I think we are making some progress in understanding each other
despite our differences. Once again, let me respond to the
points you make.
Fred wrote:
John's main point in this post is that the determination prices of
production is "A PROCESS THAT TAKES PLACE IN TIME." John, would you
please explain further just exactly what your mean by this?
My comment: If the determination of prices of production is not
a process that takes place in time, what is it? Shall we say that
the determination is made at a point in time? If so, then how can
we speak of a change in productivity or a change in the wage?
That is, if we consider matters only at a point in time, how and
when do such changes occur?
Let's go right away to Chapter 11 of Vol. 3 where Marx considers
what would happen if wages were increased by 25% *everything*
else remaining the same. Now if we agree with Marx that the
prices of production would indeed change because of this wage
increase, we would obviously see a new set of prices of
production. Is it illegitimate to ask what might happen in the next
period of time? Would not the prices of the inputs of the next
period change or, as you might put it, the sums of money capital
advanced change due to these price changes? Would not the
changing input prices generally bring forth still another set
of prices of production?
You wrote:
1. Do you mean that Marx's determination of prices of production somehow
takes place OVER MULTIPLE PERIODS of production (as in Andrew and Ted's
interpretation)? John quotes two passages from Chapter 10 which seem to
suggest this interpretation. These passages are about the actual process
of equalization of profit rates, which takes place over successive
periods, as capital migrates from industries with lower than average rates
of profit to industries with higher than average rates of profit.
My comment: Clearly for Marx, the actual process takes time.
But as time passes are there no prices of production or do they
exist only for an instant? If so, the gap between the theoretical
determination of prices of production and the manner in which
theory takes into account the actual process is so large that
the theory is simply incapable of dealing with reality.
You wrote:
However, the actual equalization of profit rates (or the gravitation of
market prices to prices of production) is not the same thing as the
theoretical determination of prices of production. Prices of production
are theoretically determined independently of the actual process of
equalization of profit rates. Even though the actual process of
equalization takes place over multiple periods, this does not imply that
prices of production are determined in some way over multiples periods
(e.g. as in Andrew and Ted's interpretation in which new prices of
production are determined in every period for many successive periods).
My comment:
Here again, I think you make a crucial distinction -- actual determination
vs. theoretical determination.
I do not think Ted or Andrew claim that they are showing an actual
determination of prices of production. Rather they were forced into
using the reproduction schemes since historically that has been the
context in which Marx's "mistake" was noted. There are no reproduction
schemes in V3, Ch 9. That said, we can turn to the question -- do the
new prices of production brought forth in any one period of production
of any duration have an effect on the input prices of the next period?
You wrote:
2. Indeed, a review of Part 2 shows that Marx's explanation of the
determination of prices of production is presented entirely in terms of A
SINGLE PERIOD OF PRODUCTION. All of Marx's famous tables are for a single
period. All of Marx's other numerical examples in Part 2 are also for a
single period. Chapter 9 begins with the words: "AT ANY ONE GIVEN TIME
My comment: Let's finish that quote: "... the organic composition of
capital depends on two factors: firstly, on the technical proportion
between labour-power and the means of production applied, and secondly,
on the price of those means of production."
No one doubts that it is perfectly legitimate to look at the composition
of capital "at any one given time." But this by no means implies that
prices of production of the outputs are determined at that same point
in time.
Now you say that prices of production are determined "entirely within
a single period of production." This is quite different than at a
point in time. Within any period of production, input prices can and
generally do differ from output prices. Now if we are claiming that
that period of production is one of many, then there is no reason
to claim that any and all changes in prices of production due to,
say a wage increase, must be completed by the end of anyone period.
Indeed, if you were to say that a period is a year, I might well say
six months and at the end of the year have the same prices of
production that you compute. But I would be computing two sets of
prices of production and you one. Obviously, our results would be
the same given that you specify some length of time for the period.
In Chapter 11, Marx suggests that the period in question lasts for one
turnover. (CH 11, para 2) Again, he compares prices of production
with and without that 25% increase in the wage. What happens to
the prices of production in the next period is unexplored.
