Justice News

Assistant Attorney General Bill Baer Delivers Remarks at the 19th Annual International Bar Association Competition Conference
Friday, September 11, 2015

Reflections on the Role of Competition Agencies When Patents Become Essential

Good morning.  I’m here today to discuss some of the ways intellectual property law and antitrust law work together to promote and protect innovation.  These concepts may not be novel.  But they are important.  Competition propels technological advances as rivals and potential entrants race to be first to bring a product to market or to improve an existing one.  Our intellectual property laws create incentives for innovation and its commercialization by establishing enforceable property rights.  The Sherman Act does not make unlawful the attainment of market or monopoly power solely through business acumen or superior skill.  Just the opposite.  We recognize that firms that take risks and innovate deserve to reap the profits from offering a superior product or service.  Consequently, we stay out of the way of disputes over the amount of royalties that patent holders might lawfully demand.

In recent years it has become well understood that the competitive process can suffer when the value of a patent is enhanced by becoming essential to a standard and patent holders seek to exploit that added value by failing to keep the commitments they voluntarily make about how they will license these patents.  Antitrust enforcers and competition advocates need to address those behaviors in appropriate circumstances.  And competition agencies need to be prepared to give guidance to standards-setting organizations (SSOs) on what ex ante rules can legitimately address concerns with patent hold up without risking antitrust challenge.  At the same time, we should be cautious about when and where the government should intervene.  Consumers and competition are not well-served by enforcement actions aimed at lowering the price of licenses when patent holders and potential licensees are engaged in a dispute solely about the terms of a license, or even the need for one.

Let me start with collaborative standard setting.  We all can agree that collaboratively set standards are critical to innovation in many industries.  Standards that are widely adopted can provide timely and effective solutions to technical problems.  By agreeing on a collaborative standard, firms may be able to avoid the costs and delays of a standards war in the marketplace, quicken the introduction of innovative products and services, and create uniform specifications for entire technological ecosystems in which these products and services can flourish.

But there are antitrust risks associated with collaborative standard setting.  Competitors working together in an SSO may have incentives to manipulate the standards-setting process to exclude rivals, fix prices, or allocate markets.  Our Supreme Court decisions over the years have made clear that such manipulation will not be tolerated under our antitrust laws.

Our concerns about the improper assertion of certain patents incorporated into standards arise from potential to harm competition when competitors get together to set standards. Over time, interoperability standards – standards that allow products or systems to work together – have grown more complex and increasingly dependent on using patented technologies.  Competition between technologies, which may include both patented technologies and technologies in the public domain, takes place before a standard is set by an SSO. Once the industry has committed to a standard, it may not be easy to shift to alternative technologies or standards. Adopters of the standard may be locked in, having no choice but to use the chosen technologies.

A holder of patented technologies essential to implementing a standard (“standards-essential patents” or “SEPs”) may be able to take advantage of this lock-in by demanding extra rents.  This is value that is not inherent to the invention itself, but instead is additional value arising from its incorporation into a standard.  These rents might be in the form of excessive royalties or more onerous licensing terms, such as requiring a licensee to give the SEP licensor the right to use the licensee’s proprietary patented technology.  If the terms of such demands were made before the standard was set, standards bodies and potential implementers would be positioned to evaluate the costs and benefits of various technological choices while competition was still vigorous.  No harm, no foul. But the bargaining position of implementers is far weaker after the standard is adopted and the patent holder’s market power, if any, is increased.

Voluntary F/RAND licensing commitments that SSOs seek from patent holders are designed to minimize the risk of hold up. Such commitments limit the delta between the inherent value of the patented technology and any value added simply by its inclusion in the standard.  The F/RAND commitment also serves other purposes.  By assuring that patent holders can expect to be compensated appropriately for the use of their technology, the F/RAND commitment also can encourage participants in the SSO’s standard-setting activities to contribute their best technologies to a standard.  In this way, the F/RAND commitment acts as a bridge that brings patent holders and implementers together and, when it works, provides a path to successful standardization.

The voluntary aspect of a F/RAND commitment is important for two reasons.  First, it precludes those setting the standard from forcing a patent holder to share its patented technology against its will by including it in a standard. Second, it creates transparency in situations where a patent holder decides not to make such a commitment.  When this happens, those working towards standardization are in a position to choose to find another technological solution, abandon the standard, or run the risk of excessive pricing demands post-standardization.

