Tuesday, April 9, 2013
Posted by Michael Carrier
How Antitrust Agencies Can Address PAE Activities
Michael A. Carrier, Professor, Rutgers Law School
Post adapted from longer article
Antitrust can—and should—regulate PAE activity. As I have explained in more detail elsewhere,
the antitrust agencies can rely on guidelines, Clayton Act Section 7, Sherman Act Section 1, and FTC Act Section 5. This post highlights the six actions the agencies can take to address PAEs.
1. Offer guidance about potential harms from patent aggregation
First, the Department of Justice and Federal Trade Commission can revise the 1995 Intellectual Property Guidelines, or at least offer guidance, on potential harms from the creation and exploitation of massive patent portfolios. The agencies should reevaluate the Guidelines’ exclusive focus on the procompetitive justifications of IP combinations in “facilitat[ing] integration of the licensed property with complementary factors of production,” which “lead[s] to more efficient exploitation . . . benefiting consumers through the reduction of costs and the introduction of new products.”
While these justifications will still be relevant in many settings, the agencies can recognize a fact that was not so obvious in the mid-1990s: that massive patent portfolios can be used offensively, and can be valuable primarily because of their size rather than the validity of each patent in the portfolio. The agencies can explain that large portfolios could present anticompetitive effects including patent holdup, raised rivals’ costs, and even increased price or reduced innovation.
2. Promote transparency
Second, the agencies could require transparency in evaluating patent-based behavior. Companies that are sued (or face threats of litigation) must know who is suing them, what patents they are being sued on, and which patents the plaintiff owns. It is difficult, for example, to engage in licensing negotiations with entities that have no “website, employees, or offices.”
Despite the need for transparency, much PAE activity today is hidden beneath a labyrinth of
shell companies. Acacia’s subsidiaries control 250 patent portfolios. And Intellectual Ventures has used at least 1276 shell companies to purchase and hold patents.
The category of secret activity includes the related concern of “privateering,” a concept that traces
back to “state-sponsored piracy” in which governments allowed private parties to “seize the property of the state’s enemies.” In the PAE setting, privateering can result in aggressive third parties scaring customers and suppliers. A practicing company, for example, could reject a license offered by a PAE, but then be sued for even more by a privateered third-party PAE. The sheer scale of such networks could offer clues about competitive effects. Gathering evidence on these entities is consistent with the FTC’s 2010 changes to the Hart-Scott-Rodino Act, which introduced the concept of “associates,” which includes entities managed by the acquiring entity. In analyzing PAE behavior, the agencies must be able to obtain as much information as they need to determine competitive consequences.
3. Prohibit transfers to PAEs that refuse to adhere to previous standards-based licensing promises
Standard-setting organizations (SSOs) allow industries to adopt a common technology. SSOs often adopt policies to mitigate the power acquired by a company with patents incorporated into the standard. One of those policies requires patentees to agree, before selection, to license their patents on reasonable and nondiscriminatory terms (RAND) if selected.
RAND promises are typical for standard-essential patents (SEPs), which are essential to the implementation of a standard. But a potentially fatal loophole is created if RAND promises could be avoided by the transfer to a subsequent party that is not bound by the promise. To pick one example, after the agencies approved the acquisition of the Nortel patent portfolio by the Rockstar consortium based on promises made by members (Apple and Microsoft) to agree to RAND licensing, the Rockstar CEO publicly stated that the consortium “isn’t bound by the promises that its member companies made” since “[w]e are separate” and the promises “do not apply to us.”
The agencies should make crystal clear that they will not allow any acquisition by a PAE (or an operating company) that does not agree to honor RAND promises made by its predecessor. Evading such a solemn obligation through transfer threatens to make a mockery of the RAND promise at the heart of the standard-setting process.
4. Use PAEs’ distinct incentives to employ Clayton Act Section 7 when “plus” factors are met
The ability and incentive to exercise market power and harm competition is crucial to the analysis of
mergers and acquisitions. PAE acquisitions can allow incumbent companies to harm rivals due to their unique characteristics. Litigation between operating companies is constrained by risk symmetry, as mutually assured destruction (MAD) fosters settlement and reduces litigation.
PAE litigation, in contrast, is characterized by risk asymmetry. PAEs have no real disincentive to sue since they do not manufacture products, and thus do not face the possibility of countersuit. As a
result, there is no MAD. PAEs also benefit from (1) producing few documents; (2) not facing disruption to their business since “litigation and licensing are their business”; (3) not confronting reputational harms when suing or threatening to sue; (4) not encountering constraints from being a repeat player in a standards organization; and (5) not facing customers exerting pressure to settle litigation or shareholders skeptical of patent enforcement.
Not every patent acquisition by PAEs will automatically violate Section 7. But the different incentives confronting PAEs call for careful scrutiny when these entities are involved in acquisitions. Six “plus” factors exacerbate the concerns, in some cases pushing the acquisition over the line of antitrust liability.