You wrote:
A few pages later (p. 258), a summary of Marx's theory of prices of
production ends with "DURING A GIVEN PERIOD OF TIME." The length of the
period of time is not always specified, but the most common period
mentioned is a year. Sometimes, Marx's examples are in terms of the
"working week" (pp. 245-46) or the "working day" (p. 248). But they are
always in terms of a single period, never for multiple, successive
periods. A single price of production or set of prices of production is
determined for this single period, not many sets of prices of production
for multiple successive periods.
My comment: Let's look again at p. 258 (Vintage)
"For each 100 units, every capital advanced, whatever may be its composition,
draws in each year, OR ANY OTHER PERIOD OF TIME, the profit that accrues to
100 units in this period of time as an nth part of the total capital."
So, again, we are talking about a period of time and not a point in time.
Further, the period itself can be as long or as short as we want. But
no matter how long or how short we could always subdivide it into several
more periods. In each and every period, prices of production would exist,
would they not?
You wrote:
According to Marx's theory, prices of production in any given period are
determined by the following equation (for each branch of production) (see
p. 265):
(1) ppd = c + v + r(c + v)
In this theory, the variables on the right-hand-side of this equation
(constant capital, variable capital, and the rate of profit) - the inputs
for this period of production - are assumed to be known. Prices of
production for this period are then determined by this equation. There
are no time subscripts in this equation because none are needed for an
analysis of a single time period.
My comment: No one doubts that (1) is true for a given period. Indeed,
the equation holds for each and every period that Andrew and Ted consider.
Here I think we can explore the matter a bit. Above we noted that Ch 9
starts with a given organic composition of capital. Indeed, given at
a point in time. We know not whether the prices are market prices,
prices of production or values. Rather they are simply prices. On
the basis of those input prices, the prices of production of the
outputs are to be determined. The difficulty is that the input
prices were somehow themselves determined.
Given that all Marx has considered prior to CH 9 is the aggregate,
the only input prices that could exist would be prices or values
of commodities that were produced as part of that aggregate. In that
aggregate there can be no difference between value and price of
production as the latter category does not exist at the beginning
of Chapter 9. Hence, the inputs must logically be priced according
to their values as we consider the composition of capital at any one
given time. We could, I suppose, go further and say that the inputs
were produced by capitals that are simply aliquot portions of the
aggregate. In this way they become prices of production as well as
values. But, more important, no matter how we see the input prices,
it is on the basis of those prices that prices of production of the
outputs are determined, using your (1) above. Once we carry out
the transformation procedure, we see that there is no requirement
that input prices must equal output prices for prices of production
to exist.
You start your description of the transformation procedure by noting
that Marx begins with given sums of money capital. Surplus value
is allocated on the basis of those given sums using the overall
rate of profit. Now if with part of those given sums of money
inputs are purchased and we assume a uniform positive rate of
surplus value, we have enough information to determine a set of
prices of production. So far there is nothing that tells us
we must go back and make sure that input prices are equal to
output prices
Again, I fail to see how and why we have to set unit input
prices equal to unit output prices as we determine the prices of
production. That is, in any given period why couldn't the unit prices
of inputs and outputs differ? We know that as technical change takes
place they will indeed differ. Should we develop a notion of prices
of production that only holds for the case in which there is no
technical change?
You wrote:
2. Marx's discussions of causes of CHANGES of prices of production in
Part 2 are usually not put in any specific time frame. Instead, the
discussion is about the general nature of the relation between the
independent and the dependent variables in the above equation for prices
of production. The question is: what causes prices of production, the
dependent variable in equation (1), to change? It follows from the
equation (1) that, if one or more of the independent variables on the RHS
of this equation changes, then the prices of production will also change.
Hence there are three possible proximate causes of changes in prices of
production: a change of constant capital, of variable capital, or the rate
of profit. Marx argued further that all three of these possible proximate
causes are one way or another the result of changes in the productivity of
labor or the real wage. Therefore, changes in productivity or the real
wage are the ultimate causes of changes of changes of prices of
production. As we have seen, prices of production change if and only if
productivity or the real wage changes.
My comment: Wait a sec. Something very subtle happened here. You
refer to changes in the real wage rather than changes in the wage.
I find it difficult to read Marx's discussion of the effect of
wage increases on prices of production (V3, Ch 11) if wages are
taken to mean real wages. Quite frankly, given the 25% increase
in the money wage, I have no idea how real wages are changing.
