[OPE-L:5311] Re: OPE-L:5136 - Duncan's reply

Duncan K. Foley (dkf2@columbia.edu)
Wed, 2 Jul 1997 06:35:18 -0700 (PDT)

[ show plain text ]

Some remarks on Claus' OPE-L:5181:

Claus says:

>1) It seems convenient to break the subject down into simpler elements. On
>the basis of Marx's theory of money, there are three relevant concepts
>concerning this subject: the general equivalent and the function of
>measure of value; the standard of prices; and credit money.
>Central bank notes, like the dollar, being a liability of the State, are a
>form of credit money. They cannot perform the function of measure of
>value, according to Marx's criteria. The measure of value has the function
>to measure the amount of social labour contained in a commodity, and the
>point here is that, according to a rigorous theoretical demonstration
>attempted by Marx in the Grundrisse (see my EEA-97 paper, section 2), this
>function can only be performed by another commodity taken as a standard
> (Marx: "But only in so far as it is itself a product of labour, and,
> potentially variable in value, can gold serve as a measure of value",
>C, I,
> Ch. 3,1).
>This measurement takes place in the market, as gold enters into
>circulation and exchanges against the ordinary commodities.
> Marx: "Money, like every other commodity, cannot express the magnitude
> of its value except relatively in other commodities. (...). Such
> determination of its relative value takes place at the source of its
> production by means of barter. When it steps into circulation as
>money, its
> value is already given" (Ib, ch. 2).

Duncan comments:

I have no problem with the consistency and plausibility of Marx's theory of
the money-commodity. I'm not sure, however, that these passages were really
addressing the type of situation we confront today. I think Marx,
correctly, was arguing against the idea that the value of money is
completely conventional (that is, that gold had value only because people
thought it had value, in a kind of bootstrapped circular argument). But the
debt of the nation state is not purely conventional, since the nation state
has assets to back its debts, including its power to tax. The conceptual
problem we face arises from the fact that once the link between national
credit and gold has been broken, both the asset and debt sides of the state
balance sheet appear in dollars which have no clear link to produced
commodities. I still am disposed to think that these systems are basically
sustained by speculation on the value of the dollar and other currencies.
If not, by what?

Claus continues:

>2) The modern standards of money, like the dollar, on the other hand, have
>a double character. Under one of them, the dollar is not a measure of
>value, but a 'measure of prices', in the following sense. The standard of
>money, or standard of prices, consists of a given quantity of the
>money-commodity fixed by law as the unit of prices, which is totally
>conventional and subject to a high degree of arbitrariness in its
>manipulation by the State. In this way, it serves as a unit to measure the
>amount of the money-commodity that expresses the value of a given
> Marx: "As the measure of value it [money] serves to convert the
>values of
> all the manifold commodities into prices, into imaginary quantities
>of gold;
> as the standard of price it measures those quantities of gold" (Ibidem).
>I would like to distinguish, for the sake of clarity, *price* form
>*standard of money-price*: *price* is more abstract, it is the money-form
>of value, i.e., the expression of value in the form of the
>money-commodity; *standard of money-price* is the price expressed in
>terms of the conventional standard of money. In this way, the 'standard of
>money-prices' of the commodities may change
>without any change in their values (or prices in the first sense),
>whenever the standard of money changes. For instance, if the State reduces
>the content of gold of the standard of prices, the prices of commodities
>rise as *standard of money-prices*, but not as *prices*. In the USA, for
>example, after the Civil War, the support to the greenbacks and to
>silver-money, and in the thirties to the rise of the price of gold, was
>meant (at least by some proponents) to rise the 'standard of money-prices'
>by way of the devaluation of the standard of prices (the dollar in terms
>of gold).

Duncan comments:

I agree with this analysis, but I'm not sure how far it gets us towards an
analysis of contemporary monetary institutions.

>In the second character, the modern standards of money are forms of credit
>money. In this sense, the dollar is a central bank note, being neither
>measure of value nor standard of prices. It is credit money, a title of
>credit, liability of the State. [It is not to forget that paper money can
>also represent a mere symbol of value, whose more developed form has been
>the "inconvertible paper money issued by the State and having compulsory
>circulation" (Ib, ch. 3,2.c)]. What is it that the State ows to the owner
>of the dollars? Originally it owed the amount of the money-commodity
>expressed in the conventional standard of prices. This amount is at
>present not declared officially and is not legally payable.

Duncan comments:

Here you put the dilemma very poignantly. As Marx's analysis of credit
makes clear, credit is a promise to pay something else in the future. I
agree that the liabilities of the central bank are credit, but the second
part of your paragraph calls into question the relevance of the category of
credit altogether. (I'm not saying I have a completely convincing answer to
these problems, but I think they're important to recognize.) I would be
cautious about this idea of "compulsory circulation". In fact inconvertible
paper issues of states always go to discount against other currencies and
gold on black markets or foreign markets, even if the government tries to
use its police power to prevent people from dealing in foreign currencies
and gold. These police powers aren't very effective, anyway, I think, for
those sectors of the population that are used to movements of capital.

