How did the IEEE Assume its Global Governance Responsibilities?

Roya Ghafele is the Director of OxFirst, a spin-out from Oxford University that focuses on the interplay of law and economics. She has also held posts as an Assistant Professor and Fellow in International Political Economy with Oxford University since 2008.

A Mismatch in Paradigms

In March 2015, IEEE, a major standard-setting organization known for issuing prominent standards such as WiFi, amended its patent policy. (Research funding for this essay was received from the IEEE.) In doing so, IEEE sought to provide further transparency and clarity to what the (F)RAND (“fair, reasonable and non-discriminatory”) promise entails. This step led to heated debates and left industry divided on the governance function that standard-setting organizations should assume. This essay discusses the central tenets of the updated IEEE patent policy in light of standard-setting organizations’ prerogatives to articulate their global governance responsibilities.

Standard-setting organizations (SSOs) have the prerogative to articulate their governance responsibilities and advance patent policies that preserve the public interest. By not doing so, they fail to mediate between the exclusionary features of patent law and the open, collaborative nature of standards.

A failure by SSOs to assume their governance responsibilities can lead to a series of shortcomings, including potentially costly litigation among market participants and, ultimately, interference with the adoption of a standard altogether. SSOs are, as the name indicates, primarily interested in establishing and promoting the widespread use of standards. They are not patent offices and not prima facie set up to manage patent policies. Hence, their approach to patent law will be driven from a standards-setting perspective. Any effort by an SSO to come to grips with patent law needs to be understood in light of its primary role.

Against this background, I attempt to establish prototypical governance functions a standard-setting organization needs to assume. While standardization easily meets the characteristics of networked market contexts, the patent system probably has to catch up. Dense webs of overlapping patent rights, through which small and medium-sized enterprises (SMEs) must tediously navigate their way, likely interfere with collaborative exchange and open innovation.

The Inherent Tensions Between Patents and Standards: Patents Confer the Right to Exclude, Which Is in Strong Contrast to the Open Nature of Standards

An important and often-overlooked aspect of the patent system is rooted in transaction cost economics, a school of thought which recognizes that contract establishment, management and enforcement operate based on costs. In light of these costs, the patent system can only be effective if contracts, such as licenses, can be obtained in a swift and uncomplicated manner. If, to the contrary, costs associated with patent contracts – be they licenses or other legal means – are exceedingly high or it is exceptionally complicated and lengthy to obtain contractual arrangements such as licenses, then licensors and would-be licensees may be better off refraining from licensing altogether. Such a decision, in turn, leads to inefficiencies and ultimately market failures associated with the patent system. Hence, the effectiveness of the patent system is strongly intertwined with the ability to manage costs. Providing further clarity of what the (F)RAND promise entails is thus likely to promote efficiency because both parties to a contract, be they the licensor or licensee, are able to save significantly on costs and time if certain key aspects of licensing standard essential patents (SEPs) are defined ex ante. This enables licensors and licensees to engage in more efficient economic transactions, which in turn is likely to benefit the reputation and efficiency of the patent system.

While the patent system is built around the logic of excluding third parties from an invention, the rationale of a standard derives from offering market participants operating within a specific innovation ecosystem a baseline on which to build and compete, as discussed by J.S. Miller’s 2006 article on Standard Setting and Patents.

The act of setting a standard remains essentially a conscious intervention in free and open technology markets. As a host of different technology solutions compete for adoption, the process of setting a standard is a deliberate choice taken by market participants to opt for one specific technological configuration over potential alternatives.

While the wide dissemination of standards is crucial for the adoption of interoperable and interconnected systems, it also means that alternative technologies may be abandoned and face great difficulties in their adoption as consumers are already locked into the established standard. If the established standard is of inferior quality, then innovation and technological progress can actually be hampered rather than promoted through the act of standard-setting, as consumers are locked in and face tremendous switching costs were they to upgrade to a more advanced technological solution.

