[OPE-L:4360] Re: Spheres and industries

From: Patrick L. Mason (pmason@garnet.acns.fsu.edu)
Date: Sun Oct 29 2000 - 19:36:51 EST


This is an empirical issue for which there is no easy answer. Whether or 
not items belong to the same sphere/industry depends on whether the 
commodities are substitutes for each other and on the strength of the 
substitution.  Or, one might want to define spheres in terms of the ability 
to produce different items using the same plant and equipment. So, in turns 
of substituability in use automobiles and army tanks are not in the same 
industry. However, the same plant and equipment and labor force can be used 
to produce both items, so from the standpoint of the latter definition they 
are in the same industry. Of course, both of these definitions are 
consistent with a broad industry definition, for example, vehicular 

The empirical difficulties of precisely measuring an industry does not rule 
out the notion of an industry as a valid theoretical concept. Few 
theoretical concepts are ever precisely captured by empirical measurement.

peace, patrick l mason

At 06:22 PM 10/28/00 +0100, you wrote:
>I'm just in the process of reading Mark Glick's thesis, which is of course
>about the gravitation of industry profit rates.
>Discussing classical theories of competition he quotes Ricardo and Marx on
>the redistribution of capital, as opposed to the workings of market price,
>as the source of profit-rate equalisation.
>Glick, along with others working in the same area, believes that one ought
>to expect to find no particular evidence of profit rate equalisation
>*within* industries, but only between them.
>The sources of *intra*-industry non-equalisation are said to be the
>possession of different techniques of production, more or less efficient
>organisational systems, etc.
>An aspect of this literature which I have always found rather unsatisfactory
>is that the subsequent empirical tests seem invariably to be conducted at
>something approximating (e.g.) the one-digit level of the 1981 version of
>the UK SIC system.
>This is partly because the intra-industry "prediction" seems entirely
>spurious at this level: for instance, industry 3 (Metal goods, engineering
>and vehicles industries) contains the four-digit industries 3222: Engineers'
>small tools, immediately followed by 3230: Textile machinery (3223 to 3229
>are unallocated).
>I'm easily persuaded that the manufacture of spanners requires different
>technology and organisation to that needed for making powered weaving looms,
>but so what?
>Even within four-digit industries the notion looks a little thin: being
>equipped to forge spanners does not entirely help one to engrave the Vernier
>scales on micrometers, for example.
>Neither Ricardo nor Marx use the term industry in the passages cited by
>Glick: in his quote from Ricardo the latter talks about capital (*not*
>capitalists) moving between "particular employments", by which he clearly
>means *enterprises*, not industries; likewise, Glick's citation from Marx
>refers to similar movements between "spheres of production", a term which
>would seem able to bear a variety of interpretations.
>Later on Glick quotes Marx as follows: "What competition brings about, first
>of all in one sphere, is the establishment of a uniform market value and
>market price out of the various individual values of commodities. But it is
>only the competition of capitals in different spheres that brings forth the
>production price that equalizes the rates of profit between those spheres."
>So a sphere is something which contains commodities: but my spanner is one
>(individual) commodity and yours is another.
>What I'd like to know, therefore, is what constitutes a sphere? Is it a
>space enclosing all engineers' small tools, all spanners, or all
>bi-hexagonal 15mm ring spanners (for example)?
>If one believes in the stratification of profit rates in the latter space,
>should one expect equalisation or stratification within industry 3222?
>Within industry 3? Within the economy as a whole?

This archive was generated by hypermail 2b29 : Tue Oct 31 2000 - 00:00:12 EST