Editorial by David A. Balto and Brendan Coffman. Mr. Balto and Mr. Coffman are antitrust attorneys in Washington D.C. whose representations include high technology firms. In addition to his practice, Mr. Balto was formerly a policy director of the Federal Trade Commission, attorney-adviser to
Chairman Robert Pitofsky, and an antitrust lawyer at the U.S. Department
of Justice.
When Patent Pools Attack: Competitive Concerns from the Devolution of MPEG LA
By David A. Balto and Brendan Coffman
Patent pools pose a unique
challenge to antitrust enforcement. On the one hand they solve
collective action problems and allow participants to achieve economies
of scale that would otherwise be impossible. Patent
pools enable market participants to join complementary intellectual
property to better manage those IP rights. As the Department
of Justice noted, patent pools may “provide competitive benefits by
integrating complementary technologies, reducing transaction costs,
clearing blocking positions, and avoiding costly infringement
litigation.”
On the other hand, patent
pools can create competitive problems by conferring market power on a
group (in the case of member-owned patent pools) or entity (in the case
of stand-alone patent pools). Thus, the antitrust
enforcement agencies have always been concerned if the pools are
over-inclusive and include competing technologies. Where this is the
case patent pools may leverage their position to interfere with
competition in at least two ways. First, a pool may interfere
with the relationships and convergent intellectual property rights of
pool members. This result may be particularly pernicious if one of the
patent pool participants seeks to introduce a new business model that
may compete head-to-head with the pool’s customers.
Second, pools may leverage their market power within the supply chain to
extract higher or duplicative royalties on downstream purchasers of the
pool’s licensed technology. These concerns are typical
in the high-tech/intellectual property field, and come up very often in discussions of patent assertion entities, or PAEs.
The key for regulators and
policy-makers alike is to ensure that patent pools remain committed to
the procompetitive collaborative efforts that justify their creation,
and do not change their business model to exploit
or create market power. In theory this sounds simple. In practice, it
is anything but.
History of MPEG LA
MPEG LA is an interesting
case study in the trade-off between the efficiencies of pooling
arrangements and the potential competitive risks. MPEG LA has been a
personification of the debate since the group’s incipiency.
Formed in 1996 to administer the pooling of 27 digital video patents
deemed essential to the MPEG 2 video compression technology into a
single portfolio, MPEG LA aggregated and offered a collective license to
27 patents from eight companies and Columbia University.
MPEG LA presented its idea to the Department of Justice, seeking ex ante approval through the DOJ’s Business
Review Procedure [pdf].
The DOJ responded in an oft-cited
14-page letter containing 49 footnotes. The DOJ concluded that the
joint licensing of the MPEG 2 essential patents is likely to provide
significant cost savings to licensors and licensees alike because “the
licensing arrangement have features designed
to enhance the usual procompetitive effects and mitigate potential
anticompetitive dangers.” The DOJ also emphasized that the structure of
the patent pool appeared well-suited to ensure that only truly
essential patents become included in the pool, and that
the patents in the pool remain complements, not substitutes. Finally,
the DOJ opined “there does not appear to be any potential for use of the
Portfolio license to disadvantage particular licensees” because the
pool’s most-favored-nations clause ensures equivalent
rates to all takers on non-discriminatory terms. In approving the
arrangement, the DOJ laid forth the following four guidelines for when a
patent pool may gain approval:
1. The patents must be valid and enforceable;
2. The pool must not aggregate competitive technologies and set a single price for them;
3. An independent expert should determine whether the patents are essential; and
4. The pool must not disadvantage competitors or facilitate collusion
The DOJ’s approval of MPEG
LA’s turned in large part on one fundamental and pivotal fact: MPEG
LA’s operating procedure called for the participation of an independent
expert at every turn. In fact, the Business Review
Letter mentions the independent expert sixteen times.
There was significant concern
raise about the MPEG LA pool. As one antitrust scholar exclaimed “the
anticompetitive potential of the MPEG LA patent pool is enormous. The
DOJ’s approval of the pool validates a collectively
enforced monopoly over a fundamental communications standard.”1 Despite
this and public criticism, the DOJ approved the behavior. MPEG LA
stated publically that the company’s mission was to “provide a service
that brings all parties together so that technical innovations can be
made widely available at a reasonable price.”
As history shows, it is not long before a company with this model
succumbs to the pressure of exercising and extending its monopoly power.
