# [OPEL:6187] Re: Historical Costs (Fred)

John R. Ernst (ernst@PIPELINE.COM)
Fri, 13 Feb 1998 16:06:34 -0500 (EST)

Hi Fred,

I appreciate your efforts in trying to work this out with
others. Let me begin by setting up my framework for this
problem and then checking out your's.

My framework:

Let me construct my picture and then attempt to fit what you
are saying into it.

Beginning of Period End of Period
Period Capital Invested Used Capital
1 C(1B) C(1E)
2 C(2B) C(2E)

______________
C(1B) Value of initial investment in fixed capital of
Period 1 which is the same as the historical
cost of the investment.

C(1E) Value of the used fixed capital at the end of
period 1.

C(2B) Value of the fixed capital at the beginning of
period 2.

C(2E) Value of the used fixed capital at end of period
2.

________________

My take on this. At the beginning of Period 1, a certain
amount of capital is invested, C(1B). Here we have no
debate as it represents funds invested. At the end of
the production period, we have an output and the unused
capital. This unused capital, say a machine, has a value.
It depends on how much of the machine was used up in
period 1 by wear and tear as well as on moral depreciation.
Now, if in period 1, better and/or cheaper machines become
available, moral depreciation occurs. At any rate,
at the end of the period, we can write:

(1) C(1B)-x(1)=C(1E)

where x(1) represents the total depreciation that occurred
in period 1. The machine has been devalued by not only
wear and tear but moral depreciation as well. In my
view, it is the selling price of the used machine. At
the start of the next period, the value it has at the
beginning of the period is equal to its value at the
end of the last. Thus,

(2) C(1E) = C(2B).

Note that the difference between C(2B) and C(1B) is x(1),
the depreciation that occurred in period 1. Note as well
that capital value in this scenario is devalued.

Now let's focus directly on your post.

You wrote:

(snip)

1. It seems to me that, in the discussion about Marx's valuation
of constant capital, there are three possible interpretations, or
that there are three different points in time at which it is
assumed that the value of constant capital is determined in
Marx's theory:

T1: the time (in the past) of the actual investment of capital.

According to this interpretation, the value of constant capital is
equal to the actual money invested at this point of time in the
past. Even if there is technological change (or a change in the
value of money), the value of constant capital remains
unchanged at this original amount. This is what I understand
by "historical cost" valuation (and I think this is the usual
meaning of "historical costs").

The difficulty I can see here is that unclarity as to when the
technological change occurs. For me, C(1B) is the initial
investment and the value of that investment at the start of the
period.

T2: the beginning of the current period.

According to this interpretation, if there has been technological
change (or a change in the value of money), then the value of
constant capital changes (is "revalued") and is not equal to the
actual amount of capital invested at T1.

But when is T2? For me, at the end of period 1, 1E; the value
of the capital invested is C(1E). As you can see from the above,
I would agree that there is a devaluation of constant capital
during that 1st period due to moral and actual depreciation.

I do not see the difference between T1 and T2, at the start of
the first period.

T3: the end of the current period.

According to this interpretation, if there is further technological
change (or a change in the value of money) during the current
period, then the value of constant capital changes (is "revalued"
again) and is not equal to the value of the constant capital at
T2. This is what I understand by "current reproduction costs"
(which again I think is the usual meaning). However, the
differences between T2 and T3 will in general not be large,
because the technological change (or change in the value of
money) during a given period is generally small. On the other
hand, the further back in time T1 is, the greater will be the
difference between T2 and T3, on the one hand, and T1, on the
other hand.

For me, C(1E) is the same as C(2B). Given I do not know when
T2 is, it's hard to deal with some of this. Note that T1, the
original investment seems to represent my C(1B) and is unchanged.
However, in the second period, the capital advanced, C(2B),
represents the value of the initial invested after depreciation
and devaluation.

All this says is that in the first period, the capitalist computes
his rate of profit on the original investment. The original
investment is the historical cost. I assume that as the first
period starts, he bought the stuff. Thus, he bases his rate
of profit calculation on the capital advanced in that first
period. In the next period, it would be based on C(2B) the
value advanced at the beginning of the period.

You continued:

2. Now, I have been arguing that Marx assumed that constant
capital is determined at T3 (i.e. by current "reproduction"
costs).

It seems to me that Ale and John in their recent posts are
arguing that Marx assumed that constant capital is determined
at T2 (both the stock and the flow of constant capital, right?).
If that is indeed what you are arguing, then I would say that
the difference between us is not great, and I would be happy to
discuss further in hopes of a perhaps reaching a consensus.

