A reply to Allin's ope-l 1634, which comments on the Kliman-McGlone
refutation of Bortkiewicz's "proof" that Marx's Vol. III, Ch. 9
transformation disrupts balanced reproduction *because* input and
output prices are unequal.
The most important point that Allin makes, in my view, is that
"The (backward-looking) rate of profit is equalized, and balanced
reproduction is possible, as Andrew says, if we allow output prices
to differ from input prices."
Thus, Allin accepts that Ted McGlone and I have indeed refuted
Bortkiewicz's alleged "proof" that balanced reproduction requires
stationary prices. I dwell on this because, firstly, the main
purpose of the illustration Allin discusses was indeed to refute
this "proof," and secondly, because this "proof" has been taken on
faith for so long that it has become the 11th Commandment--Thou
Shalt Not Let Input and Output Prices Differ. As I noted in a few
posts about 2 months ago, when Sraffians claim that their stationary
prices are the *unique* set of prices that ensure the reproduction
of the economic system, they're making the exact same error as
Bortkiewicz did.
I salute Allin for what Marx might have called his "scientific honesty"
on this question!
(Unfortunately, Allin wasn't at the EEA session where all this was
discussed, but had he been there, he would know that it "was like
pulling teeth" to get others to acknowledge that we have indeed
refuted Bortkiewicz's "proof.")
Some comments with respect to other aspects of Allin's post:
(1) The discussion of how TSS values differ from Leontiev's
vertically integated labor coefficients (VILCs), it should be
noted, is "orthogonal" to the issue of whether Bortkiewicz is
refuted. That is, whether or not one accepts the TSS interpretation
of value in Marx's work, input and output prices do NOT have to be
the same for uniform profitability, sales = purchases, or balanced
reproduction. (Call the TSS values "prices that differ from values":
they are still prices, and even though output prices differ from input
prices, reproduction takes place with a uniform profit rate.)
(2) I haven't checked Allin's period 3-through-21 numbers, though they
definitely look right, and I accept them as a basis for discussion.
(3) Allin's "Contention" from the numbers is that the TSS values, after
the initital period, are "disconnected from [their] original definition
as labour-time embodied (or required for reproduction)." This
contention repeats the contention made by Allin, Paul C., and Gil in the
debate they had with me throughout most of February. Do we need to
re-hash this? If so, I'd like Allin to answer the following: aren't
the c's of all periods sums of labor-time (by virtue of the total
price/total value equality of the prior period)? Aren't they required
for reproduction of the commodities these means of production are used
to produce? Isn't this sum of labor-time embodied in the product by
being transferred to the value of the product from the value of the
constant capital?
(4) Allin writes: "... once the transformation is complete -- in the
equilibrium achieved in table 21 ...." I maintain that *Marx's*
transformation is complete is *each* period. A uniform profit rate is
formed and, as Allin has acknowledged, supplies can equal demands and
balanced reproduction can take place under the output prices of *each*
period. Did *Marx* maintain that the transformation is complete only
when input prices equal output prices?
(5) I have nothing against iterating the damned thing. Indeed, my 1988
paper did so, until stationary prices were achieved. But then almost
*everyone* thought we had an "iterative solution" to the "transformation
problem," that we were closet "neo-Ricardians," etc. ("We" refers to
Ted McGlone and I). The point is that, whether or not one iterates ad
nauseum, *each* period's prices of production are "correct"; they give
uniform profitability and permits supply = demand and balanced reproduction.
(6) As we noted in the 1988 paper, if one understands each period to be
a real-time period (which does not imply that capitalists would actually
behave in the way the iterations force them to), then it makes no sense
to say the 21st period is "the solution." Nor is it meaningful to
compare the initial period with the last one, as we also noted. They
are DIFFERENT periods, and there's absolutely no reason why the total
price of 1996 should equal the total value of 1976.
(7) I'm delighted to be told (by Allin) that the TSS output values are
a "hybrid." Bruce Roberts was long ago told this by Steve Rankin, and
I was feeling slighted that the TSS interpretation didn't get the same
label. Apart from the notion that c is not a value sum, I personally
have no problem with the "hybrid" term. Yes, total value is a hybrid
of dead labor transferred and living labor newly added. (Aside to
Bruce: if they start charging us with race-mixing, then I'd worry :).
(8) Since, again, each period is in *real time*, it is simply incorrect
to say that we (or the TSS equations in general) choose a standard of
price to close the system. and since we have a SINGLE SYSTEM of values
and prices (which is what the SS in TSS means), the whole idea of
normalizing the price system to obtain an invariance with the "value"
system is meaningless.
Of course, one is "free" to interpret the numbers to mean something
different from what the authors specify. This was also David Laibman's
approach at the conference. For instance, I can say that v = L(I-A)**-1
is really not a value equation at all, but a cake recipe. "They
call 'v' a vector of 'values', but it really means 'sift 3 cups of flour
into a large bowl.' '=' means 'add two large eggs'. They claim that
L is a vector of labor coefficients, but it really means 'cream in 3
tablespoons of unsalted butter' ...."
I'll accept others' interpretations of my equations/illustrations instead
of what I specify them to mean the moment Allin and David Laibman
admit to being bakers and not economists.
(9) Allin writes "one should proceed until ... a consistent set of prices
is achieved (step 21)." What is "inconsistent" about the prices of
earlier periods? BTW, the FORMATION of a general rate of profit and the
EQUALIZATION of the profit rate are, for Marx, 2 different things (you
and Paul C. might like this, Allin). The former is a theoretical
construct. One takes s/(c+v) for the total social capital and allocates
profit (over cost-price) to each branch on that basis. Thus, a general
rate of profit can be FORMED in each period even if profit rates are not
equal in every period (or in any period).
(10) Finally, Allin notes correctly that revenue/outlays are not equalized
unless prices are stationary. (He calls revenue "disposable profit.")
He suggests that stationary prices should therefore "be seen as part of
the specification of equilibrium." Response: there can be different
meanings of "equilibrium." There doesn't have to be a general, all-
purpose one. Equalization of revenue/outlay has little or nothing to do
with equalization of profit/outlay. I know of nowhere where Marx ever
said that revenue/outlay tends toward equality, and I'm very sure that
Marx didn't think capitalists try to maximize revenue/outlay (so that
competition then tends to level out the different rates). Why?
Because maximizing revenue/outlay is tantamount to saying that capitalists
seek to maximize luxury consumption, that their goal is to consume and
not to amass surplus-value. For more on this, see the discussion between
Michele Naples and I in one of the 1993 Capital & Class issues (# 51?).
Andrew Kliman