Both standards and patents promote innovation. Yet both also interfere with free markets, though by different means. The success of a standard is based on its wide dissemination. Its value derives from its vast usage. This stands in sharp contrast to patents, which are negative rights built around exclusivity. Contrary to a standard, the value of a patent derives from its strength to exclude third parties from using the invention unless, obviously, they are willing to pay to do so.  The foregoing arguments – the tension between “free access and tight control” as formulated by J.S. Miller illustrate why the inherent dilemma between patents and standards is difficult to overcome. This tension is well-pronounced in this debate over SEPs. A patent declared essential to a standard is a strange hybrid that combines patent law’s negative rights aspect with a standard’s capability to disseminate a technology as widely as possible. This formula bears the potential to accrue exceptional market power in the hands of patent owners, while at the same time rendering access to proprietary technology potentially expensive. The (F)RAND promise was introduced to counterbalance this tendency.

The Ambiguity of the (F)RAND Promise

Patent holders are expected to license their SEPs on (F)RAND terms under the intellectual property rights (IPR) policy set by the relevant Standard Setting Organizations (SSOs). As such, the concept of (F)RAND has been proposed as a flexible framework for negotiations allowing SEPs owners, as well as SEP implementers, to freely license SEPs.

The (F)RAND commitment seeks to waive the patent holder’s right to refuse to license its IPR to anybody seeking such a license. For their part, licensees are expected to engage in licensing negotiations in good faith and not infringe on an SEP holder’s IPR, as discussed at length in Mario Mariniello’s 2011 article on FRAND.

Under the patent policies of most SSOs, (F)RAND commitments translate into an insufficiently complete contract between licensors and licensees. This is because of a built-in fundamental ambiguity over what “reasonable” and what “non-discriminatory” mean; an ambiguity that is not addressed by means of the policies themselves but is expected to be resolved by courts, regulators and other third parties.

The fundamental ambiguity of the (F)RAND commitment translates into significant transaction costs, which has led David J. Teece of the University of California (Berkeley), to argue that the costs of the transfer of technology between competing enterprises can be so high as to mitigate the value of the transfer altogether.

Lack of clarity can also lead to a host of other unresolved challenges, such as negotiations potentially taking place in the shadow of the law and potential asymmetrical bargaining power in licensing negotiations.

Why Standard-Setting Organizations Need to Assume Their Responsibilities

The importance of markets for novel technology to stimulating growth and the need to assure consumer welfare make it necessary to establish a well-reflected governance structure for SEPs.

Central to the determining how to best address a desire for openness, greater clarity and reflect novel market realities, is the question: How does an SSO decide to approach the (F)RAND promise? The (F)RAND promise is intended to offer a middle ground that eases the inherent tension between standards and patents. Three core policy responsibilities emerge from this purpose:

  1. The need to address the inherent tensions between patents and standards and hence promote the openness of standards
  2. The need to overcome the inherent ambiguity of the (F)RAND promise by providing greater clarity and hence promote transparency
  3. The need to reflect the market context in policy formulation

The self-organizational principles that underlie the governance mechanisms of SSOs reflect an eagerness to promote open standards; hence, standards that can be used in a wide range of industries without market participants being distracted by lack of certainty.

How IEEE Assumed its Governance Responsibilities

The very process of taking ownership and providing further clarity of the meaning of (F)RAND is likely to benefit all participants in the market, whether they seek to license SEPs in our out. The significant cost and time savings associated with operating in a well-defined legal framework are argument enough to offset potentially undesirable side effects, cited by some market participants.

The IEEE patent policy works towards those goals. IEEE has had a patent policy in place for more than two decades. Following a series of concerns raised within the UN system, the IEEE launched a comprehensive review process of its policy and in doing so, assumed its global governance responsibilities.

What the Updated IEEE Patent Policy Entails

The updated IEEE patent policy is built upon four pillars:

  • A definition of a “reasonable” rate

  • A precise definition of the term “compliant implementation”

  • Further clarity regarding use of ”prohibitive orders”

  • Reciprocal licensing

A definition of a “reasonable” rate 

Under the updated patent policy, a “reasonable” rate is defined as “appropriate compensation to the patent holder for the practice of an Essential Patent Claim excluding the value, if any, resulting from the inclusion of that Essential Patent Claim’s technology in the IEEE standard.”

A precise definition of the term “compliant implementation”

The notion of “compliant implementation” seeks to ensure that makers of components that implement an IEEE standard can benefit from “letters of assurance,” but only for use of the component conforming to the standard. Patent holders cannot refuse to honor their commitments to IEEE to grant licenses to an unrestricted number of applicants and product makers cannot force a patent owner to grant a license beyond the scope of that commitment.

