[OPE-L:2570] estimation of abstract labor

Fred Moseley (fmoseley@laneta.apc.org)
Wed, 26 Jun 1996 07:20:58 -0700 (PDT)

[ show plain text ]

This is a response to Paul C's (2539) on the necessity and possibility of
estimating abstract labor. Paul, thanks for the clarification of your
estimation procedures.

1. To begin with, I wish to note that Paul has not responded to my main
point that the concepts of constant capital, variable capital, and
surplus-value are defined in terms of quantities of money, and therefore can
be estimated directly in terms of money. One may still wish to try to
estimate abstract labor for various other purposes, but this is not
necessary in order to justify estimating the components of capital in terms
of money.

2. With respect to the concept of abstract labor, Paul wrote:

We take the position that the problems that one has in understanding a
historically lower mode of production can be solved if you consider the
more advanced form that succeeds it. Thus to understand value and abstract
labor properly you must as Marx often did, consider the matter from the
standpoint of a future communistic society.

Our initial analysis of the problem of how to define abstract labor is thus
from the context of how a socialist planned economy could perform its
calculation in terms of labor. One can then look back at capitalism and
understand to what extent the same things appear there in fetishitic form.

I disagree with this conception of abstract labor and this general
methodological principle. I find no evidence that Marx defined his concepts
from the perspective of a future communist society. Rather, Marx's
emphasized that his concepts are HISTORICALLY SPECIFIC, i.e. they are
defined in terms of their function in capitalist society. Therefore, Marx's
concept of abstract labor applies specifically and solely to capitalist
society and is not derived from a future communist society.

As an aside, I generally agree with your and Allin's arguments against
Roemer, et al. that socialist planning could be in terms of labor-time. But
this has no bearing on Marx's concept of abstract labor in his theory of

3. Paul argues that skilled labor in socialism and therefore in capitalism
is converted into abstract labor by adding the hours of training labor to
the hours of skilled labor, an interpretation first suggested by Hilferding
and which has been the dominant interpretation ever since. However, I find
no evidence for this interpretation in Marx. Rather, it seems to me that
Marx simply took the skilled labor multipliers as given and did not provide
a theoretical rule for the determination of these multipliers.

Instead, Marx assumed in Chapter 6 of Volume 1 that the WAGES of skilled
labor are determined by adding the costs of training to the reproduction
costs of simple, unskilled labor.

4. Paul and Allin's estimation procedure (I focus here on Method A which
yielded the strongest correlation between prices and values), takes the
total wages for each industry as a proxy for the total amount of abstract
labor in each industry. (By the way, is this the same estimation procedure
used by Shaikh, Ochoa, et al.?) This estimation procedure is significantly
different from their concept of abstract labor, discussed in #3. Their
concept of abstract labor, combined with Marx's assumption regarding the
wages of unskilled labor, also discussed in #3, implies that different
skills will have different rates of surplus-value (the rate of surplus-value
for skilled labor will be less than the rate of surplus-value for unskilled
labor), which is not by itself a problem. However, their estimation
procedure implies that the quantity of abstract labor, and hence the
quantity of new-value produced, is proportional to the wages paid. This
procedure implies that the rates of surplus-value will be the same for all
skills. Therefore, their estimation procedure produces biased estimates of
their concept of abstract labor.

5. By treating wages as a proxy for labor, their estimation procedure
directly tests the correlation between prices and wages. Therefore, the
resulting correlation can also be interpreted as empirical support for the
"cost of production theory of value" (CPTV) as well as for the LTV. The
CPTV was of course the main rival theory of Marx's (and Ricardo's) LTV.
Paul, how do these results or further tests discriminate between the LTV and
the CPTV?

6. Paul, do these estimates distinguish between productive labor and
unproductive labor?

7. In conclusion, I reiterate that, even if one could solve the above
problems and derive more rigorous estimates of abstract labor, this is still
not necessary to justify estimates of constant capital, variable capital,
and surplus-value in terms of money. These concepts are defined in terms of
money, not in quantities of abstract labor. Capital is defined as money
that becomes more money. Marx EXPLAINED the transformation of money into
capital on the basis of the labor theory of value (I would say on the basis
of unobservable quantities of abstract labor); but the observable phenomena
to which capital and its components refer are quantities of money that
function as capital.