[OPE-L:5117] Re: Questions to Ajit

Ajit Sinha (ecas@cc.newcastle.edu.au)
Mon, 26 May 1997 05:10:04 -0700 (PDT)

[ show plain text ]


I'll try to answer all your questions to the best of my abilities. But
before I get to that, let's clear the deck a bit. You and Andrew and the
general TSS approach, as well as others such as Fred Mosely, Glick and
Ehrbar and many others, have been calculating prices over periods of time
when prices are changing due to technical change etc. in terms of fiat
money; and you hold the value of one unit of fiat money constant over the
period of time. This is completely an indiffencible assumption. What
argument one could provide in its defence?

Once it is accepted that this assumption cannot be maintained, then the
question arises: can one have a unit of measure that would remain invariant
in the face of changes in the prices of all commodities due to changes in
technology etc. Actually classical economists were quite interested in this
problem. Adam Smith's so-called LABOR COMMANDED theory was developed in this
context. That's why he came up with the idea that corn will be the best
measure of value in the long run, whereas gold or silver would be a better
money-commodity for the short run purposes. Of course, they are not
'scientifically' precise measures. Many Sraffians (eg. Kurz and Salvadori)
and non-Sraffians (eg. Ong) have argued that Ricardo's search for his
'invariable measure of value' had also this element attached to it--an
aspect which the standard commodity does not take care of, and Sraffa wisely
never broached this issue. Ong (1983, HOPE 15, 207-27) has quite
conviencingly argued that such a measure is an impossibility, because a
measure of a change in the condition of production itself is depent on the
prevailing rate of profit (my thanks to Riccardo for recommending me Ong's
paper to read). It appears to me that a 'dynamic theory of prices' is a
logical impossibility. What I mean is that you cannot concoct a commodity
that would remain invariant in the face of a technical change in any basic
good sector.

A theory of prices is necessarily a static problem. Prices only tell us how
the different sectors of the economy are interrelated with each other.
That's why it is a ratio--an exchange ratio. We need a theory of prices to
understand the structure of an economy at any given time. Without it we do
not have an understanding of the structural relations of the economy.

Your basic concern is how to measure reassesment of 'fixed' capital in the
face of technical change. As I have suggested earlier, a consistent way to
do this is to treat fixed capital as a joint product. In this case, let us
suppose there is one new machine that produces a commodity. After the
production period you have two outputs: one is the commodity and the other
is the one year old machine, and the prices of both the commodity and the
one year old machine would be determined by input-output method. In the
second time period the input is one year old machine and the outputs are the
commodity and a two year old machine. Now, in the meanwhile, if the
technology has changed, then the fall in the value of the old machine would
reflect in the price of the two year old machine, and would influence the
rate of profit directly. I think you should take a good look at chapter 10
of Sraffa's book. I think he has the best solution to the kind of problem
you have. Let me quote the last paragraph of Sraffa's book. You may find it
quite interesting, as Sraffa credits Marx for using this procedure:

"The plan of treating what is left of fixed capital at the end of the year
as a kind of joint-product may seem artificial if viewed against the
background of the continuous flow of industrial production, but it fits
easily into the classical picture of an agricultural system where the annual
product, in Adam Smith's words, naturally divides itself into two parts, one
destined for replacing a capital, the other for constituting a revenue. Adam
Smith, however, excludes fixed capital from the annual product. It was only
after Ricardo had brought to light the complications which the use of fixed
capital in various proportions brings to the determination of values that
the plan in question was restored to. It was first introduced by Torrens in
the course of a criticism of Ricardo's doctrine. In explaining his own
peculier theory according to which'the results obtained from the employment
of equal capitals are of equal value', Torrens shows by means of examples
that his theory is varified if only 'the results' are regarded as including,
besides the product in the ordinary sense of the word, e.g. 'the woollens',
also 'the residue of the fixed capital employed in their manufacture'.
Thereafter the method was generally adopted, even by the opponents of
Torrens's theory: first by Ricardo in the next edition of his PRINCIPLES,
then by Malthus in the MEASURE OF VALUE and later my Marx, but afterwards it
seems to have fallen into oblivion." (The citation from Marx is given as:
CAPITAL vol. 1, ch. 9, sec.1, Moore and Aveling transl. p. 195, quoting
Malthus; and cf. the quotation from Torrens in THEORIEN UBER DEN MEHRWERT,
III, 77.)

Now to John's post:

>1. The difficulty with some input-output theories is that valuation
>takes place at "a point in time." This generally avoids any attempt
>to even consider asset losses as productivity increases. Now if
>asset losses must be "abstracted from" in order to obtain an
>acceptable price theory, then I am willing to consider other ways
>to take into account these losses. This may eventually mean a
>reconstruction of price theory. Until I am clearer on the
>valuation of assets as technical change takes place, I am willing
>to suspend judgment on any price theory that does not deal with
>the issue.