We agree and disagree. Note that for me C(1E) is the
"worth" of the investment at the end of period 1. Its
worth at the beginning of the period was what he paid
for it. Given that this was an act that takes place in
real time, we can't change it. What we can change, however,
is its worth at the end of period 1 and, thus, at the start
of period 2. (I think you need a T4 to capture
the idea of my C(2B) -- the value of the investment after
it has been devalued.)

You continued:

3. However, is I understand it, Andrew's interpretation (and
Alan's too I think) is that, even in the case of changes in
technological change, the value of constant capital is
determined once and for all at T1, (it is not clear to me exactly
what Andrew is assuming about the case of a change in the
value of money). Or rather, more precisely, the STOCK of
constant capital - the denominator in the rate of profit - is
determined at T1. This is Andrew's assumption in his paper on
Okishio's Theorem in Freeman and Carchedi, as I understand it.
(The FLOW of constant capital - the value transferred to the
price of the product - is assumed by Andrew, as I understand
him, to be determined at T2).

that both the stock and the flow of constant capital are
determined at T2, then I would be very happy. As I just said,
then I think our differences would be minor, and maybe could
be resolved. I can accept T2 as a possible interpretation of
Marx (although I still think the textual evidence is in favor of
T3). But I cannot accept T1 as a possible interpretation of Marx
- at least of what Marx was doing in Capital. As I have said
before, every single passage in which Marx explicitly discussed
the effect of technological change (or a change in the value of
money) on the valuation of constant capital, he stated that the
value of constant capital is determined in the current period
(i.e. that the original capital is "revalued"). He never once said
that, in the case of technological change, constant capital
continues to be valued at the original historical costs. The
whole tenor and thrust of Marx's writing on this subject is
contrary to the valuation of constant capital at historical costs.

It's hard to get into this. But I will say that my discussions
with Andrew and Alan, esp. Andrew, moved me to developing my
ideas. Again, what makes the discussion difficult is that
with T1,T2, and T3, I am unsure of where these times are
located relative to a period of production. T1 is, for me,
at the start of the period and, assuming this is a new
investment, it is both the historical and initial capital
advanced. T2 is when? T3 seems to be the worth of the
new machine at the end of period or my C(1E).

Fred continued:

(determination of constant capital at T2)?

John responds:
Again, let me ask you to locate T2 within the period of
production.

(snip)
Fred continued:

5. Furthermore, I agree with Ale and John that, given that
constant capital is revalued in some way as a result of
technological change, then the next question is: how is the rate
of profit determined, and in particular how does the capital loss
that results from the devaluation affect surplus-value or profit,
the numerator in the rate of profit? I think Ale lays out the
two options clearly: (a) the capital loss is NOT subtracted from
the gross surplus-value or (b) the capital loss IS subtracted from
I can see the advantages of (b) and maybe I can accept it.
But I see no evidence of (b) in Marx. In particular, when Marx
discussed the "cheapening of constant capital" in Chapter 14 of
Vol. 3, he did not say that the capital loss that results from
this cheapening should then be subtracted from the gross surplus-value
to obtain a net surplus-value. Indeed, Marx discussed this "cheapening"
as a counter-tendency to the falling rate of profit. If the capital loss
is subtracted from the gross surplus-value, then it is not clear
to me that this cheapening would be a counter-tendency.
Maybe the difference between (a) and (b) is higher and lower
levels of abstraction. Also, to be consistent, if one subtracts
capital losses, then one should add capital gains that results
from a decline in the value of money. Ale and John, is that how
you would treat capital gains?

This is tough stuff. But again, it's why we are dealing with this.
If there is a great deal of moral depreciation in Period 1 (my
scenario), then in that 2nd period less capital value is advanced
and there is indeed be a tendency for the rate of profit of the
2nd period to increase. But Marx also points out that if
anticipated moral depreciation is so great the capitalist will
refrain from investing.

To get at the question of whether or not moral depreciation
is to be subtracted from surplus value, I think we need to
track a particular investment from start to finish. If
there were no "moral depreciation", then the rate of profit and
that of surplus value would be higher than they would be
with moral depreciation. I suppose you could say that this
means moral depreciation is a deduction from surplus value
or profit, but it would be comparing a world with technical
change and devaluation with one in which there is none.

Take care,

John