James Miller jamiller at
Thu Nov 9 21:22:37 MST 1995


   Responding here to John Ernst's post of Nov. 6.
He starts off with a demonstration of the Sraffian
(Ricardian) approach to "value":

   9 units + 150 labor hours --->  15 units
   9 units + 3 units + 3 units  =  15 units
   225 hours + 75 hours + 75 hours = 375 hours
    c        +    v     +     s    =   w

   All are variations on the same theme. According
to this method, you don't need the distinction
between use value and exchange value to represent
commodity production in a quantitative way.
   John, not a Ricardian, takes Marx's analysis of
the commodity as his starting point. He says that
it's hard to show a falling profit rate if you use
the Ricardian method (while assuming a constant
real wage). Ricardo could only show a falling profit
rate based on rising wages (specifically, the rising
value of corn).
   Well, if you see that the value of any particular
commodity changes as the labor time required for its
production changes; that's a start. Then you see that
technological innovations cause socially necessary
labor time to fall for all affected commodities.
Then you recognize that productivity, in physical
output per hour, increases with technological
innovations. Workers process more raw material per
hour. The machinery is more massive in relation to
the labor necessary to run it, etc. The value added
by the active labor force gets smaller and smaller
in relation to the value of the capital advanced.
You end up with a falling profit rate.
   John says that he was arguing with Keen  at the
level of capital accumulation, not the level of the
labor theory of value. I had said that the argument
should go to the most basic level of disagreement.
John focused rather on the consequences of such
disagreement at a higher level of development.
   I'll let the subscribers to the list decide
which was more effective or educational.


   Then in the post of Nov. 7, John responded to my
post on TRPF. He raised the argument that if the
increases in productivity cheapen both constant
and variable capital, the profit rate will not
decline. Here's how Marx put it:
   "Outside of a few cases (for instance, if the
productiveness of labor uniformly cheapens all
elements of the constant, and the variable, capital),
the rate of profit will fall...". (Vol. III, p. 222
or 226, etc.)
   The profit rate, in a case where the relative
surplus value increases, due to productivity, can
certainly rise. Likewise, in a case where the value
of constant capital goes down while all else remains
the same, the profit rate will rise as well. You
can imagine, then, that in a case where the rate of
surplus value increases at the same time that the
elements of constant capital are cheapened, you would
have a much greater likelihood of a profit gain.
   And the amazing thing is that growing productivity
has a continuous and long term upward effect on profits,
as does the cheapening of the elements of constant
capital. You would think that, in case after case after
case, these effects would produce rising profits. And,
indeed, that is their effect. But at the same time,
rising productivity, which is the cause of these
profit-boosting effects, also causes the organic
composition of capital to rise, lowering the profit
   So the problem is that the same process that raises
the profit rate also lowers it. See Vol. III, p. 229 or
233(?): "But since the same influences which raise the
rate of surplus-value (even a lengthening of the
working-time is a result of large-scale industry)
tend to decrease the labor-power employed by a certain
capital, it follows that they also tend to reduce the
rate of profit and to retard this reduction." Then
2 pages later: " follows that they also tend to
reduce the rate of profit and to retard this reduction."
   One and the same process has counter-directional
effects on the rate of profit. This is the crux of
the problem. It has to be reckoned with, or it's just
not possible to say anything at all conclusive on
the tendency of the rate of profit to fall. I cannot
stress overmuch how critical it is to grasp this.
   So the answer to the question first posed: if you
have a "case" (i.e., a temporally circumscribed event
within a historical process) where the profit rate is
boosted by a simultaneous devaluation of constant and
variable capital, does this mean that the TRPF is in
operation? Answer: yes. The force that causes the
profit increase in this case is the same force that
causes the long-term decline in the profit rate.
   I won't say anything further on this now, except
that I deal with this in my paper posted on theinternet
at miller. This
paper doesn't add anything to what Marx said in Chap.
13 and 14. Its premises and conclusions are identical
to those of Marx. The only thing my paper adds is a
specific refutation of Sweezy, and, by implication,
all those who reproduce his method of misunderstanding
Marx's thinking on the TRPF.


