[OPE-L:4226] Re: Re: Re: Re: Part Two of Volume III of Capital

From: Rakesh Narpat Bhandari (rakeshb@Stanford.EDU)
Date: Sun Oct 22 2000 - 16:40:57 EDT

Steve, Duncan and others,

Please turn to pg. 98-99 of Duncan's book.

  Duncan takes the unit prices of production for the outputs and 
transforms the inputs in terms of them. This is exactly what Allin 
did. It's what Marxists have been doing for 100 years. I want to 
challenge the logic of this by simply introducing the reality of 
productivity change..

Duncan's transformation of the steel inputs now has the sum of their 
prices exceed the sum of their values. 15,750>15,000

I want to step back and do something exceedingly simple, but it's a 
path that is blocked if one needs to amortize the investment made in 
learning all the properties of equilibrium prices.

   Let us assume that productivity is increasing 5% a period.

So since the value of a steel unit is monetarily represented by $2 at 
the end of this period, the value of a unit of steel input would be 

Now that means where Duncan had assumed that the steel inputs for 
both branches represented 7500 physical units of steel, there is 
really only 7109 units which had a value of $15000.

Similarly, Duncan assumes that the wheat input of $10,000 represents 
6,667 units of wheat. But again if we assume that productivity has 
increased 5%, then wheat inputs with a value of 10,000 only represent 
6,329 units of wheat. The unit value of wheat as input would $1.58, 
instead of $1.50 as for the output. Again these discrepancies would 
of course be much less of course if we assumed less productivity 

Now the question becomes what should the unit prices of production be 
for the inputs?

It should be obvious now that the inputs cannot be transformed into 
the same unit prices of production as the outputs. This is the main 
point. No lecture about my ignorance of identity matrices is a 
response to this.

That is, if we use Duncan's output p's on the inputs ($2.10 for 
steel, $1.40 for wheat), then both wheat and steel would have prices 
of production less than their values.

For ease of calculation, I asked Allin to do one simple thing. Let's 
transform the inputs such that sum of the cost prices don't change. 
In Duncan's example, it's 25,000.

What makes this quite easy is that if we use the same P/V ratios 
(2.1/2 and 1.5/1.4) for the outputs and apply them to the inputs, it 
turns out almost just right. (1.05x$2.11x7109 +.93x$1.58x6329=a 
little bit more than 25,000, +45).  There would have to be only the 
smallest change in the PV ratios over time if one wanted to keep the 
total cost prices the same after the transforming of the inputs. But 
even this is not necessary; it only preserves the original 
calculation of the r.

And it becomes unbelievable to me that one could dismiss as 
implausible that  after the transforming of the inputs, Marx's 
determination of the average rate of profit will no longer make 
sufficient sense to even allow the right to be tested as a hypothesis.

It may not hold in cases of simple reproduction or equilibrium 
prices, but who cares? Why won't it hold in cases where unit values 
and unit prices of production are changing, even gradually, over time?

Again, my point is that I don't know what unit prices of production 
in terms of which to transform the inputs. I just know that they 
couldn't have been the same as for the outputs. And all this talk 
about long term *unit* prices of production has not yet been backed 
up by a single quote in Marx.

  I am just saying that as long as we are keeping the value of money 
constant (to which Duncan agrees) and the least bit concerned with 
the utterly realistic condition of rising labor productivity, one 
cannot use the same unit prices of production on the inputs as 
derived for the outputs, thereby eliminating time and characteristic 
productivity growth from the system.

And one doesn't have to allow anything to go to see how such a system 
could easily work.

All the best, Rakesh

This archive was generated by hypermail 2b29 : Tue Oct 31 2000 - 00:00:11 EST