Manish brought up a question that I had as well and I wanted to follow up
with a more general observation.
I'd need to check to be sure but I think that Ed opened the discussion with
about measuring manufacturing intensity but never clearly defined exactly
what he meant by manufacturing intensity. (He did give an example but not a
definition.) So people had a fair amount of latitude to interpret what the
concept was measuring and operationalize it as they thought appropriate
(without necessarily indicating how their operationalization actually
connected to the concept/construct.) As I am sure that everyone on the list
knows, operationalization requires a fair amount of (hopefully) educated
judgement about what is reasonable and what kind of data you can get given
the $$$ and time that you have.
As a (primarily) qualitative researcher, I think that it is worth suggesting
that (time permitting) it might be worthwhile conducting an "exploratory"
qualitative study that would help provide a better definition of exactly
what was meant by manufacturing intensity. Manish speaks to this point
(and, if I remember correctly, Ed Levitas' example made the same basic point
that Manish did.) However, based on some field work that I have done, I'd
posit that even "fabless" firms that outsource their manufacturing display
some level of "manufacturing intensity" in the sense that "designs" need to
be made "manufacturable." (There is a considerable amount written about
this topic in disciplines that look more narrowly at manufacturing
problems.) I would guess that at least some of the fabless firms do this
in-house (before they send their VLSI masks to the manufacturers). They are
also likely to have employees who stay in contact with the fabrication firm.
Similarly, on the output end, they are likely to have people involved with
quality control, inspection and the like. In this respect, even "fabless"
firms expend resources to deal with/think about manufacturing related
issues. To me, this should be captured in a measure of manufacturing
intensity. (However, I have provided my own interpretation of the construct
and it may turn out to be quite different than what Ed Levitas had in mind
with his initial question.)
Best, Bill
>From: "Manish K. Srivastava" <
msrivast@VT.EDU>
>Reply-To: Business Policy and Strategy List <
BPS-NET@AOMLISTS.pace.edu>
>To:
BPS-NET@AOMLISTS.pace.edu
>Subject: Re: question about measuring manufacturing
>Date: Sat, 23 Jun 2007 11:48:35 -0400
>
>
>
>"Any SIC code between 2000 and 4000 is manufacturing. Manufacturing
>intensity would then be sum (sales in sic codes 2000 to 4000)/all sales. "
>
>
>
>
>
>I don't think this assumption is correct. For instance, in the
>semiconductor
>industry for which SIC code is 3674 (between 2000 and 4000), there are at
>least two groups of firms: integrated device manufacturers and fabless
>firms. The fabless firms don't have their own manufacturing facilities,
>they
>are primarily semiconductor design firms and they outsource their
>manufacturing to foundries. All of their sales is in 3674 but none of that
>is from manufacturing. I am sorry, I did not offer any solution but I
>surely know this one is not correct.
>
>
>
>Manish Srivastava
>
>Doctoral Candidate
>
>2109 Pamplin Hall (0233)
>
>Pamplin College of Business
>
>Virginia Tech
>
>Blacksburg, 24061
>
>Phone: 540-231-7363
>
>Fax: 540-231-3076
>
> _____
>
>From: Business Policy and Strategy List [mailto:
BPS-NET@AOMLISTS.pace.edu]
>On Behalf Of King, Andrew A.
>Sent: Friday, June 22, 2007 10:24 AM
>To:
BPS-NET@AOMLISTS.pace.edu
>Subject: Re: question about measuring manufacturing
>
>
>
>Ed,
>
>
>
>The compustat business sector data should give you sales in up to 10
>business sectors. These sectors are given a main SIC code. Any SIC code
>between 2000 and 4000 is manufacturing. Manufacturing intensity would then
>be sum(sales in sic codes 2000 to 4000)/all sales.
>
>
>
>This will give you an approximate measure. Much better information could
>be
>gained from linking to the D&B data, but this will be a lot of wok and very
>$$.
>
>
>
>AK
>
>
>
> _____
>
>From: Business Policy and Strategy List [mailto:
BPS-NET@AOMLISTS.pace.edu]
>On Behalf Of Ed Levitas
>Sent: Thursday, June 21, 2007 3:52 PM
>To:
BPS-NET@AOMLISTS.pace.edu
>Subject: question about measuring manufacturing
>
>
>
>BPS-Netters,
>
>
>
>We are attempting to measure "manufacturing intensity" or the degree or
>amount of effort a firm directs toward manufacturing. For example, a
>company that only develops ideas for products but does not manufacture that
>product would receive a "manufacturing intensity" score of zero.
>Unfortunately, we only have access to financial data (compustat), alliance
>data, and number of products introduced. Does anyone know have suggestions
>about constructing a "manufacturing intensity" variable from these data ?
>
>
>
>Thanks in advance
>
>Ed
>
>
>
>****************************************
>
>Edward Levitas, PhD
>
>Associate Professor
>
>Sheldon B. Lubar School of Business
>
>University of Wisconsin-Milwaukee
>
>3202 N. Maryland Ave.
>
>Milwaukee, WI 53211
>
>ph: (414) 229-6825
>
>fx: (414) 229-6957
>
>
>