> Date: Tue, 30 Mar 1999 09:47:01 -0500
> From: Andrew Kliman <Andrew_Kliman@email.msn.com>
> Reply-To: ope-l@galaxy.csuchico.edu
> To: ope-l@galaxy.csuchico.edu
> Subject: [OPE-L:801] Marx's concept of price of production
>
> In OPE-L 748, I noted that
>
> "in a paper presented at the panel Fred mentions, Alan Freeman
> *demonstrated* that simultaneously determined "prices of production"
> CANNOT function as long-run centers of gravitation. He also
> *demonstrated* that temporally determined production prices DO
> function as long-run centers of gravitation. Since -- as Fred
> himself admits here -- Marx's prices of production are centers of
> gravitation, they are thus temporally determined, not simultaneously
> determined. End of story.
>
> "What is your response to this, Fred? It seems to me that you must
> do one of the following:
>
> (a) find an error in the demonstration
> (b) renounce Marx in favor of simultaneous determination
> (c) renounce simultaneous determination in favor of Marx
>
I finally had a chance over the weekend to look at Alan's
³"demonstration that simultaneously determined prices CANNOT
function as long-run center-of-gravity prices." My response is:
"error in the demonstration."
Alan's argument does not demonstrate that simultaneously determined
prices CANNOT FUNCTION AS LONG-RUN CENTER-OF-GRAVITY PRICES.
All Alan's argument demonstrates is that simultaneously determined prices
CANNOT ACTUALLY EXIST AS MARKET PRICES IN TWO SUCCESSIVE PERIODS WHEN
THERE IS TECHNICAL CHANGE. But these are not the same thing. Long-run
center-of-gravity prices do not have to actually exist in every period in
order to function as long-run center-of-gravity prices. Indeed, it
is in the nature of long-run center-of-gravity prices that they do not
exist every period, especially when there is technical change. Further
argument below.
First a brief summary of Alan's main argument (as I understand it):
1. Alan assumes that the Sraffian equilibrium prices are the actual market
prices in each period, and thus that actual input prices = actual output
prices in each period.
2. Thus, when there is technical change in Dept. 1, both input prices and
output prices of Dept. 1 will change. Input prices and output prices in
period 2 (after technical change) will equal to each other, but they will not
be equal to the input and output prices in period 1.
3. It follows that the output prices of period 1 are not equal to the input
prices of period 2.
4. But, Alan argues, this cannot be. There CANNOT BE AN ACTUAL
EXCHANGE in which the price paid by the buyer (the input prices in
period 2) is NOT equal to the price received by the seller (the output
prices in period 1). Therefore, in the case of technical change, actual
market prices cannot equal Sraffian equilibrium prices in successive
periods.
MY RESPONSE:
1. First of all, Alan's argument is entirely in terms of Sraffian
equilibrium prices, according to which the rate of profit is determined
simultaneously with individual prices and both are derived from given
physical quantities of inputs and outputs. Andrew applies Alan's
argument in terms of Sraffian prices to my interpretation of Marx's
concept of prices of production. Therefore, once again Andrew is
equating the Sraffian interpretation and my interpretation. Once again,
I insist that my interpretation is different from the Sraffian
interpretation, for reasons that I have already discussed (luxury goods,
fixed capital, etc.). Therefore, I argue that Alan's argument in term of
Sraffian prices does not apply to my interpretation of Marx's concept of
prices of production.
2. Beyond that, Alan's argument is not even correct for Sraffian
equilibrium prices, which are long-run equilibrium prices, not actual market
prices (and therefore is not correct for my interpretation of Marx's
concept of prices of production, for the same reason). Alan's argument
assumes that Sraffian equilibrium prices ARE ACTUAL MARKET PRICES THAT
EXIST IN EVERY PERIOD. (This is the same mistake that Alan made in
his interpretation of Ricardo's "natural prices" that I discussed in
Boston.) What Alan actually "demonstrates" is that, when there is
technical change, Sraffian equilibirum prices CANNOT ACTUALLY EXIST as
market prices in successive periods. But this is not the same thing as CANNOT
FUNCTION AS LONG-RUN CENTER-OF GRAVITY PRICES in successive periods.