Emerging Competitive Concerns
In recent years, MPEG LA has
been accused of inhibiting the innovation that it was designed to
foster. Notably, the company’s practice of charging high licensing fees
for patents that are near or past expiration has
led critics to assert that the firm has placed profit above its core
mission of cheap and accessible licensing of digital video patents.
Technology market players have also alleged that MPEG LA has violated
the terms of its original agreement with DOJ by failing
to invite oversight of its licensing practices by independent experts,
and neglecting to adhere to FRAND guidelines. A firm that was once a
model (at least in theory) of the potential benefits from collaboration
has morphed into one of the industry’s most
notorious and most harmful players.
Interestingly MPEG LA embodies both of
the concerns with patent pools outlined above and by the DOJ in its
approval of the patent pool’s licensing structure. First, MPEG LA may
overtly inhibit the ability of
its members to develop any technology that may compete with customers of
the MPEG LA pool. Most notably, MPEG LA was involved in a dispute with
pool-participant Google. In 2010 Google introduced WebM, an
open-source solution to uploading videos to the web.
Google designed WebM to serve as an alternative to H.264, the primary
video compression technology in Microsoft and Apple devices that is
covered by the MPEG LA patent pool. All three are members of the MPEG
LA patent pool, and
pay royalties to the company (although Apple contributed only one patent
to the pool). MPEG LA asserted that Google’s WebM product practices
on or infringes patents in the pool, and demanded that users of the
WebM product pay
royalties to the pool. A license from Google would not be enough. In fact, in early 2011 MPEG LA instructed patent owners to inform
the pool of patents they believe the WebM product uses. The DOJ
initiated an investigation into whether MPEG LA is acting
anticompetitively by trying to quash the Google WebM product through
assertion of patents as a patent pool. The investigation appears to be
on-going.
It is not only Google that MPEG LA has targeted – German
software company Nero also filed complaints with the DOJ alleging
that MPEG LA is illegally maintaining and enhancing its monopoly power
by failing to adhere to the conditions of the DOJ Business Review
Letter. Nero points out that MPEG LA does not rely
on independent experts as it said it would, but instead allows its
membership to drive patent inclusion decisions. Nero also argues that
MPEG LA did not offer the patents in question on fair, reasonable, and
non-discriminatory terms as required.
MPEG LA has also devolved
into a PAE. A PAE may accumulate and leverage patents as a business
model by taking advantage of economies of scale and the high cost of
patent litigation. A common business model of PAEs
is to divide patents among shell companies, thereby making it difficult
for defendants to identify the original patent owner and impossible to
assert a counterclaim. MPEG LA’s President, Larry Horn, is also the
president of MobileMedia Ideas, a known PAE jointly
owned by MPEG LA, Sony, and Nokia. The firm controls patents for
technologies used in mobile phones, computers, tablets, cameras, and
videogame consoles, among other devices. MobileMedia Ideas has engaged
in a significant amount of successful litigation, and
continues a business model of leveraging these patents against numerous
operating companies, including those that have business
relationships with MPEG LA.
What is the Lesson?
The lesson cannot be that
patent pools are inherently bad. Pooling complementary products
displaces litigation in favor of innovation, and is absolutely a win-win
for consumers and manufacturers alike. Instead, the
lesson must be that continued oversight is necessary in the case of
patent pools, especially when there is so much incentive for a pool to
step outside the confines of approved activity, and to begin leveraging
its unique market position anticompetitively.
With that in mind, there are
some basic steps that can be taken to prevent the problems identified
above from replicating. First, the antitrust agencies should disapprove
of patent pools that place a restriction on
how a participant may deploy its own technology. It is appropriate to
require a patent owner to make technology available on certain terms and
to as large an audience as possible, but it is inappropriate to
restrict the patent owner from actually competing.
Second, the agencies should condition approval of a patent pool on the
continued use of independent experts. A neutral voice is essential to
prevent a pool from devolving into a firm seeking to include as many
substitute patents as possible. Third, the agencies
should condition approval of a patent pool on a commitment to maintain
autonomy and to refrain from engaging in any other business activities
outside of the pooling arrangement. It is too easy for a patent pool to
learn the business of its members and position
itself in the market as another barrier to entry.
1Steven C. Carlson, Patent Pools and Antitrust Dilemma, 16 Yale
J. on Reg. 359, 372 (1999).