Naturally, then, we are concerned when a patent holder makes and then later seeks to evade the voluntary licensing commitments designed by SSOs to constrain this exercise of market power. Our recent work has focused on harm from a threat to exclude in order to demand excessive royalties under these circumstances. This is hold up, and the competitive process suffers. Alternative technologies may no longer be available. Implementers of the standard do not receive the price benefits of the competition between technologies for inclusion in the standard. Companies that thought they could rely on the F/RAND licensing commitment may be less willing to implement the standard, or future standards. The prospect of such hold up can prevent or delay products coming to market. Alternatively, they can come to market with fewer bells and whistles.

Hold up concerns are real.  In just the last couple of years, we have seen situations where SEP holders have demanded F/RAND royalties that are orders of magnitude larger than court-determined rates. A court recently rejected Motorola’s claim that one of its SEP portfolios was worth as much as $4.50 per unit, concluding that it was worth less than four cents per unit, a ratio of more than 110 to one.  In another case, Innovatio claimed that the appropriate royalty for its SEP portfolio was higher than $16 per tablet. Yet the court found that the RAND rate should be applied based on the component, here the Wi-Fi chip, and awarded a royalty of less than 10 cents per unit, a ratio of more than 160 to one.  In a third case, LSI demanded royalties for its SEP portfolio that exceeded the selling price of the component parts produced using the patents. Instead, the court determined that a RAND rate was 0.19% of the price of the Wi-Fi chips that implemented LSI’s patents, a ratio of more than 500 to one. 

In the examples above, the implementers proved victorious in court.  But not every implementer has the wherewithal to litigate.  Sometimes implementers accede to licensors’ demands, fearing exclusion and costly litigation.  In situations like these, when royalty rates for F/RAND-encumbered patents are driven by successful hold up, consumers can be harmed and innovation incentives are distorted.  A future of exciting new products built atop existing technology may be instead deferred by inefficient investment in downstream products.

The potential for such hold up can be limited if avenues of exclusion are appropriately tailored to the F/RAND commitment.  Patent holders in the United States can seek to exclude infringers from using their inventions through patent litigation in federal courts and administrative actions at the U.S. International Trade Commission (ITC), a federal agency that remedies harm to domestic industries from imports that infringe U.S. patents.  Recently, the ability of F/RAND-encumbered patent holders to get an injunction in U.S. federal courts has been appropriately limited.  This is because under our Supreme Court’s eBay standard, injunctive relief is unlikely to be granted when monetary damages suffice.  Inherent in a F/RAND licensing commitment is a pledge to make licenses available to those who practice the essential patent claims when implementing the standard, not to exclude them from using the standard unless they refuse to take a license on F/RAND terms. 

However, exclusion at the ITC to remedy infringement of F/RAND-encumbered SEPs may be possible under a broader range of circumstances because the eBay decision does not apply to the ITC.  As a result, those seeking to exclude implementers of standards from the U.S. market may be able to use the threat of an ITC exclusion order as a bargaining tool to hold up would-be licensees.

Recognizing the consequences for competition of an SEP holder using the ITC to achieve the functional equivalent of an injunction, DOJ and the U.S. Patent and Trademark Office in January 2013 issued a joint “Policy Statement on Remedies for Standards-Essential Patents Subject to Voluntary F/RAND Commitments.”  The joint statement explained that the use of exclusionary measures at the ITC to remedy infringement of F/RAND-encumbered SEPs is not in the public interest in those cases where the licensor wants monetary compensation for use of its patent needed to implement a standard.  

The joint statement also recognized limited circumstances when an exclusion order may be appropriate.  These exceptions strike an important balance to ensure that standards implementers provide appropriate and timely compensation to the owners of valid, infringed, F/RAND-encumbered SEPs.

This policy was put to good use in August 2013, when the U.S. Trade Representative, acting on behalf of President Obama, vetoed an ITC exclusion order based on infringement of a voluntarily F/RAND-encumbered SEP in a letter endorsing our joint statement. The letter sets forth an analytic framework designed to prevent ITC exclusion proceedings from being merely used as a tactic to facilitate hold up in these types of patent licensing negotiations.  This considered action by USTR to protect competition and consumers from misuse of the ITC process seems to have had its intended effect.  Many fewer SEP holders are currently seeking exclusion orders from the ITC.

Additional clarification to identify when a potential F/RAND licensee constructively refuses to take such a license, engaging in mere gamesmanship, would be useful.  Potential licensees should be able to challenge the validity, enforceability, or infringement of patents included in the offer to license.  A recent decision involving Huawei and ZTE, by the Court of Justice of the European Union, may also be instructive.  That court found that an implementer needed to express a willingness to license on F/RAND terms when properly notified by the patent holder, and diligently respond to a written F/RAND offer in a manner that demonstrates good faith.  Good faith could be shown, for example, by making a specific counteroffer that is also consistent with F/RAND together with financial guarantees of payment consistent with industry custom.

