Some comments on the posts by Alan, Paul C., and Jerry (ope-l's 3188, 3196,
3197).
Jerry writes: "What I hear Alan suggesting is that we need to drop the
"assumption" of a general r. While it may or may not be the case empirically
that there is a general r, the issue in _Marx_ *clearly* concerns the law for
the GENERAL r to tendencially fall. If that "law" is to be attacked or
defended, it must -- at least initially - concern the analytics of changes in
the general r during the course of accumulation."
I think this evinces a widespread confusion concerning the category "general
rate of profit" (GROP). It is not synonymous with "equal rates of profit."
This is clear from _Capital_ III. Ch. 9 deals with the *formation* of the
GROP, while Ch. 10 deals with the "Equalization of the GROP." Quite
obviously, if the GROP is to be equalized, it must already be formed, exist,
prior to and apart from that equalization. Thus, what Ch. 9 deals with is the
determination of the general, i.e., weighted average, rate, and the *ideal*
prices associated with it, while Ch. 10 deals with the process by which
competition *tends* to equalize profit rates around this ever-present, always
existing, GROP.
BTW, this answers Bruce who, several months ago, said that when I talk about
"the" rate of profit, I am assuming profit rates are equalized. That isn't
necessarily the case. He also said that in every Kliman piece he ever read,
there was always an equalized profit rate. Again, not so---sometimes I (and
Ted) simply give the *ideal* prices associated with the GROP (we explicitly
state this in our paper in the Freeman/Carchedi book).
Alan writes: "In the last chapter of our book I tried to present a general
refutation
of the Okishio theorem for the case of general market prices and hence, any
general dispersion of profit rates."
I know what he means, but the theorem, especially in Roemer's versions,
*states explicitly* :) that an equalized (or "equilibrium") profit rate is
being assumed. To refute the theorem in the *narrow* sense, then, one *must*
also assume this.
Alan also writes: "My take on Andrew's Okishio refutation (as well as
Andrew/Ted
on transformation) has always been - and this has come up before - that the
equal profit rate assumption is strictly for the purposes of illustration, not
least because this assumption is made by everyone else in the debate. But I
don't think any of us think it is a necessary assumption and it certainly
isn't true in the real world."
I agree with the 2nd sentence. Concerning the transformation, Alan is also
basically correct. The key thing is to show that total price and profit equal
total value and total-surplus-value, respectively, and this needs to be shown,
and can be shown, independent of any assumption that profit rates are equal.
But to defend the logical coherence of Marx's own account, it needs to be
shown for the special case he assumed, in which profit rates are equal (again,
this was not a claim concerning real prices, but a statement concerning
*ideal* prices and associated profits). Alan is of course right, however,
that if one shows this in the general case, one has a fortiori shown it in the
particular case. He took the one tack, we the other. There's no
disagreement.
Concerning the tendency of the GROP, there's no need to assume equalized
profit rates, but again, to refute the Okishio theorem in the *narrow* sense,
one does need to assume it. Here I don't think it is sufficient instead to
show an FRP is *possible* in the general case (to show its *necessity* would
be sufficient); one needs to show that the equal profit rate case is among
those in which the FRP is possible.
The other aspect of the Okishio issue is this. Especially as Roemer poses the
issue, the assumption of an equalized profit rate is meant to duplicate Marx's
own assumption in *one* specific passage. Why? Because it is meant to be a
refutation of Marx's law of the FRP, not a mere result in its own right! Marx
says that the innovator's own profit rate initially rises, but as the new
technique becomes diffused, it becomes subjected to the general rate, which
falls. Now, this is not the same as assuming an equalized profit rate, as I
explained above. But Roemer is another one who labored under this
misconception. I could have made a point of this and said, "Marx doesn't
assume an equalized profit rate, so the theorem doesn't refute him except in
that special case," but what good would that have done? He would have
remained "refuted" in that case, and I would also have had to argue about a
fine point of interpretation ad nauseum. I think my procedure was more
robust: refute the theorem in the special case and *thereby* refute it in the
general case. (One counterinstance suffices to refute a general theorem.)
Paul writes: "What should be at issue is how useful his [Okishio's] theory is
as a predictor of what actually happens in capitalist economies. To assess
this one
has to judge the extent to which the preconditions of the theory actuall exist
in the real world, what testable predictions it makes about actual economies,
the extent to which the available historical evidence is compatible with its
predictions."
I think this seriously misunderstands the nature of the Okishio theorem. It
is not a "theory." It makes NO predictions concerning the tendency of the
rate of profit, even an equalized rate of profit. There is therefore nothing
to be tested empirically. Roemer is very clear about this, and quite right.
It is an "impossibility" theorem. It claims that no Okishio-viable technical
change can, by itself, lead to a fall in the equalized rate of profit (though
sometimes the word "equilibrium" is used instead of "equalized"). It thus
makes a claim concerning the CAUSES of any possible fall in the profit rate,
and a claim concerning the INFLUENCE OF TECHNICAL CHANGE on the profit rate.
If the theorem is true, then no amount of data could falsify it. If it is
false, then no data are needed. Whether technical change is or is not
Okishio-viable, whether or not profit rates are equal, is irrelevant.
One can, of course, ask whether Okishio-viable technical change can lower the
GROP when profit rates aren't equal. (One would have to assess viability
using actual current prices and profit rates, not equilibrium ones.) I doubt
one would get very far with that in a simultaneist framework. I think it may
be possible to squeeze out an FRP under certain assumptions, but I'm pretty
sure that the overall results would not substantiate Marx's claims concerning
the impact of technical changes that increase productivity and the organic
composition of capital on the GROP.
In any case, ignorance is not sufficient reason. I don't think it is helpful,
or true, to say that one doesn't know what can happen if profit rates are
unequal and therefore anything can happen. This confuses an ontological
question with an epistemological one.
Andrew Kliman