[BITS] Week 9 Standards Readings

Rohit Khare (rohit@bordeaux.ICS.uci.edu)
Tue, 03 Mar 1998 14:29:30 -0800


[BITS] Week 9 Standards readings

[As usual, the indented bits are Phil's words -- RK]

Week 9 / Antitrust

The most topical policy question relating to standards is antitrust
law. A company that owns proprietary de facto standards possesses a
license to print money, particularly when those standards enable
the company to leverage its control into new markets. The law is
struggling to comprehend this phenomenon in traditional antitrust
categories, and we will consider a range of opinions.

As usual, I have spent long hours reading and rereading the entire
set, but don't have the time to do them proper justice... One of the
main functions of this set was to yank me out of a myopic focus on
Internet standards and recall the wealth of other examples we have in
common discourse about the computer industry: Dell's and Unisys's
post-facto assertion of submarine patent rights in VL-bus and GIF; the
underpinnings of the IBM case (including predatory preannouncement);
and the slew of early '90s Microsoft 'abuses': Stac's drive
compression, Sidekick's tools, the Intuit buyout, the MSN
'juggernaut,' the OS/2 'false dealing', ... Myopia is a funny thing.

Mark A. Lemley, Antitrust and the Internet standardization problem,
Connecticut Law Review 28, 1996, pages 1041-1094.

This is a lengthy but lucid explication of the theory of standards
lock-in as applied to antitrust law. Its conclusions are skeptical.
Antitrust law is held to be incapable of regulating
standards-driven industries, and standards organizations are held
to be overwhelmingly pro-competition.

The main source of all this contextualizing is Lemely's able
recapitulation of the brief history of "The Internet Antitrust
Problem." He recapped some of the causes of swift and decisive
standardization: network externality, compatibility links between
products/markets, and the inherent economics (high fixed costs on both
sides: sw development and training). He directs attention away from
the standards-setting process as the likely locus of antitrust abuses,
though -- at least in the Internet space, if not for OS design. In his
recommendations, he reemphasizes that natural monopolies are not in
Sherman or Clayton's balliwick; and even if there were flaws in
standards competition on today's Internet, there's no effective
structural remedies, so why bother prosecuting?

As for predatory pricing, he is correct in noting that participants
willingly and knowingly discount the fact of competition into their
business plan -- no one goes into business expecting Microsoft to
charge full freight on v1-3 -- but incorectly dismissive that we
should be "concerned... only to the extent it is asymmetric... one
firm is much better situated to survive a predatory battle." The
latter point, to me, is the critical issue in the Microsoft context,
and not addressed thoroughly by any of the readings -- although it's
not prima facie illegal, either. The bottom line is that Microsoft has
such an overwhelming bank account that it can wait out any other
competitor in any one market of the software industry because of their
interrelated rents. The software startup, or content startup, seems
altogether a flea on an elephant...

[Note for my W3C case study: "the use of antitrust doctrine to compel
access to a standards-setting organization should probably be rare"]

Bryce J. Jones, II, and James R. Turner, Can an operating system have
a duty to aid its competitors?, Jurimetrics 37(4), 1997, pages
355-394.

When markets are dominated by proprietary standards, it is commonly
argued that the standard constitutes an "essential facility" in the
sense provided by American antitrust law. This argument is almost
never successful in court, but Jones and Turner lay it out
nonetheless in the case of Microsoft Windows.

I can see why it's so rarely successful in court: even in the Terminal
Railroad Association case (a cadre of railroads owned all the
switching and bridging capacity across the river in St. Louis), we see
a very reluctant decision to offer equal access, without presumption
of past abuse. Furthermore, unilateral refusals to deal are also
pretty defensible, as when Intel cuts off Digital or Microsoft
withholds betas. But, the inkling of hope is that the courts did open
up Bell's technical standards, after all. And there certainly are
curiousities, like Microsoft suing developers reverse-engineering
undocumented calls their own code uses. Finally, they consider if
Microsoft is 'leveraging' its monopoly markets in any classical sense:
using rents from Office to drown out web zines.

