COMMENT ⊗   VALID 00003 PAGES
C REC  PAGE   DESCRIPTION
C00001 00001
C00002 00002	.require "memo.pub[let,jmc]" source
C00009 00003
C00010 ENDMK
C⊗;
.require "memo.pub[let,jmc]" source
.cb NATURAL MONOPOLIES IN INFORMATION SERVICES


	For some years it has been technically and economically
feasible to make a transition to a computer and communication
based information and commercial system.  In such a system, all
books, magazines, business data and personal files will be stored
in a variety of electronic media, and all commercial communication
will take place between computer systems.  Decisions will be manual
in so far as is desired, but people and institutions will increasingly
discover how to delegate the more routine aspects of their business
lives to computer programs.

	In my view this development has proceeded more slowly than
is desirable, and the reasons for the delay have been primarily
a lack of enterprise in business rather than a lack of technology.  The
Government and technophobic sentiments have also contributed
obstacles, but much that could have been done and to which no
objections have been raised has been neglected.  However, as
the technology advances, the payoff to the innovators becomes
more and more apparent, and those companies that pioneer will
reap increasing rewards relative to those who lag.
Probably there are more obstacles than I know about, so the
lag may not be as great or as disgraceful as I am inclined to
think.

	One of the problems is that some of the new systems that
are required are natural monopolies and others are not.  The
concept of natural monopoly is an old one, and the prototype
natural monopolies are the distribution of natural gas, and the
telephone switching system.  With our present level of technology,
it would be extremely expensive to have two competing sets of gas piping
or electric wiring or roads in a neighborhood.  These services
are natural monopolies rather than artificial monopolies arising
from government action or one company buying up the others.
Changes in technology can change whether a service is a natural
monopoly.  For example, it is now apparent that the provision
of telephone instruments is not now a natural monopoly, and
competition in this area has already produced useful products.

	It seems that in future computer based information systems,
some services are natural monopolies and others are not.  Those
services that are natural monopolies are appropriately performed
either by the some level of government or by regulated private
monopolies.  Which will be done depends partly on political
factors, but some are more appropriate for government than others.
Thus few propose that police and defense are appropriate for
private industry, and almost as few want the government to take
over the American telephone system.

	The object of this paper is to identify the natural monopolies in
information services and to discuss various ways in which they might be
provided.  I shall assume that our society continues capitalist with
government regulation.  My bias is against monopoly, and I will propose
that some services be provided by competing suppliers in some cases
in which monopoly might be more efficient.

	One aspect of this preference arises when a service can be
provided either by a product which each user of the service must
buy for himself or by a system designed to provide the service
to everyone or at least to all subscribers.  A product is preferable
to a system even at somewhat greater cost, since different tastes
can be accomodated.  Even more important, we can't be confident that we
fully understand all the characteristics of the system being
designed and the demands of future users.  Therefore, it is preferable to
have products get obsolete than whole systems.

	Nevertheless, sometimes it is impossible or too expensive
for everyone to buy his own.  Try to imagine everyone supplying
his own roads.

	We begin with electronic mail.  Almost everyone seems to
assume that it is naturally a system, but it seems to me better to
develop it as a product.

	The world is supposed to be in transition to the %2post-industrial
society%1 in which more of the GNP will be in services rather
than in production.  Very nice but many of us in the computer
field are aware that much of the growth in non-production parts
of the economy is merely an expression of the fact that in spite
of computers, productivity in administration has grown far more
slowly than in production; in fact in may have even declined.

Maybe in was never a natural monopoly, but maybe the telephone
companies were once correct that the design of a