Even though Sraffian equilibrium prices cannot actually exist as market
prices in successive periods when there is technical change, they can
nonetheless function as long-run center-of-gravity prices around which
actual market prices fluctuate, both before and after the technical change.
Long-run center-of-gravity prices seldom actually exist as market prices.
That is the nature of long-run center-of-gravity prices - that they do not
actually exist as market prices in every period, but exist only as the long-
run average of actual market prices. And, when there is technical change
(as in Alan's example), long-run center-of-gravity prices will never exist
as actual market prices in the next period after the technical change.
Technical change creates disequilibrium, which initiates a process of
adjustment to the new equilibrium. But these new equlibrium prices can
still function as long-run center-of-gravity prices in subsequent periods,
even though they cannot actually exist as market prices in the next period
after technical change.
All Alan's argument "demonstrates" is that actual market prices will not
immediately jump from the old equilibrium price to the new equilibrium
price from one period to the next when there is technical change. Even
assuming that the old equilibrium price was the actual market price before
the technical change, the adjustment to the new equilibrium price will take
time. The actual market price will not immediately jump to the new
equilibrium price in the next period. But the new equilibrium price will
still function as long-run centers-of-gravity around which actual market
prices fluctuate in subsequent periods.
3. The second part of Alan's argument modifies the Sraffian determination
of prices and the rate of profit to allow for input prices equal to output
prices and shows that this modification leads to contradictory results -
either rates of profit across sectors will diverge or Sraffian prices will
diverge from actual prices. However, this second argument presumes that
Alan's first argument is correct. Since Alan's first argument is
incorrect, we don't need to concern ourselves with his second argument.
4. Therefore I conclude that Andrew's statement in the passage quoted at
the beginning of this post is wrong. Alan's argument does not demonstrate
that simultaneously determined prices cannot function as long-run center-
of-gravity prices. All Alan's argument demonstrates is that
simultaneously determined prices cannot actually exist as market prices in
two successive periods when there is technical change. But long-run
center-of-gravity prices do not have to actually exist in every period. Alan
and Andrew seem to be assuming that in order to function as long-run
center-of-gravity prices, Sraffian equilibrium prices must actually exist in
every period. But this is a misunderstanding of the nature of long-run
center-of-gravity prices. It is the nature of long-run center-of-gravity
prices that they DON'T exist in every period. They are the hidden
gravitational poles toward which actual market prices are drawn through
the equalization of profit rates across sectors, i.e. they are long-run average
of actual market prices, not the actual prices in every period.
5. Indeed, I have argued that that essentially the opposite of Andrew's
argument is true. Only simultaneously determined prices (i.e. prices in
which input prices are equal to output prices) can function as long-run
center-of-gravity prices, at least with respect to Marx's concept of price
of production.
I have argued (with substantial textual evidence) that one of the
characteristics of Marx's concept of price of production is that they
change if and ONLY IF productivity or the real wage changes. This
characteristic is contradicted by the temporal determination of prices of
production, which keep changing every period, even though productivity
and the real wage remain constant, as Andrew and Tedıs numerical
examples clearly show. Therefore, the temporal interpretation of Marx's
concept of price of production cannot be correct. Instead, this fourth
characteristic of Marx's prices of production is satisfied ONLY IF input
prices are equal to output prices, i.e. only if input prices are valued in
current reproduction costs. In other words, Marx's prices of production,
as long-run center-of-gravity prices, require that input prices are valued in
current reproduction costs, and in this sense require simultaneous
determination, and are incompatible with temporal determination (which
results in prices of production changing every period, even though
productivity and the real wage remain constant).
I look forward to Andrew's and Alan's responses to my arguments and to
further discussion.
I also hope that others will comment on this debate - Alejandro R.,
Abelardo, David L, and others. Does anyone else think that Alan has
demonstrated that simultaneously determined prices cannot function as
long-run center-of-gravity prices? (the argument is on pp. 13-16 of his
paper in the Boston conference volume. Does anyone else agree or
disagree with my argument that Marx's concept of price of production
requires that input prices are equal to output prices?
Comradely,
Fred