And that's the nub of the issue: I believe the socially-harmful
behavior is the the leverage across MS's vast product line, but it's
not clear it's from leveraging specific monopolies. There are
successful products, but I hesitate to call any of them a monopoly
(except the base OS). It's just that the net social effect is that a
lot of alternative tools, like the AOLPress editor I'm using right
now, never survive the development test of bankrolls against arguably
inferior MS products. But it may be that it's just a matter of scale
(they have grown to be in so many markets and are 'learning' in all of
them simultaneously) and greed (they don't have any committment to
letting a thousand flowers bloom: they'll move into any market as soon
as it's possible) -- can't argue with either, really, except on
artistic merit!

Note: some may complain I'm skipping the obvious tying between
monopoly OS share and applications: nowadays, I increasingly fear the
interactions between the applications. The damage has already been
done on 'undocumented calls' and their feints on emphasizing COM/OLE,
Windows networking, fonts, etc, where they announced their strategy
early and often, but no one listened except their internal developers
-- who got years of lead time in the market for their 'integrated'
products.

S. J. Liebowitz and Stephen E. Margolis, Should technology choice be a
concern of antitrust policy?, Harvard Journal of Law and Technology
9(2), 1996, pages 283-318.

These scholars, as we have already seen, are highly skeptical of
the network externality theory of market failure. In this article
they apply their critiques to antitrust issues, arguing that the
foresight and sophistication of market participants is sufficient
to ensure that the best technology wins in a standards competition.
The consequence, of course, is that the government has no reason
intervene.

A useful survey article, neatly recapping the false fables behind the
oft-retold QWERTY, VHS, and Win3.1 fables. The basic framework is that
contemporaneous compatibility standards *are* winner-take-all markets;
split-decisions are invariably unstable. And, the best technology can
win. (see below).

Recommended

W. Brian Arthur, Competing technologies: An overview, in Giovanni
Dosi, Christopher Freeman, Richard Nelson, Gerald Silverberg, and Luc
Soete, eds, Technical Change and Economic Theory, London: Pinter,
1988.

Arthur's economic model of technological lock-in through positive
returns to scale suggests that monopolies can arise through
self-reinforcing market mechanisms, and that those mechanisms do
not necessarily select the optimal technology. The model abstracts
away from far too many aspects of real-world technology markets to
be evaluated in isolation, but it suggests many possible lines of
research. I have included it as a recommended reading because is
cited heavily in both the popular and legal literature.

Arthur thinks the worse technology can win because it's really a race
to the critical-share point, and adoption is an ergodic random walk --
the very meaning of path-dependent. True, systems like NeXTstep failed
because not enough people adopted it -- but was that because some
costs weren't internalized? That's the flaw in overbroadly applying
Arthur: he really does think there are substitutes, and that's not
true of any technological competition I know of. NeXTs, for example,
may have been doomed by Windows, but not because of the number of
Windows boxes, but because of the (sad?) truth that most people aren't
that high on the demand curve: they *don't* develop custom software,
don't particularly care how multimedia their mail is, and certainly
don't care about geek-joy-buzzers like DSPs (at least ten years ago,
they didn't. Man, I feel old...)

James J. Anton and Dennis A. Yao, Standard-setting consortia,
antitrust, and high-technology industries, Antitrust Law Journal 64,
1995, pages 247-265.

Standardization activities, whether conducted through formal
standards organizations or through private consortia, strongly
resemble illegal marketplace collusion. Society obviously needs
standards, however, and so it is important to determine the
difference between socially necessary standardization activities
and standardization activities that mask illegal market fixing.
This article applies some of the economic theories that we've
already read to an analysis of the problem.

This is a critical, critical article in the context of my W3C case.
I'll have to discuss it in detail there. Short version: W3C's probably
safe, but I'd be happier if it understood the risks in the first
place.

Joseph Kattan, Market power in the presence of an installed base,
Antitrust Law Journal 62, 1993, pages 1-21.

On one analysis, the dispute between the Justice Department and
Microsoft concerns what antitrust lawyers call "tying". This
article analyzes the leading tying case, in which Kodak required
the companies that bought its photocopiers to purchase parts and
supplies for those copiers exclusively from Kodak. The intuition is
that this constitutes restraint of trade to the extend that the
vendor has market power, and the question then arises of what
market power is and how one measures it.