Litigating the meaning of F/RAND at the ITC and in courts is costly and inefficient, regardless of the analytic framework applied.  One way to promote competitive outcomes and mitigate risks to competition arising from collaborative standards-setting activities is for SSOs to “fix it first.”  SSOs are positioned to make standards-setting less susceptible to hold up by clarifying their patent licensing policies.  This was one theme of a thoughtful talk my colleague Renata Hesse gave three years ago.  We stand ready to help SSOs sort through the antitrust risks associated with reducing the likelihood of ex post facto disputes about the meaning of F/RAND commitments.

Although we do not dictate what patent policies SSOs should adopt, it certainly is our view that increasing the clarity in these policies can be good for competition and, therefore, good for consumers.  Similarly, our call for patent holders to honor their F/RAND commitments to SSOs or risk antitrust challenge is well anchored in competition policy.  These commitments are designed to be relied on to constrain the exercise of market power gained when a standard is set and thus avoid harmful patent holdup.

In 2014, the IEEE, an SSO that develops standards in the electronics and communications sectors, requested a DOJ antitrust business review of a proposed update to its patent policy.  The proposed update sought to clarify the meaning of the patent licensing commitment made by patent holders to IEEE.  We analyzed the competitive effects of the proposed policy under the rule of reason and concluded that the policy has the potential to benefit competition and consumers by facilitating licensing negotiations and reducing litigation.  We balanced those competitive upsides against the antitrust risks and concluded that we were unlikely to challenge the proposed update if IEEE were to adopt it. We reached this conclusion for several reasons.  First, licensing rates under the proposed update are ultimately determined through bilateral negotiations on a case-by-case basis.  Second, the specific provisions were consistent with the direction of current case law interpreting the meaning of F/RAND licensing commitments. Third, patent holders could avoid making a licensing commitment under the updated policy and still participate in IEEE standard-setting activities.  Lastly, technology holders were not bound to contribute their patents to future IEEE standards: they were free to take their technology elsewhere and seek to attract implementers willing to adopt their technology.

As our joint statement with the PTO and our business review guidance letter to IEEE demonstrate, we see a role for competition enforcement, guidance and advocacy in the situations described above.  But there are other circumstances where competition enforcement may not be warranted. We have seen recent examples where companies that simply would like to pay a lower royalty or obtain access to important patented technology, look to enforcement by antitrust authorities to achieve this goal for them by whatever means necessary.  As I remarked at the front end of this talk, it is hard to justify antitrust intervention in a basic commercial dispute.  If there is no bad conduct by the patent holder, no improper use of enhanced market power, but rather an assertion of lawful patent rights, competition enforcers need to stand down.  Otherwise we are penalizing lawful innovation.  Companies that know they may easily gain access to the patents or other intellectual property of their competitors have less incentive to undertake the risky and expensive research necessary to be innovators themselves.  Likewise, innovative companies have less incentive to continue their efforts.

So we are skeptical when manufacturers complain to us about high royalty rates in the absence of bad conduct.  We don’t use antitrust enforcement to regulate royalties.  That notion of price controls interferes with free market competition and blunts incentives to innovate.  For this reason, U.S. antitrust law does not bar “excessive pricing” in and of itself.  Rather, lawful monopolists are perfectly free to charge monopoly prices if they choose to do so.  This approach promotes innovation from rivals or new entrants drawn by the lure of large rewards.  In this regard, we make common cause with our European enforcement colleagues.  Even though Article 102 of the Treaty on the Functioning of the European Union (TFEU) authorizes actions against excessive pricing, the European Commission has said that “addressing excessive prices is an area of antitrust where limited and very cautious intervention is warranted.” 

In addition, although getting access to certain patents, especially differentiating patents owned by one’s competitor, may be commercially desirable for companies who wish to use the technology in their own products, patent holders may not want to license them. In our view, there are extremely limited situations, if any, in which a patent may be considered “necessary” or “essential” to compete in a market and, on that basis alone, make a refusal to license a patent an antitrust violation.  Forced sharing of patents that does not remedy some cognizable harm to competition is a misappropriation of assets that creates disincentives to innovation and investment.

The challenge for competition authorities is to calibrate correctly our enforcement work and our competition advocacy to ensure that patent rights serve to promote innovation and consumer welfare.  When patents are incorporated in standards, we accomplish that goal by focusing on the meaning and voluntary nature of the F/RAND commitment.  As we look forward, the Antitrust Division will continue to promote sound competition policy by antitrust enforcers, administrative agencies, and courts that will allow us to reap the benefits of innovation that arise from patents, standards, and competition.

Reflections on the Role of Competition Agencies When Patents Become Essential (90.39 KB)

Updated February 4, 2016