[OPE-L:4960] Production and Circulation

Duncan K. Foley (dkf2@columbia.edu)
Fri, 9 May 1997 10:41:39 -0700 (PDT)

[ show plain text ]

Way back in March Alan asked

(i) do Ajit, Paul, Allin and David agree that the value of
goods are given prior to sale?

(ii) what do Bruce and Fred think?

(iii)what does Duncan think?

I found this a hard question, and spent quite a bit of time thinking about
it. In the meantime Jerry has started a thread on "ideal and real value"
concerning the principle of the "conservation of value", which seems to me
closely connected. As I try to formulate my thoughts, the issues also seem
to touch on some of the problems John Ernst has been raising about the rate
of profit as well.

As I read Marx, he thinks value is created in production through the
expenditure of living labor, and realized in circulation when commodities
are exchanged against money. (The one exception is the money commodity
itself, which does not need to be exchanged, since it's already in the
general equivalent form.) If we look at an economy over a specific period,
we can in principle (subject to a number of problems about measurement of
labor, reduction to simple labor, and distinction between productive and
unproductive labor) observe how much labor time was expended. We can also
observe in principle the money value added realized in the sale of those
commodities (taking account of the time lag between production and sale).
The stipulation that the "monetary expression of labor time" is equal to
the ratio of the money value added realized in the sale of the commodities
to the living labor time expended in their production is supposed to
correspond to this idea. So I would say that the labor time expended in the
mass of commodities sold in a given period is "given prior to sale", but
the expression of this labor time in money is not necessarily given prior
to sale. The definition of the "monetary expression of labor time" above is
inherently ex post. It tells us what happened after the dust has settled.
This might not be a completely satisfactory answer for Alan, because I
think he wants to know whether I think the "value" of the non-labor means
of production is determined before sale. While it's obvious that its
historical cost is predetermined, I tend to be convinced by Fred's evidence
that Marx thought of these intermediate inputs as being revalued to their
reproduction cost.

The ex post is quite different from the ex ante viewpoint, which is the
other moment of the dialectic, I suppose. Ex ante capitalists produce on
speculation: they don't know whether or for what price the product is going
to sell. They may ex ante have inconsistent and conflicting plans or
expectations. The process of circulation one way or another (in extreme
cases by crisis) mediates those differences. It's much harder, in my view,
to devise interesting theories that describe this ex ante side of the
process than it is to devise methods and measures that describe the ex post

Capitalist accounting has to confront some of the puzzles that Jerry
raises. Typically in the process of production a certain amount of the
output is spoiled or discarded. (For example in manufacturing computer
chips they have to discard a significant percentage of the finished chips
because they fail quality-control tests.) When these losses are actuarially
predictable accountants tend to view them as part of the costs of
production. There are also losses of other kinds, such as the destruction
of saleable output in accidents. These losses are often charged against
profits (or in Marx's terms, surplus value). When there's a recession or
depression and demand falls sharply the producers of spoilable commodities
are forced to mark them down to sell them, or, if they cannot recover their
variable costs, discard the spoiled output altogether. These losses are
charged against profits.

To my mind what the "conservation principle" means is that the circulation
of commodities (and even the inevitable wastage that accompanies it) cannot
change the amount of living labor expended in producing them, so that if
the money value added they realize fluctuates because of aggregate demand
failures, or gluts on particular markets, the consequence is a change in
the monetary expression of labor time, which implies that the "value" in
labor time units doesn't change.

John Ernst's questions about the treatment of depreciation seem to raise
the same ex ante/ex post dichotomy. The "rate of profit" is typically an ex
ante concept, which depends on some estimate or expectation of economic
depreciation of the assets, which may or may not turn out to be justified
ex post. It seems to me to be desirable to distinguish between two sources
of gains or losses for the capitalists in talking about the rate of profit:
one is the profits on production and sale, and the other is profits or
losses on changes in the prices of stocks of assets held as an aspect of
the process of production. I think Marx's discussion of the falling rate of
profit concerns profits on production. This rate of profit will fall if the
productivity of capital falls (or the value composition of capital rises)
and the rate of exploitation does not rise enough to offset it: in
particular if the rate of exploitation (or, in my terms the value of
labor-power) stays constant. Real capitalist firms, of course, face both
issues, and their ex post profits depend both on the profitability of their
production process and on the gains and losses they experience on their
stocks of assets held to accomplish production. The Okishio theorem seems
to me to address just one aspect of the problem, the profitability of
production per se, abstracting from changes in the valuation of stocks of
assets. I take Paul C.'s remarks in OPE-L: 4929 to lead in the same
direction. Alejandro's OPE-L:4941 seems to me also to be making this

The intuition I sense behind John's and to some degree Andrew's concerns is
that losses on assets due to price changes might offset profits generated
in production from the exploitation of labor. Furthermore, one could
imagine circumstances in which rapid improvements in capital productivity
would lead to large losses to holders of existing capital, despite a rise
in the underlying rate of profit on production. (Dumenil and Levy's data
suggests that if the Great Depression had something to do with the profit
rate, it was probably due to a _rise_ in the productivity of capital, and
hence probably on the underlying profit rate on production per se, leading
to a catastrophic revaluation of existing stocks of assets and a general
crisis of debt and liquidity.) I don't think this is what Marx was talking
about in his discussion of relative surplus value and the falling rate of
profit, though he clearly envisions the revaluation of existing stocks of
capital as one of the restorative aspects of crises.


Duncan K. Foley
Department of Economics
Barnard College
New York, NY 10027
fax: (212)-854-8947
e-mail: dkf2@columbia.edu