The first occurs when facts demonstrate incentives to harm competition beyond those confronting
the typical PAE. For example, after acquiring 2,000 Nokia wireless patents in 2011, Mosaid agreed
to fund the portfolio acquisition “through royalties from future licensing and enforcement revenues,” and its ownership of the patents was “subject to minimum future royalty milestones.” If Mosaid did not reach certain milestones, it would lose the right to transfer the patents and cede ownership of the patents altogether.These provisions encourage the aggressive filing of lawsuits.
A second, related, plus factor could be present when a PAE follows an investment model designed to
provide a certain level of returns to investors. In this case, again, the PAE has a strong incentive to sue or threaten to sue. Intellectual Ventures has raised more than $5 billion from operating companies, institutional endowments, and wealthy individuals who expect venture-capital-like returns. One study estimated that to be successful, Intellectual Ventures would need to amass a
25% return each year, leading to a lifetime revenue expectation of between $40 billion (over a 10-year period) and $244 billion (over a 20-year period).
A third plus factor could be revealed by a pattern of late filing of suits in relation to patent issuance.
Patents that are used to protect companies against rivals are typically litigated shortly after their products are introduced. In contrast, PAEs tend to assert patents late in the patent term and litigate their patents “to the verge of expiration.” While this third factor alone would not be sufficient to
show antitrust harm, PAEs that consistently file lawsuits at the end of the patent term provide evidence of patents that are not directly used to recoup the rewards of the invention.
A fourth factor could be specific PAE conduct blocking needed disclosure. For example, in the
operating-company context, Barnes & Noble alleged that Microsoft refused to disclose the patents on which it was suing Barnes & Noble. Microsoft apparently would not disclose this basic information unless Barnes & Noble executed a non-disclosure agreement.
A fifth factor could be presented by the combination of a PAE and an operating company where
the operating company has an incentive and ability to harm its rivals. For example, the operating company could acquire patents with a PAE and then enlist the PAE to sue its rival. This could present a combination of the operating company raising its rivals’ costs and the asymmetry advantages possessed by PAEs.
A sixth factor could be presented by disaggregation of a portfolio. The issue of “royalty
stacking” occurs when multiple claims for royalties are “stacked together,” which can lead to higher costs that increase the price charged to consumers or that drive rivals out of the market.
This list is not exclusive, and other plus factors could be identified. The point is that the unique factors presented by PAEs should raise the antenna of the agencies in considering acquisitions under Section 7. The existence of plus factors could push the case over the threshold of antitrust challenge.
5. Monitor collusive activity
Although Clayton Act Section 7 will be the most natural setting to analyze the antitrust effects
of PAE activity, Sherman Act Section 1 also might be applicable in certain settings involving collusion. One example discussed in the literature involves alleged monthly calls between RPX and Acacia in which “Acacia describes the producers they are in the process of targeting and the patents they will assert against the producers.” In return, Acacia “names a price for the patents in question” and “RPX purchases the patents if it wishes.” Such conduct implicates collusive activity that should be explored under Section 1.
Another type of interaction that could conceivably implicate Section 1 involves collusive relationships
between operating companies and PAEs discussed above. For example, if Nokia and Microsoft pooled their patents and enlisted Mosaid to use the patents to sue competitors, that could present a concern of raising rivals’ costs or removing a rival from the market. In any of these cases, conduct could be analogized to that considered in United States v. Singer Manufacturing Co., in which the Supreme Court struck down an arrangement in which three companies pursued a “common purpose” to
suppress competition “through the use of [a] patent.”
6. Consider the use of FTC Section 5
PAE behavior also could potentially constitute an unfair method of competition prohibited by FTC Section 5. Section 5 reaches beyond antitrust but needs a justifiable framework based
on well-defined limiting principles. One potential setting involves incipient or “frontier” conduct that has recently developed and that does not fit into well-established antitrust categories. Some commentators underscore the propriety of Section 5 in this setting on account of its
prospective application and the lack of a private damages recovery.
Although the FTC has not yet offered a precise framework for Section 5, one conceivable framework could require the factors of (1) market power, (2) a lack of a non-trivial efficiency (i.e., behavior not
justified by purposes of patent system), (3) causation, and (4) consumer harm. In the PAE setting, (1) market power is possible in technology markets, (2) PAEs’ revenue-driven licensing will often not be connected to product creation, and the behavior (3) seems likely to cause competitive harm and could result in (4) higher prices or reduced innovation for consumers. The “plus” factors discussed above
in the context of heightened ability and incentive to exercise market power are relevant here in exploring the existence of justifications and consumer harm.
PAEs confronting Section 5 prosecution should be given the opportunity to show that the proceeds they collect benefit inventors in non-trivial proportions. But absent such a showing, and given the articulation of limiting principles, the application of Section 5 could be appropriate given the potential unjustified consumer harm in a setting that lacks antitrust precedent.
PAEs present a challenge to antitrust law. But even if the empirical evidence on PAEs across society is not yet fully collected, and even if certain PAEs can justify some of their conduct, that does not
mean that all PAE activity is immune from antitrust scrutiny. For if it was so protected, then the most aggressive and unjustified behavior, undertaken by PAEs with the greatest market power and
largest portfolios, and inflicting the greatest harm on rivals and consumers, would fall through the antitrust cracks. Antitrust enforcement cannot automatically be shunned in a context that presents new and powerful opportunities to inflict anticompetitive harm.