[OPE-L:907] Price-Value Equivalence

Alan Freeman (100042.617@compuserve.com)
Wed, 31 Jan 1996 18:39:12 -0800

[ show plain text ]

Thanks to Allin for his detailed response [OPE 890] to my post
[OPE 884] both of 31/1/96.

I wrote

"Marx does not take the case of price-value equivalence as an
assumption in Volume I"

I fully expected people to jump up and down at this proposition
and thought I got off lightly when I introduced the question
earlier this month.

However since Allin has come into the debate after some of the
exchanges are over (to which of course there's no objection) it
may avoid repetition to just note that Jerry has listed the
earlier contributions in his review of the month. As far as I
can I want to deal with the new points introduced by Allin.

First I cannot resist taking one point out of order and dealing
with his last citation from pp327-328 because this illustrates
very clearly how easy it is, if one walks into Marx and reads
only what one wants to see, to miss what he actually says.

The passage Allin refers to seems to me to establish exactly
the opposite of what he sustains. The 'constant capital' of the
spinner includes the following:

"The rent of the building we suppose to be L300 per year"

Now without wishing to stir up sleeping horses, I always understood
rent to be an irrational price, that is, a price independent of
the labour embodied in the product. Yet we find that Marx
calmly and without batting an eyelid includes it in the constant
capital of the spinner. Nor can we assume he simply means the
depreciation of the building, since in the immediately preceding
passage he refers to the depreciation on the machinery, making it
clear that the rent is a distinct category.

If Marx rules out, in Volume I, sale at prices other than
values, what is rent doing here, since rent is a *systematic*
deviation of price from value?

This passage on the contrary rather strongly supports the
points which John Ernst has been making, namely that the value
of constant capital is given for Marx as it is in real life,
namely the value of the money expended by the capitalist in
order to secure the inputs to production.

It is moreover further, and rather damning evidence, against
the traditional intepretation of Marx which is based on what I
termed Equation (1) in [OPE 884], namely

v = vA + L

The given citation simply does not square with this equation.
Why does everyone miss this or pass it over? Because they read
the text with spectacles bequeathed them by an interpretation
which is not Marx's. They see what they expect to see, not
what is actually there.

To turn to substantive points.

Allin writes:
I, for one, disagree. At the very abstract level of Chapter 1,
Marx posits such equivalence as a precondition for the
intelligibility of the exchange relation. For example (refs. to
the Penguin edition throughout):

"The equation 20 yards of linen = 1 coat, or 20 yards of linen
are worth 1 coat, presupposes the presence in 1 coat of exactly
as much of the substance of value as there is in 20 yards of
linen, implies therefore that that the quantities in which the
two commodities are present have cost the same amount of labour
or the same quantity of labour-time." (pp. 144-5, ch. 1)

The problem with these citations and this general argument,
over which Gil and I have been slugging it out - with some
results I think useful - is that the fact that labour is the
quantitative measure of the value contained in them does not
impose the restriction that they must exchange in the ratios
given by this value, and nor does Marx's argument depend upon

Nor will you find any general ruling (such as you *do* find,
very clearly, in Volume II) that in Volume I it is assumed that
goods exchange at values.

In the citations you give, Marx in my view uses the case of
exchange at value to illustrate his argument and to debate with
the economists of his day. The first passage does not provide a
definition of value but assesses the effect of changes in
value; and it is hence perfectly natural to illustrate this in
the 'pure' case where exchange ratios are undisturbed by price-
value deviations.

The sense in which this case is 'pure' - and this may partly
deals with a similar point of Gil's on the use of the word
'pure' - is that if we allow price-value deviations to enter
this discussion at this point, we cannot be certain whether the
conclusions we draw arise from the price-value deviations, or
the phenomenon under study, namely changes in value.

I see no difference between this method and, for example,
deseasonalisation where we extract a trend and a seasonal
variation. If, in order to study nonseasonal effects, we study
only the trend, this in no sense means that we consider
seasonal effects do not exist or ignore the actual observed
data points. It merely means we make a separation of the
effects into two parts, one due to the motion of the earth and
one due to longer-term factors.

