[ show plain text ]
Per my reply to Alfredo, I would say that what you've identified isn't
really a problem. That is, in constructing input-output systems how do we
know what activities to aggregate into differing industries. The answer is
that it depends on the issue that we are exploring. At one extreme (and I
know you are not saying this), every firm is a unique industry because
every firm produces a unique product. At the other extreme, every firm is
part of giant entity called "capital," which is Marx's assumption in volume
1. For empirical work, we may move from 1 digit SIC code industries (about
16, I think) to 4 digit SIC aggregations (about 500, if my memory is
correct). Conceivably, there is no reason that we cannot have even finer
aggregations. The only problem is cost.
Note however that this is not simply an empirical issue. As soon as we
allow that firms use heterogeneous technologies and managerial strategies,
employ heterogeneoous workforces, and produce a heterogenous products, we
have some difficulty in defining an "industry." Well, whatever procedure we
use to solve this theoretical problem is exactly the same procedure we can
use to solve the empirical aggregation problem in constructing an I-O matrix.
At 03:41 PM 5/24/00 -0400, you wrote:
>> >From: Paul Cockshott <firstname.lastname@example.org> >Subject: Re: [OPE-L:3300]
>> RE: Re: objectivity of value > >At 12:20 24/05/00 +0100, you wrote: >
>> > > >> 3299] Re: objectivity of value >> > >> >Paul C writes: >> > >>
>> > This is a point where we part company, I would go along with Ajit >>
>> > in saying >> > that the objective measure of value is labour time.
>> This exists prior to >> > sale as a >> > property of the input/output
>I don't understand how this is possible. The input output matrix already
>presupposes a value measurement: how else would you know which branches to
>include (cars instead of buggies) and how can you determine the average
>condition of production in each branch unless you know on the basis of
>demand levels the last marginal producer to include in the determination
>of that average?
>It does not seem to me that the input output sysem can preexist the value
>measurement. It only becomes real when commodity output is measured as
>values in terms of a system of prices.
This archive was generated by hypermail 2b29 : Wed May 31 2000 - 00:00:11 EDT