A reply to Fred's ope-l 2699.
I had written: "The inequality of input and output prices in these examples
in caused by a previous inequality between input and output prices (beginning
in the initial period).
Fred has responded: "What exactly does this mean: "a previous inequality
between input and output prices"? According to KM's logic, the input prices
in the initial period are taken as given prior to production and thus prior to
the determination of output prices. At the time these input prices are taken
as given, they cannot be unequal to output prices because output prices (the
other side of the inequality) do not yet exist."
Andrew: This is quite right. What I should have said is this: in all
periods *but the initial one*, the inequality of input and output prices is
caused by an inequality (in earlier periods) between input and output prices
(given that physical quantities are the same).
So what determines the inequality between input and output prices in the
initial period? Fred says:
"As a result of this transformation of output prices, input prices will not be
equal to output prices."
Andrew: I have already shown that this is false. I showed that it is the
difference between input and output prices that makes prices and the general
profit rate change (given the same physical quantities), and that the
transformation of COMMODITY VALUES into production prices is neither necessary
nor sufficient for these changes to occur. Now, what this also implies is
that the transformation is neither necessary nor sufficient for input prices
to equal output prices.
Not sufficient: assume unequal compositions of capital, and equal rates of
exploitation, so that production prices don't equal values: there is a
transformation. But prices can still be stationary. Example:
industry means of production real wages living labor
output
steel 5 steel 2 corn 5
10 steel
corn 5 steel 4 corn 10
15 corn
Assume 1 unit of labor = $1, and that the input price of steel = $1.20, the
input price of corn = $1.00. Then we have
industry C V S C+V+S PP Av. Profit
Unit PP
steel 6 2 3 11 12
4 12/10 = $1.20
corn 6 4 6 16 15
5 15/15 = $1.00
total 12 6 9 27 27
9
Input and output prices are equal, despite the transformation, so the
transformation isn't sufficient to produce nonstationary prices.
Not necessary: assume equal compositions of capital, and equal rates of
exploitation, so that production prices equal values: there is no
transformation. But prices can still be nonstationary. Example:
industry means of production real wages living labor
output
steel 5 steel 2.5 corn 5
10 steel
corn 5 steel 2.5 corn 5
10 corn
Assume 1 unit of labor = $1. The *only* initial input prices which will
result in production prices (of outputs) equal to these initial input prices
are: input price of steel = $1.00 and input price of corn = $1.00. All other
sets of initial prices will result in nonstationary prices, given uniform
profitability. For instance, assume that the initial input price of steel =
$1.20 and the initial input price of corn = $1.20.
Then we have
industry C V S C+V+S PP Av. Profit
Unit PP
steel 6 3 2 11 11
2 11/10 = $1.10
corn 6 3 2 11 11
2 11/10 = $1.10
total 12 6 4 22 22
4
Or assume that the initial input price of steel = $1.20, but that the initial
input price of corn = $1.00. Then we have
industry C V S C+V+S PP Av. Profit
Unit PP
steel 6 2.5 2.5 11 11
2.5 11/10 = $1.10
corn 6 2.5 2.5 11 11
2.5 11/10 = $1.10
total 12 5.0 5.0 22 22
5.0
In both these examples, input and output prices are unequal, and in the latter
example, there is also a difference between the *relative* input and output
prices. This occurs even though there is no transformation, so the
transformation isn't necessary to produce nonstationary prices.
Hence, the transformation of values into production prices is neither
necessary nor sufficient for nonstationary prices, and therefore neither
necessary nor sufficient for changes in production prices and the general
profit rate over time. In other words, the transformation is not a cause of
either, not even in the "initial period." No talk of new examples vs.
previous examples can alter these facts. Because transformation is neither
necessary nor sufficient, it cannot be a *cause* in *any* example. In some
examples it merely ACCOMPANIES a difference between input and output prices
(and thus accompanies changes in production prices and the general profit rate
over time), that is all. And since transformation is never a cause, in any
example, it doesn't matter at all which example comes first historically or
what is published. Again, I can produce example after example, and even have
these examples precede contrary ones, in which I command, "Let There Be
Light!", flick on the light switch, and have the lights go on. But since my
command is never necessary or sufficient for the lights to go on, then in NO
example can it be a cause, and it matters not a whit whether the examples in
which I do command "Let There Be Light" precede the ones in which I don't
historically, or whether I publish the results in the former cases and not the
latter.
Let me pose the matter even more simply. Assume a set of input/output
relations and uniform profitability. Now, Fred, can you tell me the
conditions under which input and output prices will be equal, and the
conditions under which they will be unequal? Can you do so WITHOUT knowing
anything about the values of individual commodities, about differences in
value compositions, about whether values equal or don't equal prices, and thus
about whether there is a transformation of values into production prices or
not? If you can do so, then this means that transformation is ENTIRELY
irrelevant to whether input and output prices are equal (and thus to whether,
given the same input/output relations and the same profitability assumption,
output prices and the general profit rate will change over time.
Assume, for instance, the following:
industry inputs
output
steel 15 steel 50 corn 30
steel
corn 9 steel 30 corn 90
corn
you may, if you wish, assume real wage advances are included among the inputs,
but I'm not going to let you know the distribution of inputs between means of
production and real wages. Nor am I going to let you know the amounts of
living labor used in either industry. The above is all the information you
have, and it is all the information you need. If you can't solve the problem
with this information, you can't solve it at all, because all other
information is completely and utterly irrelevant.
