IP Ties and Microsoft's Rule of Reason
| Author | Herbert Hovenkamp |
| Published date | 01 June 2002 |
| Date | 01 June 2002 |
| DOI | http://doi.org/10.1177/0003603X0204700206 |
| Subject Matter | Article |
The Antitrust Bulletin/Summer-Fall 2002
IP ties
and
Microsoft's
rule
of
reason
BY HERBERT
HOVENKAMP*
I.
Introduction:
tying
arrangements
and
the
rule
of
reason
369
In its
Microsoft
decision the D.C. Circuit Court of Appeals broad-
ened the scope
of
intellectual property (lP) rights by adopting a
rule of reason for tying arrangements that met all the requirements
for tying's per se rule.' Microsoft was accused of unlawful tying
because it "commingled" the code for its Windows operating sys-
tem and its Internet Explorer (IE) browser, making it impossible
to sell them separately or for users to separate thern.? Microsoft
also refused to permit computer manufacturers (OEMs) to remove
the IE icon from the desktop or the Windows "Start" button, and
did not place IE on the add/remove programs utility. It required
several Internet service providers to favor IE as a browser, gave
independent software vendors preferred treatment if they devel-
oped interfaces and commands that either invoked IE exclusively
*Ben V. &Dorothy Willie Professor
of
Law, University of Iowa.
AUTHOR'S NOTE: Thanks to
Roger
D. Blair, Christina Bohannan
and
Mark Janis for commenting on a draft.
United States v. Microsoft Corp., 253 F.3d 34 (D.C. Cir.), cert.
denied, 122 S.Ct. 350 (2001).
See 253 F.3d at 64.
©2002 by Federal Legal Publications. Inc.
370
The antitrust bulletin
or required the software to favor IE, and agreed with Apple on a
set
of
protocols for Mac Office that required Apple to favor IE in
its operating system.'
Under the idiosyncratic and controversial per se rule for tying
arrangements, Microsoft's tie was unlawful
if
Microsoft held sig-
nificant market power in its Windows operating system; if Win-
dows and Internet Explorer were "separate products," a tying law
term
of
art; and if a
"not
insubstantial" amount of tied market
commerce was involved."
The
first and third requirements were
clearly met. The D.C. Circuit did not decide the "separate prod-
ucts"
issue, but under existing law that requirement was almost
certainly met as well. Nevertheless, the court refused to apply the
per se rule to this tie. Rather,
it
remanded the case for treatment
under the rule of reason, saying almost nothing about what this
entailed. The government subsequently dropped the tying claim.
Here, Iargue three points. First, the D.C. Circuit's develop-
ment
of
a rule
of
reason for the software ties before it was com-
mendable
and
long overdue. Second, concerns for stare decisis
need not prevent alower court from fashioning such a rule of rea-
son until after the Supreme
Court
had spoken on the issue. But
third, the
court's
limitation
of
the new rule to computer software
bundles
involving operating systems is both irrational and per-
verse, providing more lenient treatment for one of the relatively
few types
of
ties that threaten serious competitive harm.
Abroad view
of
IP rights embraces relatively easy patentabil-
ity
or
copyrightability,
liberal
scope,
and
expansive
immunity
from antitrust challenge or misuse claims. This view is built on
the premises that the IP statutes serve mainly to create marketable
property rights, that most
of
these rights confer little or no eco-
nomic power, that the transaction costs
of
identifying and trading
them
are low,
and
that the
simple
expectation
of
amarketable
Id. at
84-85.
4
For
a
brief
synopsis
of
the
requirements,
see
HERBERT
HOVENKAMP,
FEDERAL
ANTITRUST
POLICY:
THE
LAW
OF
COMPETITION
AND
ITS
PRACTICE
§§ 10.1, 10.3-10.5, 10.7 (2d ed. 1999).
Microsoft :371
right
spurs innovation. As a
result,
broad
protection
generally
poses
a
modest
threat to
competition
and
the
maximum
set
of
incentives to innovate.
By the criteria we ordinarily apply to markets, however, one
would predict that IP markets function poorly. Two characteristics
seem
particularly problematic:
lack
of
good
information
about
ownership and extreme product differentiation.
In a well functioning market the bargaining parties are clear at
the outset about who owns what. There is little dispute about who
the buyer is, who the seller is, and whether atransaction is neces-
sary
at all.
When
I
walk
into
an
A&P
grocery
to
purchase
a
banana, Ibegin with a fairly strong assumption that prior to the
transaction the banana is not "public domain." Rather, it has been
made
subject
to ownership:
A&P
is the
owner,
so
if
I
want
it
I must purchase it from A&P. By contrast, patent rights are ren-
dered ambiguous by uncertainties about patent validity, scope and
claim construction, misuse, and infringement. In the world
of
IP
rights, simply determining who owns what
or
whether anything is
owned
at
all
is costly.
Further,
broad
IP
coverage
for
modest
improvements
creates
the
anticommons
problem
that
useful
improvements require many transactions.'
Asecond characteristic
of
IP markets that can cause
them
to
function poorly is extreme product differentiation. Something that
is no different from public
domain
technology
or insufficiently
distinct from other patented technology is not patentable. So the
fact of patentability implies at least modest uniqueness. Markets
For a particularly
pessimistic
view, see
LAWRENCE
LESSIG,
THE
FUTURE
OF
IDEAS:
THE
FATE
OF
THE
COMMONS
IN
A
CONNECTED
WORLD
(2001); and Lawrence Lessig, The Death
of
Cyberspace, 57
WASH.
&
LEE
L.
REV.
337 (2000). Other good literature
includes
Robert P. Merges,
Intellectual Property Rights
and
the New Institutional Economics, 53
V
AND.
L.
REV.
1857 (2000); Michael A. Heller, The Boundaries
of
Pri-
vate Property, 108
YALE
L.J. 1163 (1999); Michael A. Heller &Rebecca
S.
Eisenberg,
Can Patents
Deter
Innovation?
The
Anticommons
in
Biomedical Research, 280
SCIENCE
698
(1998); Michael A. Heller, The
Tragedy
of
Anticommons: Property in the Transition from Marx to Mar-
ket, 111
HARv.
L.
REV.
621, 621 (1998).
To continue reading
Request your trial