At long last I return to John's challenge #2 with respect to my post
1492 on the TSS interpretation re Marx's theory of the tendentially
falling rate of profit. I apologize to John and Andrew, whom this
post also addresses, for the long lag. It's end of semester time and
very hectic--OPE-L participation will require snatching odd bits of
time, at least until mid-May.
Anyway, John's point #2:
> 2. Again, with reference to Andrew's post, Gil writes:
>
> "...as I understand Andrew's argument, it requires that
> capitalists calculate rates of profit on the basis of *historical*
> rather than *replacement* costs of capital (i.e., that capitalists
> fail to ignore sunk costs). [Please correct me if this understanding
> is wrong, Andrew.] Now, I doubt that this is the case, but that's
> really beside the point: whether or not capitalists ignore sunk
> costs is an empirical issue that does not depend on the comparative
> *theoretical* fidelity of the TSS system to Marx's conception."
>
> I assume all of us mean the same thing by "sunk costs."
Unfortunately, [blush], I didn't mean "sunk costs" in the standard
sense here. But some preliminary spadework is necessary before my
use of the term here can be made clear.
> Is the
> idea that a capitalist measures his rate of profit by measuring
> the profits made against the capital advanced an empirical issue?
> Or, perhaps, we are excluding "sunk costs" from capital advanced?
I'm going to come at this indirectly. I went back to Andrew's 1988
RRPE article to reconstruct my argument and discovered that the source
of my questions began earlier than the preceding account suggests.
Answering these questions will render the above either redundant or
irrelevant (I think), so if John and/or Andrew don't mind I'd like to
re-start the dialogue with two more basic issues.
1. Rates of growth of output, fixed and circulating capital, and
living labor used in production are taken as exogenous (p. 284).
This implies that the rate of capital accumulation is constant, or at
least independent of the rate of profit. But this rules out by fiat the
"self-correcting mechanism" Marx identifies in section 1, Ch. 25 of
Volume I. It seems to me that this procedure is doubly suspect:
first, because it eliminates a possible countervailing tendency
without any justification, and second, even if Marx's Ch. 25 account
weren't there, the corresponding microeconomics are problematic: why
*isn't* the rate of accumulation sensitive to the rate of profit?
2. Price determination: Andrew writes "To determine the path that
the unit price, p(t), takes over time, Marx's concept of price
formation is adopted. He holds that the total value of output is the
sume of the value of the constant capital, plus the value added
(labor-time extracted from workers)."
This strikes me as a _non sequitur_, since the purpose is to explain
price formation, not value formation. Any connection between the two
must be established, not assumed. At the very least, it should be
*possible* to determine prices independently of values, even if this
process is held to be "superficial". Otherwise one runs the risk of
imposing conditions which make no sense from the standpoint of the
logic of exchange (i.e., the world of relative prices).
Another warning bell goes off when I read this: this is a one-output
model. Therefore I expect the unit price of the one output good to
be indeterminate, i.e. absolutely arbitrary. So I don't see how it
can have any particular "path".
To me, these points are illustrated quite forcibly in equation (6).
Take for example the expression for t = 1, which simplifies as
follows (I think: Andrew's expression for pi has an extra
parenthesis, so I can't tell whether -a goes in the denominator of
n/[d/b] or not; on contextual grounds I assume it does, but please tell me
if otherwise): using m for "mu",
p(1) = p(0) [A(0)/Q(0)] + N(0)/mQ(0).
This formulation strikes me as problematic for at least 3 reasons.
A) I'd expect production prices to depend on the wage rate and an
imputed rate of profit, but these don't show up here. Why not? I
note that if the 2nd term on the right hand side included a money
wage rate term, then the p(0) term could at least be isolated on the
right-hand side. More on this in a second.
B) Second, I don't see why m, the "value of money" has anything
whatsoever to do with price determination. At the very least
something problematic is being smuggled in with this stipulation.
C) There is no uncertainty in the model, so the price path is
absolutely determinate. In that case I'd expect futures markets
(arbitrage across time) to equate prices across periods, up to a
discount factor. If individuals do not discount the future, then
prices should be equal in every period, or the model is economically
nonsensical
[Scenario: you know you can buy a commodity today for $5 and sell it
for $10, and you don't discount the future. Barring laziness or
scholarly disinterest, why wouldn't you arbitrage?]
In other words, the price formation story has been forced onto the
Procrustean bed of a (particular) value story. But even if I
accepted the latter on its own terms (which, er...), I wouldn't
necessarily be prepared to accept the implicit microeconomics of the
latter as applied to the world of price formation. For the reasons
given above, I think the relevant microeconomic underpinnings are
dubious at best.
So how am I wrong?
In solidarity, Gil