You wrote:
Marx did not say anything in these discussions of causes of changes of
prices of production about the LENGTH OF TIME required for prices of
production to change or for changes of productivity or the real wage to
bring about changes in the prices of production. The analysis is not
presented in terms of specific time periods. Rather, the discussion is
about the general nature of the relation between productivity and the real
wage, on the one hand, and prices of production, on the other hand. The
main point, again, is that prices of production change if and only if
productivity or the real wage changes.
My comment: Here again, we encounter your introduction of the real
wage. Why not go with Marx and simply speak of a 25% wage increase?
Let me suggest that the 25% increase in the money wage creates
problems for your reading of Marx. If the only input price we
vary is the money wage, then the prices of production change assuming
that workers add the same value to the means of production before
and after the wage increase. Marx demonstrates this in Ch 11.
He does not address the question of changes in the real wage. To
do so, he would have to go further and place the analysis within
the context of reproduction. He does not do this. If we do so,
we would have to consider several periods of production and reproduction
in which no additional changes are introduced. Prices of production
would change after each and every period and eventually we could
see the change in the real wage. Given the possibility of aggregation
problems, we may or may not be able to say by how much it changed.
You avoid all this by reading "wage increases" as "increases in
real wages." Hence, you are able to collapse everything into that
single period of production. The process that starts with an
increase in the money wage and its effect on prices of production
is now lost and seemingly pointless. Hence, your critique of
Andrew's and Ted's efforts.
You wrote:
In Chapter 11, Marx analyzed the effect of a change of wages on prices of
production (Ricardo's question). This analysis is conducted entirely in
terms of a SINGLE PERIOD of production, as in Marx's general theory of
prices of production discussed above. Marx assumed three capitals
(average composition of capital, higher than average, and lower than
average) and one period of production ("annual" is mentioned several
times). Wages are assumed to increase by 25%. This increase of wages
changes all the variables on the RHS of equation (1). As a result, new
prices of production for this period are determined for the two
non-average capitals (the price of production for the average capital
remains the same). Marx compared the new set of prices of production with
the old set of prices of production. The most interesting conclusion is
that the price of production for capitals of higher than average
composition falls, rather than rises as in Smith's cost-of-production
theory of value.
My comment: Let's look at that 25% wage increase a bit more. What is
significant to me is that while Marx notes that this wage increase
changes the prices of production, he does not proceed to change the
prices of production of the other inputs. His "c" is indeed independent
of changes in "v". For you, the change in "v" means that "c" must
change as well in the period of production we are considering.
Put another way, I am unclear about how you would deal with a 25%
increase in the money wage. That is after all what Marx is talking
about in Chapter 11.
You wrote:
Nothing is said here about HOW LONG it takes for prices of production to
change or possible problems or obstacles in this process of change. The
method of analysis is similar to what today we would call "comparative
statics." One of the independent variables is changed and the effects on
the dependent variable are analyzed. I know comparative statics is a
dirty word among TSSers, but it seems to me that this is essentially what
Marx did Chapter 11. Are there any other interpretation of Chapter 11?
My comment: Given that the independent variables include "c" and "v",
it's hard to imagine cases in which "c" does not change as well in
your manner of proceeding. Marx's discussion in Chapter 11 can
be easily placed within the framework of Ted's and Andrew's efforts.
It is not a problem nor do they have to say that the increase is
in real terms and not in terms of money. Given no other changes
you accomplish in an instant what may take them several periods.
To be sure that 25% increase in the money wages probably will not mean
a 25 % in the real wage as the prices of wage goods will change with
the changes in prices of production. Again, I am unclear about you
could start with an increase in the money wage and claim that
any and all changes in the prices of production due to that change
must be accomplished in one period or, more accurately, at a point
in time.
You wrote:
I should note that Marx's overall logical method is much richer than
comparative statics. But when he discussed the possible causes of changes
of prices of production, he conducted essentially a simple type of
comparative static analysis.
My comment:
Agreed. His comparative static analysis is, as you put it, richer.
We could assume that prior to the 25% wage increase the prices
of production are equilibrium prices. This gives us one static
state to compare. If in that static state we increase the money
wage by 25%, we arrive at new prices of production. The wage
increase did indeed change the prices of production. That's it.
That's Marx of Chapter 11. There's no discussion at all of any
change in the real wage with a new set of equilibrium prices.
