The Antitrust Division of the U.S. Department of Justice (“DOJ”) has published a business review letter that it will not challenge the Institute of Electrical and Electronics Engineers (“IEEE”) adopting changes to its IPR Policy discussed in our Feb. 3, 2015 post. In a related DOJ press release, DOJ states that it does not dictate patent policy choices of standard setting organizations (“SSOs”), and did not believe the proposed IPR policy change “is likely to result in harm to competition”, stating:
“The department supports standards setting organizations’ efforts to clarify their patent licensing policies,” said Acting Assistant Attorney General Hesse. ”IEEE’s decision to update its policy, if adopted by the IEEE Board, has the potential to help patent holders and standards implementers to reach mutually beneficial licensing agreements and to facilitate the adoption of pro-competitive standards. Where, as here, the department does not believe that adoption of a policy change is likely to result in harm to competition, IEEE and other standards setting organizations are free to adopt those modifications to their policies that they believe will benefit their standard setting activities. The U.S. government does not dictate patent policy choices to private standards setting organizations.”
DOJ also explained the limited import of its business review letter as a current intent not to bring an antitrust challenge against the business activity, though it may do so later if there are anticompetitve effects:
Under the department’s business review procedure, an organization may submit a proposed action to the Antitrust Division and receive a statement as to whether the division currently intends to challenge the action under the antitrust laws based on the information provided. The department reserves the right to challenge the proposed action under the antitrust laws if it produces anticompetitive effects.
Below is a summary review of DOJ’s action here.
IEEE Request for Business Review Letter
DOJ was responding to IEEE’s request for a business review letter. IEEE stated it was requesting such a review because some comments in response to the proposed IPR Policy change “voiced either vague or specific antitrust concerns,” including concerns that revisions “to the term ‘reasonable rate’ … could amount to ‘buyer-side price-fixing.’” The IPR Policy change generally is directed to four areas:
- A definition of “Reasonable Rate” that includes reference to “the smallest saleable Compliant Implementation”
- A definition of “Compliant Implementation” as being a component, sub-assembly or end-product so that any such implementers along the chain may seek the benefits of a RAND-commitment
- A blanket preclusion on seeking injunctive relief prior to adjucating through appeal reasonable license terms even as to unwilling licensees
- Reciprocity provisions that may condition providing a license in return for a cross-licensee to the licensee’s SEPs on that same standard, and permitting voluntary flexibility with other reciprocity terms.
The IEEE’s letter states that it does not believe there is “buyer-side price-fixing” for several reasons. With respect to what IEEE considered the most specific anti-trust issue raised concerning the term “reasonable rate”, IEEE emphasized that the enumerated factors–e.g., the smallest saleable Compliant Implementation–are not required to be considered, only recommended:
1. The proposed policy does not set a maximum royalty, either for a specific patent or for a group of all patents essential to a particular standard. It generally defines the term “reasonable rate” and recommends (but does not require) additional factors for consideration in determining an appropriate rate. The proposed policy does not prevent parties from discussing any other factors that they believe appropriate.
IEEE also stated that the proposed IPR Policy change “does not retroactively amend previously Accepted Letters of Assurance” and patent owners are not required to submit new Letters of Assurance under the proposed IPR Policy change if they don’t want to.
IEEE’s Process In Proposing IPR Policy Change
DOJ received concerns by some entities that “parties desiring lower royalty rates commandeered IEEE-SA” and the IPR Policy revision “was the product of a closed and biased process.” DOJ does not go into details about those concerns, but states that most concerned an Ad Hoc committee within IEEE that drafted the revision. Below is a summary of where that Ad Hoc committee fit into the process. Ultimately, DOJ determined that, “[g]iven the numerous opportunities for comment, discussion, and voting at different levels within IEEE, the Department cannot conclude that the process raises antitrust concerns.”
IEEE is a large non-profit professional technology association. The IEEE Standard Association (IEEE-SA) is an operating unit within IEEE that develops technical industry standards and which developed the IPR Policy at issue here. The IEEE-SA is governed by its own Board of Governors, who appoint members to the IEEE-SA Standards Board that oversees the IEEE standards-development process. The IEEE-SA Standards Board has various working committees, including the Patent Committee (“PatCom”) that oversees the use of patents in developing IEEE standards.
Participants in the IEEE standard-setting process may submit letters of assurance (“LOAs”) for patents that may be essential to a developing standard and indicate their willingness to license such patents, and under what terms, if they ultimately are essential to the finalized standard. Those terms include, for example, an agreement to license a patent that is essential to the standard on reasonable and non-discriminatory terms (“RAND”).
The PatCom chair appointed an Ad Hoc committee to consider changes to the IPR Policy given divergent views on what constitutes “reasonable rates” under a RAND-encumbered patent. The Ad Hoc committee then drafted changes to the IPR Policy, which was revised and ultimately approved by the PatCom to forward to the IEEE-SA Standard Board. The IEEE-SA Standards Board approved forwarding the revised IPR Policy to the IEEE-SA Board of Governors, which then approved the revision conditioned upon receiving a favorable business review letter from DOJ (i.e., this one here) and ultimate review by the IEEE Board of Directors. The IEEE Board is expected to vote on this revised IPR Policy soon.
