On Mon, 11 Dec 2000, you wrote:
> "Duncan K. Foley" wrote:
>
>
> Duncan, By money supply i meant MV of the QTM, but in any case the problem i was
> referring to is not the same as the NI concept of the 'value of money'. Paul C. and
> Allin C. seem to suggest that you can take value of fiat money at the beginning of
> the transformation and hold that constant during and after transformation since it
> is not a commodity and so it should not be affected by the transformation. I think
> this argument will simply not work. Because, first of all, how are you going to
> determine the 'value' of the fiat money?
This is at the one time both a completely abstract issue in dealing with
hypothetical economies, and at the same time a practical methodological
issue in Marxist statistical work.
Suppose that starting from real I/O tables for a a given country I want
to work out the value of the output of each industrial sector, and also to
work out the price of production for each industrial sector.
Note that the actually recorded figures for the gross output of
each sector will be
1. Measured in the national unit of account, pounds, dollars, francs etc.
2. Will probably coincide neither with prices of production nor with
values.
We have to work back from the observed monetary figures to the
value and price of production figures. We are trying to construct
counter-factual examples of what the price structure would be
under assumptions of goods exchangeing at values or at price
ratios taking into account an equal rate of profit on capital.
As a first step one can calculate the newly added value
per dollar of output for each industry, assuming that one
has figures for hours worked in each industry.
This figure will underestimate the value of each $ of
output because it neglects consumed means of production.
Call this vector of labour contents V1.
As a second step one can impute to the consumed means
of production in each industry their previously calculated
labour contents in terms of hours per $ given in V1 If one then
uses these you can calculate a second estimate of the
hours per $ output for each industry, call these estimates
V2.
This is clearly an iterative and convergent procedure that
affter sufficient iterations will get aribrarily close to the correct
figures for $ per hour for each industry.
We now have n distinct figures for the value of money dependent
upon the industry on which the money is spent.
In conjuction with the I/O table we can derive distinct
figures for the value flow of each industry in terms of
hours per year, and adding up, a single figure for the
GDP in hours per year. If one divides this by the monetary figure
for GDP one gets a mean value per $ figure.
Thus for the value per dollar figure, there is a determinate
process to compute it.
There exists a similar iterative process by which one can
go from the I/O table to figures for
$ price of production / $ market price
for each industry. Call these successive estimates
P1,P2,,,,,etc
Again one can compute the average $ price of production
per $ market price for the whole economy.
Thus there does exist a determinate procedure to arrive
at the prices of production and the values starting out from
observed data in terms of market prices.
Of course Ajit may object that this does not explain why the
price level is what it is. I would reply that this can not be
determined by a theory of commodity production. It needs
a theory of state finance to uncover its determinants. A commodity
production theory can only take the price level in
state money as a given.
--
Paul Cockshott, University of Glasgow, Glasgow, Scotland
0141 330 3125 mobile:07946 476966
paul@cockshott.com
http://www.dcs.gla.ac.uk/people/personal/wpc/
http://www.dcs.gla.ac.uk/~wpc/reports/index.html
This archive was generated by hypermail 2b29 : Sun Dec 31 2000 - 00:00:04 EST