Or, to use a physical analogy, it is the same as the calculation
in mechanics which breaks down the motion of a body into the
motion of its centre of gravity, on which is superimposed the
motion of its component parts. 'Pure' motions are discussed
in mechanics by treating the centre of gravity as if it were
a dimensionless point; this does not mean that any engineer
believes all rigid bodies really are dimensionless points.
And it is only from analysis which does *not* assume that a rigid
body is a dimensionless point, that one can deduce the equations
which prove it is legitimate for certain points, to treat it

If, epistemologically, we treat the centre of gravity as *prior*
to the rest of the body or as part of the *definition* of the body,
we end up with absurdities. It is perfectly possible to have bees
without a swarm, but no-one has ever seen a swarm without the bees.

If, in order to study price variations due to two causes,
namely changes in productivity (value changes) and fluctuations
in supply and demand (price value deviations) this equally does
not mean that we deny the existence of price-value deviations.
It merely means we separate the causes of price movements into
two parts, one due to changes in value and one due to price-
value deviations; one due to the movement of the centre of
gravity, and one due to the movements from which this centre
is constituted.

Similarly in Marx's derivation of surplus value (as I stated in
my original post on this question) Marx does assume exchange at
values explicitly. But he has a specific and limited aim: to
refute the argument of Proudhon, that surplus value arises from
price-value differences. The most effective (and, I think,
devastating) way to do this is to demonstrate that even if we
eliminate price-value differences altogether, exploitation
still arises.

This does not mean the derivation of surplus value depends on
price-value equivalence, and it is quite easy to deduce surplus
value from the earlier part of this chapter in the general case
where goods exchange at any arbitrary price, as I indicated in
the post Allin is referring to.

The critical point is that Marx does not at any point, as far
as I can see, impose price-value equivalence as the basis of
his definition of value, nor as a general ruling assumption in
Volume I.

In fact he goes out of his way to deny it. That is precisely
why the qualification you refer to is necessary:

"The possibility ... of a quantitative incongruity between
price and magnitude of value, i.e. the possibility that the
price may diverge from the magnitude of value, is inherent in
the price-form itself. This is not a defect, but, on the
contrary, it makes this form the adequate one for a mode of
production whose laws can only assert themselves as blindly
operating averages between constant irregularities." (p. 196,
ch. 3)

It is also why, as has emerged from the debate with Gil, it
becomes very important to attach a meaning to the types of
statement we find on p265 where price-value differences are
systematically introduced, which immediately precedes the
definition of surplus-value, and which are meaningless if one
assumes price-value equivalence.

I don't think the formation of an average removes the logical
problem at issue and I think the function of the p190 and p265
passages are deeper than you accept.

If Marx's only purpose was to raise the question of price-value
divergence in order to dismiss it by asserting that goods
exchange at their value on average, then we cannot understand
why, in Volume III, he categorically asserts that goods do
*not* exchange at their values, even on average. If this is our
interpretation, I cannot but see that we must accept
Samuelson's famous 'eraser' critique.

The problem with the idea that value is merely an average,
attractive and fruitful though it is, is that one is immediately
driven to ask: an average of what? Unless, Nils-Bohr-style, you
take the view that *only* the average has meaning, so that we
are not entitled to enquire into an individual exchange, I do
not see how you can avoid this question. Particularly since
Marx asks it.

The real logical problem is: since goods can in principle
exchange in any arbitrary ratio, and since the category of
value is deduced from exchange, if it is accepted that even
on only one occasion goods can exchange in an arbitrary ratio
- never mind systematically and regularly as in Volume III -
then any category deduced from the assumption that they exchange
at one *particular* ratio cannot tell us anything about what
happens when they do not exchange at that particular ratio.