Claus continues:

But, as Duncan reminds in his OPE-L:5163:
>>It's rather striking that since about 1986 the price of gold in
>>dollars has not fluctuated a whole lot. Might the Fed be managing a de
>>facto gold standard?
>This is a difficult subject in Marxist theory of money today, but in my
>view it deserves better analysis before a stronger judgement can be made.
>Theoretically, in Marx's system, the standard of prices must express a
>given amount of the money-commodity. However, in today's economy it
>apparently doesnt do it. There is a contradiction between the internal
>laws and the observable phenomena. Can they be reconciled in Marx's
>theory? The answer to this problem requires, as I see it, to review the
>real development of the monetary system, under the light of Marx's theory,
>in order to find out if Marx's categories still hold and, if so, in what
>form they manifest themselves as the circumstances changed.

Duncan comments:

I wholeheartedly agree with these points. What has always struck me is how
much of Marx's theory of money, especially the relation between labor time
and money, continues to make perfect sense even with the very dramatic
change in monetary institutions.

Claus continues:

>Anyway, in my judgement we can in no way attribute the function of measure
>of value to forms of credit money, like the dollar.

Well, then, shouldn't we be able to state just what the measure of value
is? According to your analysis above, it must be a produced commodity. It
would be strange if so central a feature of the capitalist economy were
somehow hidden from everyone's view. Is it oil? or gold? or what?

Claus again:

>3) A final comment about Duncan's observation numer 2:
>>2) Why can't a "fictitious capital" (the capitalized value of land rent or
>>tax revenues) function as the measure of value? This surely requires
>>capitalists to impute a value to the fictitious capital at the time they
>>price the commodities, but wouldn't they have to do the same thing with
>In the way I understand it, this is a different meaning of *measure of
>value* than the one referred to money's function of determining the value
>of a commodity, in the sense mentioned above. "Fictitious capital" is a
>very complex category that pressuposes both money and capital as such. It
>is an asset whose money-value is determined not by its content in labour,
>but by the interest it is able to bear. In this sense the existence os
>fictitious capital implies the previous existence of a system of prices,
>which means that the problem os measuring values has already been solved
>by the market.

Duncan comments:

I find this a serious objection, though I might try to find a way around
it. Doesn't your description of a fictitious capital exactly fit the
valuation of the liabilities of the state, insofar as they capitalize the
income flow of future taxation?

Claus concludes:

>Theoretically you cannot define the value of one commodity in terms of the
>value of another commodity. You have instead to define it in terms of some
>other element. One may make an analogy with Marx's critic to the
>definition of capital as the value of an investment: since the value of an
>investment is a sum of prices, this means that capital is defined as a sum
>of prices, which is again merely a price. Thus, this is not a definition
>of capital. If this is correct, the attempts to atribute to forms of
>credit money the function of measure of value, in Marx's sense, are
>inconsistent. This does not mean that Marx's concept is correct, but it
>cannot be denied that it is theoretically consistent. This means that you
>cannot reject his theory without proposing another consistent theoretical
>explanation of the way the labour contents of the commodities express
>themselves as values.

Duncan responds:

Again, I agree completely with the program you propose. I don't know any
way to move forward on this except to consider some concrete options. At
the moment these appear to me to be:

1) to stick with the money-commodity theory, and to propose that the world
capitalist system is really on a (perhaps highly mediated) gold standard
(or perhaps some other standard.)

2) to move to the Keynesian point of view that the real value of the state
debt is not determined by any other forces in the system, so that it can
drift up and down, and that the main forces making it drift up and down are
the movements of money wages over the business cycle (a kind of money wage
standard, as Keynes clearly has in mind in some passages in the General
Theory). This is the point of view I put forward in my 1983 paper on Marx's
theory of money, but I must confess I'm having considerable doubts about it
both on theoretical and empirical grounds.

3) to recognize that the valuation of gold in the gold-standard system has
a speculative element in it which can be transferred to the speculative
valuation of the government debt. This is the point of view some
adventurous mainstream economists like Michael Woodford are adopting, and
that I put forward in my paper for the Bergamo conference.

4) to find some kind of synthesis of points 1)- 3).

This is all particularly difficult because of the immense importance the
fear of inflation has had in the last couple of decades in shaping the
political economies of the advanced capitalist nations, and increasingly
the not-so-advanced ones, too.

Maybe we can gradually shed some light on these dark matters.


Duncan K. Foley
Department of Economics
Barnard College
New York, NY 10027
fax: (212)-854-8947
e-mail: dkf2@columbia.edu