Further clarity regarding use of “prohibitive orders”

The update clarifies further the relationship between standards and patents by providing insights into when prohibitive orders (also known as injunctions) are an appropriate means of recourse. The update states that parties should negotiate over license terms and makes it clear that a patent holder is not prohibited from seeking an injunction if an implementer declines to participate in or comply with the outcome of an adjudication, as specified in the policy.

Reciprocal licensing

The IEEE patent policy update states that a patent holder cannot require an implementer to take a license to a non-essential patent or require an implementer to grant to the patent holder a license to the implementer’s own patents (except essential patents on the same IEEE standard). Patent holders and implementers are free, however, to negotiate any kind of cross-license or portfolio licenses that they mutually wish.

Prior to IEEE’s adoption of its updated patent policy, the U.S. Department of Justice prepared a business review letter in which it stated that the IEEE patent policy bears the potential to benefit competition and consumers and that it had no intention at this point to take antitrust enforcement action.

IEEE further recognized that the degree of openness of a standard depends to an important degree on how that standard responds to the patent rights retained by a SEPs owner. If usage is fraught with transaction costs and access to the standard is not based on a “reasonable” licensing rate, then the standard will be closed rather than opened to the wide community, as discussed by Mark Lemley in a 2002 article. This peculiar aspect was addressed by IEEE.

Overcoming the Ambiguity of the (F)RAND Promise

In the absence of clear definitions, determining what “negotiating in good faith” means can be a lengthy and hence costly undertaking. By making the definition of “reasonable” and “non-discriminatory” under the (F)RAND promise core to its updated patent policy, IEEE assumes its governance responsibilities at the global level.

The mere act of shedding light on what a “reasonable” rate means, what “non-discrimination” entails, when prohibitive orders should be available and what reciprocal licensing means, translates into a decrease in transaction costs for SEPs owners and implementers alike. If core aspects of the SEPs licensing regime were to remain unclear, ill-defined or ultimately opaque, it would likely result in lengthy disputes and potentially costly litigation might well ensue. Lemley thus argues that SSOs bear the potential to become “transaction cost minimizing governance structures.”

This statement is true simply because a contract, where the terms are too vaguely defined, is hard to enforce. Simply put, without sufficiently clearly defined patent policies of SSOs, the bargaining in the beginning of a licensing negotiation is affected by potential court and other legal involvement later should the SEPs holder and implementer not come to an agreement. This bears an inherent risk for both implementers and owners of SEPs. If courts need to increase their involvement and hence fill the void of ex ante set governance frameworks, then the distribution of the economic stakes are altered.

In other words, if all market participants operating in a technology field that relies strongly on standards fail to determine the “rules of the game” in advance, then there is an inherent risk of opportunistic behavior. Downstream innovators and implementers can find themselves using standards protected by patents, only to find themselves later on facing exceedingly high royalty rates. SEPs owners may also see their efforts in investing in the standard unrewarded. Significant transaction costs that occur in the absence of adequate governance structures harm both and in that sense erode the value proposition of patent law as the “right to exclude” may become so costly as to be ultimately unenforceable in practice.

Ultimately, it already takes the court system to determine which patents are essential to a technology standard and if patents are at all valid. Deferring to the courts even more aspects of the governance system of SEPs leaves a democratic deficit, influences the respective bargaining positions in SEPs negotiations and leaves both parties no choice, other than to negotiate licensing terms in the shadow of an ill-defined patent policy, a situation posing serious business risks.

How IEEE’s Patent Policy Reflects Novel Market Contexts

The much deeper question as to how patent law fits the characteristics of clusters and networks is difficult for even IEEE to address. However, IEEE’s updated patent policy is starting to seek to encapsulate some of the most pressing issues in the field. By adopting its novel patent policy, IEEE sought to offer a governance structure functioning for global corporations, SMEs, universities and public research institutions alike. It furthermore provides information useful to arbitration bodies, courts and regulatory agencies alike.

In doing so, the IEEE patent policy recognizes that the conditions under which standards will find their application in technical devices depends to an equal extent on how contributors to the standard will be remunerated for their efforts and how standards will find adoption by implementers.

What Rules Apply to Information Privacy and the Internet of Things?

How Can SHODAN Help Prevent Cyberattacks?