Indeed, after the 25% increase in wages, we would generally no longer
have equilibrium prices equal to the prices of production. Should
we seek a set of equilibrium prices given this change, we would
have to track matters over several periods.
You wrote:
It should also be noted that Marx's main objective in Part 2 was NOT to
derive a time series for prices of production over multiple successive
periods. Instead, Marx's objective was to demonstrate that all possible
causes of changes of productivity ultimately boil down to a change in
productivity, i.e. a change in the value of commodities. In other words,
prices of production do not contradict the labor theory of value, but
rather can be explained on the basis of the labor theory of value
(something that Smith and Ricardo and all the other economic writers
before Marx were unable to do).
My comment: No one here is into deriving a time series for prices of
production. Nor are Ted and Andrew in their efforts. But if the
concept of prices of production is to have meaning in real time, must
we not say that they exist continuously. If they do not, then
the idea that prices of production are part of the labor theory of value
makes that theory valid if and only if we limit it to our analysis to
distinct sets of points in time.
You wrote:
For this objective, a simple kind of comparative static analysis will
suffice. One does not need to derive a time series of prices of
production in order to explain that all possible causes of changes of
prices of production boil down to a change in the values of commodities.
I would think that the derivation of a time series of prices of production
was the farthest thing from Marx's mind when he was writing Part 2.
Especially since equal rates of profit is only one aspect of the
distribution of surplus-value and there was much more for Marx to write
(indeed the rest of Volume 3) about the other forms of the distribution of
surplus-value (merchant profit, interest, and rent).
My comment:
That simple comparative static analysis you mention is, let's face it,
not so simple. Again if the only initial difference between the two
states you are comparing is a 25% increase in the money wage, matters
become quite complex since you insist on making prices of production
equilibrium prices.
You wrote:
3. In any case, there is nothing in all of Marx's discussions of changes
of prices of production like John's interpretation, according to which a
one-time change of productivity results in a SEQUENCE of many changes of
prices of production in many later periods, even though productivity and
the real wage remain constant in these later periods. In Marx's
discussions, a one-time change of productivity results in a one-time
change of price(s) of production. There is a one-to-one relationship
between productivity and prices of production. If time is specified at
all, it is a single time period.
My comment: Marx's discussion is quite simple. He shows the effect
that the 25% money wage increase has upon the prices of production for a
given period. Again, that's it. He does not bother with much else
in Chapter 11. But given only this one-time change it's hard to
see how the prices at the end of the period can be viewed as equilibrium
prices. To get to those equilibrium prices and take into account the
time it takes to get there, you'd have to do something like Ted and
Andrew do. I think they go there as a way of answering Bortkiwewicz.
Why you insist that equilibrium prices must prevail at the end of each and
every period is beyond me. Clearly, in Marx's Ch 11 the set of prices
of production brought about that increase in money wages is not one in
which we see equilibrium prices.
You wrote:
Therefore, John's attempted rescue of Andrew and Ted's interpretation must
be judged a failure. According to Andrew and Ted's interpretation, prices
of production CHANGE EVERY PERIOD, even though productivity and the real
wage remain constant, which seems clearly and explicitly to be
contradicted by Marx's texts. John's rescue attempt is to assume that a
one-time change of productivity results, not in a one-time change of
prices of production, but rather in a SEQUENCE of many different prices of
production in many successive periods into the future. With this
assumption, it could be argued that productivity changed BEFORE PERIOD 1
in Andrew and Ted's numerical examples, and that this prior change of
productivity sets in motion the continuing every-period changes of prices
of production that we observe in Andrew and Ted's examples, even though
productivity and the real wage remain constant in these later periods.
However, we have seen above that there is NOTHING in Marx's texts to
support John's interpretation of a one-time productivity change resulting
in many successive changes in prices of production.. If I have missed
something in Marx's texts, I am sure John will let me know.
My comment: My challenge to you is to start with two identical, distinct
and entirely theoretical periods of production. Do the transformation bit.
Now increase only the money wage in one. Do the transformation bit for
that one you changed. Note the difference in the prices of production for
the period in which wages were increased. That's Marx of Chapter 11.
To be sure, the new prices of production are probably not equilibrium
prices. Indeed, there's no need to continue unless one insists that we
must seek out a new equilibrium point. But why would we do so?
John