[OPE-L:2547] status of the assumption

Fred Moseley (fmoseley@laneta.apc.org)
Wed, 19 Jun 1996 23:08:36 -0700 (PDT)

[ show plain text ]

In (2514) I wrote:

Paul C. has asserted in several recent posts that "one cannot have a theory
of surplus-value without a theory of relative prices." This assertion is
simply false.

I have presented an aggregate theory of surplus-value in my 1993 paper on
the transformation problem in Marx's Method (which I summarized in a
recent post (2441)), which does not depend in any way on a theory of
relative prices. This aggregate theory of surplus-value depends
primarily on the assumption that the total money value added in the
capitalist economy as a whole is proportional to the total current labor
in the capitalist economy as a whole. From this assumption, it is easily
derived that the total amount of surplus-value is proportional to
the total
amount of surplus labor, where surplus labor is defined NOT as the
labor-time embodied in surplus-goods, but rather as the number of hours
that workers produce money value added over and above the equivalent of
their money-wage.

Allin responded to me in (2515):

My question is: what is the status of the "assumption" that total money
value added is proportional to total current labour? Presumably the
idea of "proportionality" must mean that if we compare the economy in
various different periods, we find a common ratio of money value
added to labour performed. Now this is obviously empirically false,
and presumably to save the assumption we need to add in the
cetris paribus clause, "if there is no change in the value of money".
But then the assumption becomes, as Duncan puts it, a "stipulation".
What we have, it seems to me, is not any sort of proof or demonstration
that surplus labour-time is the source of profit, but rather a
proposition of the form, "let us consider the consequences of supposing
that surplus labour-time is the source of profit." If I understand
him right, Duncan may reckon that that is all we are entitled to
propose. On the other hand, I suspect that Fred (and others) are
in search of a "stronger" formulation. If relative prices are
(to a first approximation) proportional to labour-contents, then we
can justify a stronger formulation -- and I take this to be
Paul's point.

Duncan responsed to Allin in (2518):

This is pretty close to my position. In favor of it I'd point out that
this is the usual way that "core" propositions in theories come out to
confront empirical evidence. One considers the consequences, and
evaluates their predictive and explanatory value. If this value is high,
that tends to make us believe the theory.

My response:

I pretty much agree with Duncan that the validity of this assumption depnds
primarily depends on its explanatory power - i.e. on the range of important
phenomena that can be explained on the basis of this assumption, compared to
rival theories.

I have presented such an empirical test of the explanatory power of Marx's
theory in a recent debate with Mark Blaug. Blaug argued (in a set of
distinguished lectures at the U. of Amsterdam in 1980) that most of the
major conclusions of Marx's theory have been refuted by the historical
evidence. I reexamined the empirical evidence related to these conclusions
(the falling rate of profit, the relative impoverishment of workers,
technological change, conflict of the length of the working day, conflict
over the intensity of labor, periodic depressions, and the concentration of
capital) and came to the opposite conclusion: that Marx's theory is
largely consistent with the empirical evidence and has an impressive range
of explanatory power. To the empirical evidence I cited in this paper, I
could add the estimates for the UK economy by Paul and Allin on the
impoverishment of workers and the falling rate of profit.

I argued further that a comparative appraisal of the relative explanatory
power of Marx's theory of profit and the neoclassical theories of profit and
interest would reveal Marx's theory to be superior, i.e. would show that
Marx's theory of profit provides a much more substantial theory of
capitalism's important dynamic tendencies than does neoclassical theory, and
that the empirical evidence supports Marx's conclusions more strongly than
the neoclassical conclusions.

Blaug wrote a very interesting reply to my response to his empirical
appraisal of Marx's theory. Blaug disputed a few points, but at least
implicitly accepts most of my arguments, including that Marx's theory does
make important definite predictions regarding technological change,
conflicts over the working day and over the intensity of labor, the
increasing concentration of capital, etc., and that these predictions are
strongly supported by the empirical evidence. It appears that my
reexamination of the empirical evidence forced Blaug to concede that Marx's
theory does have considerable explanatory power.

Blaug also appears to agree that Marx's theory provides a better theory of
the dynamic tendencies of capitalism than does neoclassical theory. Blaug
quotes my conclusion to this effect and then says:

The point is made as soon as it is said: neoclassical economics is
essentially a static theory of resource allocation in a competitive
market structure, and as such does not make predictions about the
evolution of the capitalist system.

This is the type of empirical test that I think is the most appropriate way
to evaluate the validity of Marx's theory. I think that Marx's theory is
primarily a macroeconomic theory about the broad dynamic laws of motion of
capitalism and the appropriate empirical test of this theory has to do with
the extent to which the conclusions of this macro analysis are consistent
with the historical evidence.


P.S. My response to Blaug and Blaug's reply to me are included in a book
published last summer entitled "Heterodox Economic Theories: True or False?"
(edited by me and published by Elgar). This book also includes a response
to Blaug's appraisal of Sraffian economics and the capital controversy by
Ian Steedman, and a response to Blaug's appraisal of radical economics by
Michael Reich, and also includes replies by Blaug to all these two and
comments by Roger Backhouse, Wade Hands, and Bruce Caldwell. I think it is
an interesting debate, not only about the validity of these heterodox
theories, but also about the appropriate methodology for apparaising the
validity of economic theories.