly does Marx make this assertion?  I understand him to
assert more typically that, except transiently, capitalists take care to
maximize the bottom line, and thus to minimize waste.  Let this passage
from Volume I represent a number of others that make a similar point:

"[Labour] must be expended with the average amount of exertion and the
usual degree of intensity; and the capitalist is as careful to see that
this is done, as he is to ensure that his workmen are not idle for a single
moment. [Notice that with fixed-coefficient techniques, which Marx
generally assumes in numerical examples, this is not possible to do with
excess labor.--GS]  He has bought the use of the labour-power for a
definite period, and he insists on his rights.  He has no intention of
being robbed.  Lastly...*all wasteful consumption* of raw material or
instruments of labour is strictly forbidden, because what is wasted in this
way represents a superfluous expenditure of quantities of objectified
labour..."[I, 303, emphasis added].  

Seemingly, to Marx, the representative capitalist assiduously *avoids*
wasteful or superfluous expenditures.  

Moreover, notice that your claim works against the contention driving
Marx's "general law of capitalist accumulation,"  namely that they
continually strive to develop techniques that substitute constant capital
for labor.  Why would capitalists incur the trouble and cost to do this if
they're already using more labor than is socially necessary?

Second, though, I'm not talking about inefficient use of labor inputs; I'm
saying rather that *efficient*  techniques (which must be adopted at least
some of the time), and thus corresponding labor values, are determined by
relative input commodity prices.

> The value of a product by the definition of socially necessary labour
>does not depend upon which technology the capitalists actually use, since
>the high rate of exploitation may cause them to find it cheaper to produce
>the product at a labour expenditure above its value.
>The extent to which this occurs is a measure of the social product
>lost due to the accounting system imposed by capitalism.

I don't agree with this assessment.  You're asserting here that efficient
techniques may differ from cost-minimizing ones.  But that will not be the
case under a wide range of input market conditions, including of course
"competitive" conditions.  I note in this connection that if, as you assert
two paragraphs above, that capitalists are "wasting" labor, then that means
they are not minimizing cost.  I've registered my doubts about this
contention as a general statement of capitalist behavior above.  Here I'll
just note that even if this were true, it merely modifies my point rather
than eliminating it.  Since, by your representation, this phenomenon of
wasting labor occurs only when the rate of exploitation is relatively high,
then the phenomenon must disappear when wages become relatively higher.  My
point would still apply for any value of relative wage higher than that
"tipping point."

>I dont dispute that in practice the interest rate and the wage rate are
>relevant for the choice of techniques actually used. What I dispute is
>that these are relevant for the definition of value.

It's not the *definition* of value that's at issue here, but the
determination of value *magnitudes*. If input commodity prices affect the
choice of technique, and the technical composition of capital varies across
techniques, then in general they must affect realized value magnitudes.

>> But second, suppose that your point was apropos.  Shaik et al's empirical
>> result in no way guarantees that the indicated result would *always* hold;
>> we might therefore be a set of technical innovations away from the reverse
>> outcome, and my analytical point would still be relevant.
>It is possible, but my suspicion is that the existence a real i.o table
>with hundreds of products in which reverse reswitching occured
>is vanishingly improbable.

That sounds plausible to me.  But my point is that even if this were true,
the existence of alternative techniques with unequal technical compositions
implies that labor values will typically depend on relative input prices.
Or perhaps more accurately, value and price magnitudes will in general be
simultaneously determined, so that it wouldn't be legitimate to assert that
values are in any coherent sense analytically prior to commodity prices.


This archive was generated by hypermail 2b30 : Sat Feb 02 2002 - 00:00:06 EST