Re: Why isn't semiconductor memory made in USA anymore?

From: John Conover <>
Subject: Re: Why isn't semiconductor memory made in USA anymore?
Date: 2 Sep 1999 18:24:13 -0000

Don Dale writes:
> I don't follow you here.  Economists are the very folks who do the analysis.
> Economists have analyzed the costs and benefits of virtually every trade
> barrier that ever existed.  We promote the analysis.  We also are the ones
> who say things like, "You know, if you really wanted to help America's
> textile workers, you could do x, instead of that trade barrier."  We're
> looking for efficient ways to meet equity goals all the time.  But nobody
> listens, so what can you do?

Hi Don. I've been in the semiconductor industry for a long time. It is
a different kind of industry. Its a very fickle industrial market and
goes up, and down-sometimes 20% in just a few days. The average time
for a company to be on top is about 5 years, (Fairchild, National, TI,
NEC, Toshiba, Intel, Samsung, all were number one in less than 32
years-which just happens to fit into a random walk entropic/fractal
theory of 4.3 years, quite nicely.) There are over two hundred major
players in the market internationally, so it is quite vicious, (105
listed in the Yellow Pages in Silicon Valley within a 10 mile radius.)

The issue of dumping (trade indiscretions, vs., protectionistic
lobbying,) has been discussed at some length in the executive suites
of the Valley. Many economists have been hired for advice.

Its a complicated issue. If it was not for dumping, your PC would cost
about an order of magnitude more than it does. So, in this sense, the
society is better off not intervening in dumping issues-I mean if
someone wants to sell something below cost, why not take advantage of

On the other hand, it is very hard for a company to prosper in an
environment where the chances are that someone in the world is selling
competitive products at below costs to gain market share. Developing
new products, market share, R&D, the payroll, etc., all suffer.

Note that it is a different take off on Kenneth Arrow's so called
Impossibility Theorem which states that there is no rational way to
determine which has the higher priority, manufacturing infrastructure,
or inexpensive products for the masses, (ie., the intransitiveness of
the social welfare function.) Its a complicated question.


BTW, traditional methods of running a business don't work very well in
the SC industry. The ROI curves are very non-linear. After the first
part is manufactured, the next millions cost only incremental cost of
sales to deliver. (I mean, what's a teaspoon of sand cost?)
Incremental costs are everything. That's how the US lost to Japan in
the late 80's. They taught us that. One runs a factory at capacity,
and gets what it can in the marketplace for the production. The SC
industry is more like the nuts-bolts-and-rivets of the electronics
industry-everything is value added from there on up.


John Conover,,

Copyright © 1999 John Conover, All Rights Reserved.
Last modified: Sun Nov 14 00:02:00 PST 1999 $Id: 990902112427.19167.html,v 1.0 2001/11/17 23:05:50 conover Exp $
Valid HTML 4.0!