(The answers you should get are as follows. Assuming a uniform profit rate,
for the relative input price to equal the relative output price, the input
price of steel must be exactly 5 times the input price of corn (a relative
price of 5). Any other relative input price will give a relative output price
that differs from the relative input price. For the *absolute* input prices
to equal the *absolute* output prices, you must have a relative input price of
5 and, in addition, a profit rate of 20%. Where does this 20 ome from? In
Marx's theory, it depends on the initial capital advances, the distribution of
advances between constant and variable capital, and the amount of living labor
extracted, so it seems as though it might depend on knowing something about
price/value differences and thus it seems as though it might depend on
transformation. But this is not the case: the level of the general rate of
profit, as you know and agree, is determined independently of the
*distribution* of living labor across the economy; it depends only on the
aggregate amount. Hence, living labor in this economy could be distributed
such that prices = values or prices don't equal values. For a profit rate of
20%, all that matters is that *aggregate* surplus-value is one-fifth the size
of the initial advance of capital.)
Therefore, as I've noted several times already, and will keep doing so until
I'm blue in the face if need be, holding my breath throughout the summer until
Fred is back on ope-l, if need be, the following is simply not correct:
"Therefore, it makes no sense to say that the continuing change in prices of
production and the rate of profit in KM's published articles are caused by
the inequality of input prices and output prices, and not by the
transformation of output prices. Because the former are caused by the
latter.
"Therefore, KM's interpretation of the transformation process as presented in
their published articles requires multiple periods, as evidenced by the
continuing changes in prices of production, which are caused by the
transformation of output prices."
Fred also makes a perplexing statement:
"In recent ope-l posts, Andrew has introduced an entirely new element to
explain the inequality of input prices and output prices and thus to explain
the continuing changes in prices of production and the rate of profit:
MARKET PRICES, which are not equal to prices of production. Market prices
are never mentioned in this sense in KM's published articles; it is never
mentioned that the input prices might be equal to market prices which are
not equal to prices of production. Andrew has recently introduced this
assumption in response to my criticisms of their published articles."
Fred, please see our 1988 article, p. 72: "we must therefore begin ... with
given input prices. ... _Solely_ in order to facilitate comparison with
'transformation problem' 'solutions', ... initial values are equal to the
values [not production prices-AJK] of means of production and labour-power."
Please also see p. 41 of our article in _Marx and Non-equilibrium Economics:
"the collective capitalists ... purchase means of production (MP) and labour
power (L), at given prices representing given values. (Any initial values
could be assumed. _Solely_ to facilitate comparison with transformation
problem solutions, the initial values here equal the values [not production
prices] of the means of production and labour power)."
This seems clear enough to me, not only because we assume in both examples
that the initial prices equal the values and not the production prices of MP
and L, but also because we take the initial values and prices as GIVEN, i.e.
data, i.e., they are what they are, period.
Fred continues:
However, I know of no textual evidence to support this new interpretation
[...;] he never said that the prices of inputs in his theory of prices of
production might be market prices."
Fred's view at first seems plausible, because there are places in which Marx
does say that the cost-price is determined by the production prices of means
of production and subsistence, just as there are places before Ch. 9 of Vol.
III in which he says that it is determined by their values. But both kinds of
statements are just simplifying assumptions that do not exhaust his theory.
First, note that in the beginning of Ch. 12, he discusses a change in the
general profit rate as one cause of changes in production prices. He
explicitly refers to a "depression of wages below their normal level" as a
cause of the change in the profit rate (p. 307, Vintage). Similarly, all of
Ch. 11, concerns the impact of changes in "wages" (not the value of
labor-power, not the production price of means of subsistence, etc.) on prices
*of production*, and also here considers the case in which "wages previously
stood above the normal price of labour, or if they are now to be pushed below
it." In addition, on p. 1001 (Ch. 50), Marx notes that "If the commodity with
the monopoly price is part of the workers' necessary consumption, it increases
wages and thereby reduces surplus-value, as long as the workers continue to
receive the value of their labour-power."
Moreover, all of Ch. 6 concerns the impact of changes in prices on the general
profit rate. As I have just mentioned, changes in the general rate affect
production prices, and changes in input prices are shown in Ch. 6 to affect
the general profit rate. Hence, variations in input prices affect production
prices. But perhaps Marx is excluding all changes in input prices except
those caused by changes in production prices? No. On p. 208 (section 2 of
Ch. 6), he writes: "But since we are dealing here with the effect that these
price fluctuations have on the profit rate, IT IS ACTUALLY A MATTER OF
INDIFFERENCE WHAT THEIR BASIS MIGHT BE. THE PRESENT ARGUMENT [that input
price changes affect the general profit rate] IS JUST AS VALID IF PRICES RISE
OR FALL NOT AS A RESULT OF FLUCTUATIONS IN VALUE, BUT RATHER AS A RESULT OF
THE INTERVENTION OF THE CREDIT SYSTEM, COMPETITION, ETC." [emphases added].
The reference to competition could conceivably be construed as referring to
production prices only, but certainly not the reference to the credit system.
As is well known, Marx argued that interest rates are determined by supply and
demand, fluctuate daily on the basis of supply and demand factors, *and* that
there is no "natural" interest rate. (See Ch. 22). Hence, supply of and
demand for credit can affect input prices, which alter the general profit
rate, which alters production prices, according to Marx's theory. And in
general, it is a "matter of indifference" what causes the input prices to
change; when they change, so does the profit rate and thus so do production
prices.
These observations also, BTW, completely demolish the view that Marx's
production prices are some long-run equilibrium (Moseley) or "long-period"
(Garegnani) prices, or "structural abstractions" (Roberts).
Andrew Kliman