I think what I wrote above ahould constitute an answer to this.

>2. Implicit in what you say in the above is that Marx was clear
>on the entire matter of asset revaluation. Indeed, it would
>seem that we are to regard him as an early input-output theorist.
>Yet, when Marx introduces the concept of depreciation in CAPITAL,
>we find something a bit less clear than you indicate. (Here, let
>me add that I would appreciate your comments on the following
>passage from Vol I, Ch 15, Sec 4 of CAPITAL.)
>Marx wrote:
>The material wear and tear of a machine is of two kinds. The one arises
>from use, as coins wear away by circulating, the other from non-use, as
>a sword rusts when left in its scabbard. The latter kind is due to the
>elements. The former is more or less directly proportional, the latter
>to a certain extent inversely proportional, to the use of the machine.

This is more of a polemic against the capitalists. He uses this polemic more
than once. To counter the capitalists argument that since they provide the
means of production they are intitled to a share of the output. Marx
suggests that they should be thankful to the workers for working on their
machines, which maintains its value, otherwise the machines would rust and
lose all value.
Marx again:
>But in addition to the material wear and tear, a machine also
>undergoes, what we may call a moral depreciation. It loses
>exchange-value, either by machines of the same sort being produced
>cheaper than it, or by better machines entering into competition with
>it. <Note[64]> In both cases, be the machine ever so young
>and full of life, its value is no longer determined by the labour
>actually materialized in it, but by the labour-time requisite to
>reproduce either it or the better machine.

Supports my point against Fred and Duncan. The quotation continues:

It has, therefore, lost
>value more or less. The shorter the period taken to reproduce its total
>value, the less is the danger of moral depreciation; and the longer the
>working-day, the shorter is that period. When machinery is first
>introduced into an industry, new methods of reproducing it more cheaply
>follow blow upon blow, and so do improvements,
>that not only affect individual parts and details of the
>machine, but its entire build. It is, therefore, in the early days of
>the life of machinery that this special incentive to the prolongation
>of the working-day makes itself felt most acutely.

This is a kind of quotation I cherish so much. Marx is arguing like a
structuralist. He is explaining a certain kind of behavior on the part of
individual capitalists, i.e. the prolongation of the working day in the
factories, as an outcome of the structuralist competitive relation of
capitalism. Given that technical changes, which is caused by the
structuralist relations of the system itself and is not in control of
individual capitalists themselves, amounts to loss of value on the old
machines, the risk of such losses can be minimized by working up those
machines as quickly as possible. The point to note here is that this is an
explanation of, as one of the reasons for, the prolongation of the working
day during the early period of factory production. Any connection with a
theory of value or prices has not been made here.
Marx again:
>Note 64
>The Manchester Spinner (Times, 26th Nov., 1862) before referred to says
>in relation to this subject: "It (namely, the "allowance for deterioration
>of machinery") is also intended to cover the loss which is constantly
>arising from the superseding of machines before they are worn out, by
>others of a new and better construction."

The joint-production method of depreciation, which Sraffa credited Marx for,
will be consistent with this sort of depreciation.
>John continues:
>Here we have capitalists setting aside funds as part of the depreciation
>charges because of what Marx calls "moral depreciation." Are we to
>take this seriously or should we ignore such losses? Can we construct
>structure of production models that capture Marx's idea that capitalists
>depreciate their assets in the manner he suggests? For me, these are
>real questions that cannot and should not be written out of any economic
>theory and, especially, any that purports to represent Marx's.

Sure! Take it seriously as Sraffa did. I personally don't deal with fixed
capital because it is an unnecessary complication for the questions I'm
interested in, which is: what is the significance of a theory of value and
prices in economic theory and the question of the choice of the unit of measure.
>Ajit wrote:
>The problem with Andrew is that he neither understands value, nor prices, nor
>money as a unit of account; and these are the three concepts he is dealing
>with. His basic problem with the input-output theory of prices is that
>inputs are bought before outputs are produced and so their prices should be
>different. But he should know that at any given time point there could be
>only one price for a commodity.
>John comments:
>Again, the problem here is that production takes place in time. Thus,
>as we consider inputs becoming outputs there are at least two points
>in time involved. All would agree that at any point in time a commodity
>can have only one price. But given that the prices at which a
>commodity used as a capital input is purchased can differ from that
>at which it is sold after production takes place due to decreases in
>its value as productivity increases, the question becomes how one
>takes into account those changes in asset valuation. Perhaps we
>need to include within our concept of money the idea that it is
>also a store of value. The preservation of that value becomes
>problematic within the capitalist production process itself. It
>is an issue that cannot be simply solved by appeals to authority or
>by definition.

I think, I have answered this point. On the question of money as store of
value, you should keep in mind that the value of money changes when prices
change. That's why during a period of high inflation, fiat money is a very
poor store of value.

Cheers, ajit sinha