[OPE-L:5069] Re: RRI and The Rate of Profit

john ernst (ernst@pipeline.com)
Mon, 19 May 1997 01:07:17 -0700 (PDT)

[ show plain text ]

Hi Fred,

Below I respond to your OPE-L 5068.

Fred wrote:

1. John suggests a highly controversial interpretation of Marx's concept
of "moral depreciation" (I would say a dubious interpretation), and then
seems to argue in effect that this controversial definition of " moral
depreciation" implies that Marx must have valued constant capital at
historical costs. John's controversial definition is that "moral
depreciation" is a kind of accelerated depreciation: because capitalists
anticipate future technical change and the resulting devaluation of their
existing constant capital, they depreciate their constant capital at a
faster rate in order to avoid this devaluation. If capitalists'
anticipations are correct, then capitalists recover their full investment
of constant capital valued at historical costs.

I have argued in several posts over a year ago that John's definition of
"moral depreciation" is not Marx's definition, which is instead the loss of
capital that results from unanticipated technical change. As I remember
it, this is also the way that "moral depreciation" has usually been
interpreted in the literature (although I haven't had the time to check
this out). I know of no one else that defines "moral depreciation" like
John does. (John: if others have a similar definition, please give me the
references). I will not go back over the details of this debate now. But,
however this issue is decided, I think that it has to be admitted that this
controversial defintion of "moral depreciation" is very slim evidence to
justify the conclusion that Marx assumed that constant capital is valued at
historical costs. ESPECIALLY when there is so much direct, explicit
evidence to support the opposite conclusion - that Marx generally assumed
in Capital that constant capital is valued at current costs. Whatever the
merits of John's concept of "moral depreciation", it does not provide
convincing evidence that Marx himself assumed historical cost evaluation.

John responds:

Let's take the 2 issues, one at a time.

a. Moral Depreciation. Without going back over the old battle of
quotes, let me simply point out that when Marx first introduces
moral depreciation in Vol. 1, he states that it shortens the life
of fixed capital and that capitalists know it and set aside
additional depreciation funds to allow for it. In Vol. II, he
notes that there can be losses to surplus value due fires, natural
disasters, etc. but does not include moral depreciation as such
a loss. In Vol. III, he states that capitalists may, at times,
withhold investments because moral depreciation is too high.

It does not surprise me that the secondary literature fails to
recognize my reading of Marx on this. For years, simplistic
readings of Marx on a variety of issues passed for scholarship.
My challenge on this is simple. Look at Marx's introduction to
this concept and examine the way he uses it. I maintain that
there is "slim evidence" to interpret it any other way. Here,
I do note that at no point do you attempt to back up your
position with excerpts from CAPITAL which actually contain Marx's
definition of the concept as well as its use. (Let me point
out that there is also "slim evidence" that you, Duncan or
TSS have correctly interpreted the transformation procedure.)

Let me suggest that those who want to revalue the constant
capital after each and every period of production must misread
Marx on moral depreciation. It makes the immediate simultaneous
valuation of inputs and outputs impossible. Thus, the rule
that "if it doesn't fit, you must omit." is immediately applied
with no textual evidence.

I do recall having a discussion with you on moral depreciation.
I, honestly, cannot recall how you argued against Marx on this

b. Current Cost Valuation. Note that I am not arguing against
this type of valuation. I am arguing against the idea that it
occurs at the end of each and every period of production. To me,
this a bit like assigning individual values to commodities at
the end of each period of production without recognizing that
the task is to show how social values fall to the level of
individual values or even lower.

Fred wrote:

2. The second issue raised by John is the RRI. I have not followed all
the recent discussion about the RRI, so I may not understand all that is
involved here. But it seems to me that John is saying that his recent
discussions with Andrew made him realize that his definition of "moral
depreciation" is inconsistent with Marx's rate of profit as usually defined
- as the objective, ex-post ratio of the total surplus-value in the economy
as a whole to the total stock of capital invested. However, he argues that
his definition of "moral depreciation" is consistent with the concept of
the RRI - the subjective, ex-ante "rate of return on investment". John
says that, rather than give up his concept of "moral depreciation", he has
decided to give up instead Marx's objective ex-post rate of profit in favor
of the subjective ex-ante RRI (or is at least considering this). This
seems to me to be a very high costs to pay just to hang onto John's
controversial concept of "moral depreciation". But, in any case, this
would clearly be a departure from Marx (John does not seem to dispute
this). And, most important for the current discussion, however one decides
this issue, John's discussion of the RRI cannot possibly be an argument
that Marx assumed that constant capital is valued at historical costs.

