Re: More money or better distribution?

["John B. O'Donnell" <jackodonnell@xxxxxxxx>]

William B. Ryan wrote:
>
> John O'Donnell:
>
> 1.  You misuse mathematics by writing dQ/dM = 0, where Q is the level of
> economic activity, and M is the quantity of money.  Such an equation is
> meaningful only if either Q or M is the actual dependent variable of the
> other, which you specifically deny.

I do hope that you are not confused by the use of "d" in place of the
conventional lower case greek delta since the ratio is to "dM" and is
obviously only a partial.

That is, dQ/dt = SUM[{dQ/dX-sub-n)(dX-sub-n/dt)] for all X-sub-n's
including the X-sub-n that is "M" which while dM/dt <> 0 and dQ/dt <> 0,
dQ/dM does equal zero. [dM/dt can be likewise presented.]

> If the relation between two variables is truly functional, even if one
> is constant, changing the value of that constant may profoundly effect
> the result.

The meaning of dQ/dM = 0 is that they are _NOT_ functionally related.

> The most dramatic example of this is the value assigned to
> the speed of light.  Newton and his contemporaries had assumed that the
> speed of light is infinite.  If infinity is substituted for c in
> Lorentz's transformations, they become identical to those of Galileo.
>
> But Q and M are neither constant nor static.

True but irrelevant. [At least I assume this is true of Lorentz
transformations as I have no idea nor do I have reason to doubt it is
so.]

> Let me suggest a line of argument that Paul Davidson avoids, probably
> because of his fixation with "non-ergodicity."
>
> The only thing we can be certain of is that economic processes are
> individually dependent variables of time.  Taking time out of the
> equation can lead to much error.

And assuming that time is the only, or even the most relevant, variable
to cause change in the variable being examined can also lead to much
error. That is, leaving out of any search for possible causes any of the
many possible X-sub-n's described above may lead one to believe that
time is the only, or is the most significant, relevant variable. This
does not make it so. So?

> If Q and M are plotted against time from historical data and
> superimposed on the same graph, you will find that they are indeed
> closely correlated.  Mere correlation does not infer the chain of
> causation, which must be determined by other means.

Also true. Again -- so?

> 2.  Keynesians and Monetarists concur that money is non-neutral, that
> money matters.  They differ in their understanding of the nature of the
> existing system, and the utility of various reforms.

This is one of the assertions that keeps me using the dQ/dM=0
description rather than saying "the quantity theory of money" or
"neutrality of money" to describe the issue. The implications of either
of these terminologies is too often used to drag along some belief I do
not hold and whenever I allow such terminology it is then used to assert
that the aforementioned belief I do not hold is part of my argument.

"Neutrality of money" may accurately describe my position [I can't say
without some _VERY_ specific definition of the meaning of the phrase.]
but I do not deny that "money matters." Money matters very much as it
creates a very significant improvement in the ability to conduct trade
and that is a _VERY_, _VERY_ significant characteristic of money.
Further, as I have said many times before on this list and elsewhere,
the most significant characteristic of money in making this contribution
to the efficiency of trade is its reliability in representing value.

> Your assertion that monetary growth may affect distribution but not Q
> more closely resembles certain premises of the trade-cycle theory of the
> "Austrian" economist Ludwig von Mises.

I keep meaning to read some of Mises. Oh well, I'm such a delinquent.

The assertion concerning the effect of money creation on the
distribution of wealth is nothing more than a statement of the obvious
consequence that for any money increase to be meaningful someone(s) must
receive that money. It is _VERY_ unlikely that even if intended the
distribution of newly created money can be uniformly distributed such
that no change in the distribution of wealth results.

> 3.  It is almost beyond dispute that the existing money supply is
> largely endogenous, and cannot be otherwise--in which case money must be
> a dependent variable of Q.

Only in the case of models that so oversimplify reality as to lose
contact with the very concepts they are trying to examine. Money is a
function of the existence of the stuff of money [if money means some
substance] or is a function of the choices of a central bank [or it's
equivalent] if money either is or includes credit issues that are
generally accepted in trade. Q has nothing to do with the quantity of
money even though, because the actions of banks [central or otherwise],
may create a correlation between the two _MOST_ of the time.

> Such money naturally arises from transactions in trade.

No, it is usually created by banks or alternatives are found to conduct
trade. That's a very big difference and is a major reason why the
quality of money is so important.

> Every
> institution that grants loans is operating on the basis of fractional
> reserves, even if its reserves are putatively one-hundred percent.
> Credit is being created whenever a business sells to its customers on
> open account, which is greatly facilitated if the business can factor
> its receivables to a third party.  It was the generalization of this
> concept that was the great innovation of banking.

I don't think it is necessary to go into the detail of the semantics of
"fractional reserves" here.

> Davidson said that the case is closed.  Very nearly, it is.

Nope!

> 4.  But money can also be the independent variable.  Keynes admitted as
> much in the General Theory, where he said that banknotes could be buried
> in mines and then dug up, thus stimulating the economy.

In most cases it _IS_ independently variable, the only exceptions being
a semantic one of defining "money" as a substance with a fixed quantity
in existence or capable of being made extant. Calling it otherwise is
just getting lost in the world of incomplete models. [i.e. -- Those that
leave out most of the X-sub-n's that are relevant to dM/dt or whatever
total differential is being examined.]

			-- jbod

		Tax Privilege, Not People
___________________________________________________
Come visit and see a new economic perspective --
       http://www.geocities.com/CapitolHill/1067
           Comments/arguments welcome.
..