   John had asked about the relation between technical
composition and productivity. I said that this is a
technical question. Some innovations improve productivity
a lot, others only a little. It depends on the specific
character of the innovation. John says that I didn't
answer the question. He had asked, which is increasing
faster, the inputs or the outputs? But again, this is
a technical question. In the case of machines which only
increase the rate of raw material processing per labor
hour, obviously the inputs and outputs increase to the
same degree, not counting the increase in the machinery.
But this usually requires an increase in the mass of
machinery in relation to the labor force. Thus here
the input would increase more than the output, as far
as that particular firm is concerned. And the same
would be true of the total social capital. The input
increases more than the output in relation to the
labor force, in physical terms, because in addition
to the faster processing of raw materials, there is
additional machinery and equipment per worker.
   A related issue is the ratio of the annual physical
output to the existing mass of the means of production.
As extractive and productive machinery becomes more
and more efficient, the output of products grows in
relation to the mass of the productive apparatus. (See
Marx's quote from Thompson, Vol. II, Chap. 17.) So I
would say that the tendency is for the input (in
material) to grow more than the material output, for
the total social capital; but that the mass of annual
physical output grows also in relation to the mass of
the existing means of production. But I'd like to hear
what John has to say about this.
   In any case, whether the inputs grow faster or
slower than the outputs, in material terms, this
will have no effect on the tendency of the rate of
profit to fall. The only necessary condition for
the rate of profit to fall, in the long run, is that
the productivity of labor keeps increasing. This
will produce an increasing technical, and therefore
organic, composition of capital.


   John says: A machine "embodies" 50 hours of labor and
costs $500. Production commences with this machine.
Meanwhile, a new machine is made which would embody
25 hours of labor and would cost $250 (all else the
same). Is the capitalist using the costlier machine
"out $250"?
   Let's assume we're talking about a typical capitalist
functioning under normal conditions of competition. Then
it is an everyday fact of life that newer, better and
cheaper machines are being made and are appearing in the
market, and our capitalist is always calculating and
recalculating whether it would be better to keep the old
machine running, or replace it.
   Our capitalist's machine, like that of the other
capitalists in his branch of the economy, begins to lose
in value as soon as the cheaper and better machines
begin to go on line and begin putting out product. (And
this aside from the value of the machine being
transferred to the product.) Moral depreciation begins
to become a fact of life.
   Maximizing profit means getting the most out of your
current equipment, while buying the new at the earliest
time. This is one of the reasons capital tends to run
up against the barriers of circulation, creating
generalized overproduction. But there's no theoretical
problem here. The goal is to minimize production costs,
and to do this the capitalist has to make decisions
regarding purchase of plant and equipment accordingly.
   The unlucky capitalist would be the one who has to
buy a new machine, and it turns out to be the last one
to come off the line before the newer, cheaper and
better models go into production. He buys the old
machine in July, and the new machine comes out in
   At the other extreme, you have the capitalist who
buys the new machine before anybody else does, and
gets the most out of it before it has to be replaced
by the next generation. In between the two extremes
you have the typical capitalists who pay average
production costs.
   Of course, it's not primarily a matter of luck, but
of knowledge about the trends going on among the
companies that make the machines you're going to
buy. This is always a challenge. Sometimes there's
an inside track, monopoly agreements, etc. Sometimes
you buy out the machine maker.
   John is concerned that moral depreciation is
running amok, and capitalists are losing big time
as their machines become obsolete almost overnight.
He says, "it would seem that assigning values in
usual fashion would force us to take up our erasers
in earnest as replacement costs fall."
   He could be right. But this is something that
has to be subjected to empirical measurement. This
is one of the ways (but not the main one) in which
labor is wasted under capitalism. A lot of labor
time goes into making a machine, then it goes to
the scrapyard halfway through its effective life,
just because it's cheaper for the capitalist to
replace it.
   How many perfectly good 386 PCs are thrown on
the trash heap every day by major corporations
and replaced with Pentiums? It's a loss for sure.
   But there are other ways to demonstrate the
wastefulness of capitalism, some a bit more
dramatic. No need to drag this out.


   John said that Sweezy used the term "machino-
facture" in a 1974 argument against Marx's theory
of the falling profit rate. Sweezy said that the
manufacture of machines by machines cheapens the
elements of constant capital so fast that the
rate of profit does not fall. As you can see by
what I said above, this argument has some merit.
All it is, is a variation on Sweezy's earlier
argument that the cheapening of the elements of
constant capital offset the falling profit rate
   The mechanization, computerization, etc., of
manufacturing technique doesn't change anything
about the tendency of the rate of profit to fall.
You still have an increase in the value of the
means of production in relation to the new value
created daily in production. Of course the mass of
the means of production grows at a faster rate
than its value, because its elements are
constantly being devalued as they grow in mass.
   As far as "machinofacture" is concerned, it is
a term that is superior to "manufacture," because
"manu" is Latin for hand, and the machine has
replaced the hand. But I don't think the term will
catch on.

Jim Miller

     --- from list marxism at ---


More information about the Marxism mailing list