John comments:

I think that both the rate of profit and the RRI can be determined ex ante
as well as ex post. I did indeed state that my discussions with
Andrew concerning moral depreciation and the rate of profit did force
me conclude that Marx's concept of moral depreciation requires the
use of the RRI. But I do think there are additional reasons for using
the RRI.

a. The concept was not available to Marx. Hence, its non-use by Marx
does not furnish an argument for excluding it from his effort.

b. Can anyone seriously maintain that capitalists' investments are
based on the rate of profit? To be sure, in the days of
classical political economy this may have been a factor as there
was little fixed capital within production processes. But given
the presence of fixed capital no capitalist will invest in a
process in which he anticipates a higher rate of profit but
a lower RRI. That same capitalist would, of course, invest
in process for which a lower rate of profit and a higher RRI
are expected.

c. It is no longer surprising to me that after finishing Vol. I,
Marx focused on Vol II and never really came back to his draft
of Vol. III. He never could finish Ch 4 of Vol. III, "The
Effect of Turnover on the Rate of Profit" since he had no
idea how the turnover of fixed capital influenced the rate
of profit.

Fred continued:

Therefore, I think one has to conclude, on the basis of the evidence and
the discussion thus far, that Marx himself assumed in Capital that constant
capital is valued in current costs. We can to on to discuss the other
issues rased by John (at a lower level of abstraction), but unless further
evidence and arguments are presented, this conclusion should be acknowledged.

John comments:

It was acknowledged. But as you point out below, the issue is what happens
as capital moves from one period to the next. You state that this
issue is at a lower level of abstraction. However, as we both know,
most models that purport to represent Marx revalue constant capital
after each and every period of production. Only of late, have I seen
more than TSS folk acknowledging that this is problematic. For
example, in his EEA paper, Laibman noted that the losses due to
asset revaluation are a problem that must be treated. In his more
recent posts, Duncan has mentioned the need to consider asset
revaluation as accumulation occurs.

Fred wrote:

3. To clarify a point raised by John, I am not saying that Marx assumed
that the devaluation of constant capital as a result of technical change is
necessarily immediate from one period to the next. Marx's theory in
Capital is at a high level of abstraction. He was not trying to explain
actual market prices in each and every period. Rather, he was trying to
explain the long-run tendency of these prices. It seems to me that Marx
argued something like the following:

Small devaluations of constant capital take place all the time as a result
of ongoing technological change. Competition forces prices to be based on
current costs, maybe not in the next period, but sooner rather than later.
These small ongoing devaluations do involve the loss of capital, but they
do not by themselves necessarily cause a crisis.

Furthermore, I think that Marx argued that, in spite of these small ongoing
devaluations (i.e. in spite of the ongoing "cheapening of the elements of
constant capital"), the rate of profit - evaluated in current costs - still
falls, and this fall in the current cost rate of profit evenually causes a
crisis, in which much bigger devaluations take place, as a result of
bankruptcies, which thereby restores the rate of profit and makes possible
a recovery from the crisis.

John comments:

I'm a bit unclear on your "level of abstraction." On the one hand,
"CAPITAL is at high level of abstraction." On the other hand, in
CAPITAL we find crises caused by a falling rate of profit which
leads to larger devaluations and the restoration of the rate of
profit. This doesn't seem so abstract to me.

I read Part III, Vol. III a bit differently with respect to crisis.
That is, I go with Grossmann's notion that, at times, in some period
there is not enough surplus value to meet the demands of accumulation
in the next. This does not require that the RRI or the rate of
profit visible to capitalists actually falls.

Here, both of us stand accused of orthodoxy. I would maintain with
either interpretation the notion of how valuation takes place from
the period immediately prior to crisis and that of crisis can and
should be considered at this level of abstraction.

Fred wrote:

John commented in an earlier post that he thinks that a decline in the
"current cost" rate of profit in contemporary capitalism is unlikely.
Adding to Duncan's response to this comment in (4968), according to the
available empirical evidence, mine and other's, the CURRENT COST rate of
profit DID DECLINE significantly in the post World War II period, at least
in the US and Mexico, and I think also in the UK and other countries.
(Alan, are your estimates for the UK, which are published in the Dunne
volume, in terms of current costs or historical costs? I can't remember
and I don't have it with me). In other words, Marx's theory of the decline
of the current cost rate of profit is generally supported by the available
empirical evidence for the postwar period (i.e. for "late capitalism").

John comments:

a. Given that much of the recent theoretical controversy stems from
attempts to deal with the Okishio theorem, I think we need some-
thing more than macro studies of the rate of profit. That is,
insofar as discussions around Okishio focus on choices of technique
made by individual capitalists, it seems to reasonable to ask
how individual capitalist invest. That is, it would seem that
a study or two of an individual firm would be in order. Again,
let me point out that for sometime I've been asking for
individual examples of the introduction of machines that
represent "Marx biased" technical change on the level of the
firm. I have seen none.

b. I would now admit that rate of profit may fall since that
is completely compatible with a rising RRI, the primary variable
for rational capitalist decisions. As far as I know, only
Dumenil and Levy have examined the RRI over time. They found it
falling although they observe a lag in its decline relative to
their falling rate of profit. I wish they had spent a bit
more time on RRI and less on the rate of profit. Given the
short life spans governments assign to fixed capital and their
use of government stats, I'd like to know more of the way
they arrive at the falling RRI.