This is a very basic epistemological issue and the root of
my objection, for example, to the reasoning provided by Duncan
in replying to Andrew in [OPE 887 31/1/96]. The whole of economics,
as far as I can see, operates with categories of value derived
from simplified models. However, not all simplified models are
valid simplified models.

If these simplified models impose restrictions which cannot be
removed, then what poses as a simplification is not actually
a simplification but a straightjacket; the category of value
thus derived cannot be used without the restrictions.

Thus, one simplified model of the solar system is that a
planets is like a conker on a string, moving in circles round
the sun. This special case of a circular orbit can be used to
*illustrate* many important questions and even deduce important
constants, such as the minimum orbital velocity. But it cannot
be used as the basis for the definition of gravity. Because if
we assume at the outset that the planets move in circles, the
magnitudes we deduce from this will no longer apply when the
planets move in ellipses or worse, as in general they do.

Alternatively, it is as if we defined the weight of a body to
be the extension of a spring balance which would result if
weighed on the earth. This tells us nothing about what will
happen on the moon. In order to understand what happens on the
earth and what happens on the moon in a single conceptual
framework, we need a more basic concept which connects up
the two different acts of weighing, namely the concept of *mass*,
which is independent of any particular act of weighing and
allows us to say what happens in any arbitrary act of weighing.

'Mass' is to 'weight' what 'Value' is to 'price'; it is the
quantity in which weight must be reckoned, if we wish to abstract
from weighing on any particular planet or comparing against any
particular standard.

Marx's derivation of value is the only derivation in economics
that I know of which does not depend on the assumption of
reproduction at prices equal to values. It is therefore, I
contend, the only derivation suitable for studying general,
dynamic phenomena.

Allin goes on to cite:

"The same value, i.e. the same quantity of objectified social
labour, remains throughout in the hands of the commodity-owner,
first in the shape of his own commodity, then in the shape of
the money into which the commodity has been transformed, and
finally in the shape of the commodity into which this money has
been re-converted. ... Insofar, therefore, as the circulation
of commodities involves only a change in the form of their
values, it necessarily involves the exchange of equivalents,
provided the phenomenon occurs in its purity." (p. 260, ch. 5)

But Marx then proceeds precisely to discuss what happens when the
phenomenon does *not* occur in its purity, with the following
example: suppose wine worth L40 exchanges for corn worth L50.
(which means, as I said in reply to Gil, that they both sell
for a price of L45).

We require a category of value in order to make the statement
that has just been made. But this category of value is *not*
derived from exchange. The fact that wine is worth L40 is not a
statement about exchange. It is a statement about production.
It says that the wine was produced using goods and living
labour which contributed a total of L40. If the value of money
is 1 hour per pound, it means that the wine incorporates 40
hours of dead and living labour. In and of itself, this says
nothing about what that wine exchanges for. It is simply a
measure of the quantity of value in the wine, just as the mass
of an apple is simply a measure of the quantity of matter in
the apple. It does not tell us whether this L40 will exchange
for L50, L60 or L10,000, just as the mass of the apple does not
tell us whether the apple will actually stretch a spring
balance 6 inches, 12 inches or 3 miles. The actual extent to
which the spring stretches depends on the other partner to the
weighing event, namely the planet on which the apples is
weighed. The actual amount of value against which our wine,
worth 40 hours, exchanges depends on the other partner to
exchange, namely the corn or, more strictly, the money against
which it exchanges.

So, I think the citations given by Allin allow us further to
illustrate the precise role played by the *temporary and
contingent* assumption of value-price equivalence in *certain
limited deductions* of Marx, a role which it could not play
unless it was a *special case* of exchange at any arbitrary
prices, for which purpose Marx uses a category of value
*derived from* exchange at any arbitrary prices.

Furthermore, the concept of 'pure' exchange is not to be
understood that 'impure' exchange does not take place, but that
the effects of arbitrary exchange can be separated into those
effects due to changes in underlying values, and those effects
due to the deviations of price from value, which is a perfectly
normal and standard analytical technique for analysing any
composite motion.