“Prohibitive Order” (e.g., Injunctions)
DOJ considered the proposed IPR Policy change that would preclude seeking an injunction until after licensing terms are adjudicated and go through at least a first level of appeal. DOJ observes that a RAND commitment under IEEE’s current IPR Policy does not preclude seeking an injunction against an unwilling licensee, stating that the RAND commitment includes an “[i]nherent … pledge to make licenses available to those who practice such essential patent claims … in other words, not to exclude these implementers from using the standard unless they refuse to take a RAND license.” DOJ also observes, however, that the proposed amended IPR Policy “place[s] additional limits on patent holders’ ability to obtain injunctive relief” and is more restrictive than current U.S. case law and guidance from U.S. agencies, stating:
This provision may place additional limits on patent holders’ ability to obtain injunctive relief in a U.S. court, but it appears that, in practice, it will not be significantly more restrictive than current U.S. case law, and the added clarity may help parties reach agreement more quickly. Although this provision is more restrictive than recent guidance on this issue from the U.S. government, the U.S. government does not dictate patent policy choices to private SSOs.
That noticeably understates the point because, as explained in our Feb. 3, 2015 post, there is universal agreement in statements by courts and agencies–including DOJ itself–that injunctive relief should be available in certain circumstances involving unwilling licensees in order to deter patent hold-out. DOJ does not directly address this issue that the IPR Policy would preclude seeking an injunction against an unwilling licensee. DOJ does so indirectly, indicating that implementers may be deterred from seeking to “delay payment” or being “recalcitrant about taking a license” because (i) they can avoid uncertainty caused by the absence of a license, (ii) they can avoid the cost of litigation, (iii) they may obtain a pre-litigation discount, and (iv) courts may require them to pay a bond or escrow payments, stating:
Nevertheless, patent holders have expressed concern that this damages remedy is insufficient because it permits potential licensees to benefit by delaying paying reasonable compensation for a portfolio of patents until a patent holder has litigated each patent in its portfolio individually. The Department encourages patent holders and implementers to negotiate licensing agreements that are mutually acceptable, and there are incentives favoring a negotiated outcome. For example, implementers have incentives to reach agreement on licensing terms to reduce cost uncertainty as they bring products to market. In addition, litigation is expensive for both parties and licensees risk that a court will award a higher royalty for a patent that is found to be valid and infringed than a discounted pre-litigation rate offered by a licensor.
… [W]here potential licensees appear recalcitrant about taking a license, courts and other third-party decision makers may seek to ensure payment by requiring alleged infringers to post a bond or make escrow payments. Moreover, other potential licensees will be less likely to litigate once a patent holder has demonstrated the value of its patents (or a subset of the patents in its portfolio) through successful infringement litigation.
But, again, those considerations always have been present, yet were not deemed sufficient deterrents by courts and U.S. agencies that have consistently stated that injunctive relief should be available against unwilling licensees. This revision, therefore, actually may increase litigation over IEEE standards even though reducing such litigation is the suggested reason behind these revisions.
DOJ ultimately concludes that the proposed IPR Policy revision “is unlikely to result in competitive harm,” because it “is consistent with the direction of U.S. case law and patent holders can avoid its requirements by declining to submit an LOA.” As discussed above, however, precluding injunctive relief against unwilling licensees is against current U.S. case law and prior guidance from U.S. agencies. Challengers of the proposed IPR Policy change may argue this raises questions on the reliability of DOJ’s business review letter.
Mandatory Factor. DOJ reviewed the defined “Reasonable Rate” as having a “mandatory factor” that “a Reasonable Rate ‘shall mean appropriate compensation … excluding the value, if any, resulting from the inclusion of [the patent claim's] technology in the IEEE standard.’” The general concept of apportioning value to the patented technology finds support in current case law, such as the Federal Circuit’s recent Ericsson decision. DOJ also found that this provision aligns with goals of providing patent holders appropriate compensation without improper “hold-up value” connected with standardization, citing the Federal Circuit’s recent Ericsson v. D-Link decision that “the patentee’s royalty must be premised on the value of the patented feature, not any value added by the standard’s adoption of the patented technology.” That Federal Circuit decision, however, counseled as being “neither necessary nor appropriate” instructing a jury on patent hold-up based merely on some “general argument that these phenomena are possibilities;” rather, patent hold-up may be considered if there is evidence of actual holdup, such as a patent holder seeking higher royalty rates after the IEEE standard was adopted (see our Dec. 5, 2014 post for a summary of the Ericsson v. D-Link decision).
DOJ does not refer to any actual evidence of hold-up for patents involving IEEE standards, it is not clear that IEEE itself had considered such actual evidence, and IEEE standards (e.g., the 802.11 WiFi standard) have comprised the bulk of standard essential patent litigation, but they have no provided any actual evidence of patent holdup. We have reported on one instance where a patent owner–that had not participated in the standard-setting process or made any associated commitment–argued its royalty rate could be enhanced by the fact that the defendant could not practice the standard absent a license to the patent, but the court in that case unsurprisingly precluded the patent owner from seeking such “hold-up” value. (See our Apr. 23, 2014 post). Challengers of the revised IPR Policy may argue that the lack of any evidence of actual hold-up for patents involving IEEE standards raises questions about reliance on DOJ’s position here that seeks to justify a significant IPR policy change in order to address a problem for which there is no evidence actually exists.
Three Optional Factors. DOJ also considered the remaining “three recommended factors” that include considering the “smallest saleable Compliant Implementation” and limiting consideration of comparable licenses to those not entered under threat of an injunction or in circumstances not consistent with the defined “Reasonable Rate.” The DOJ found comfort from these factors not being required and no specific methodology being mandated:
Significantly, the Update makes clear that the determination of Reasonable Rates ‘need not be limited to’ these factors. The Update, then, does not mandate any specific royalty calculation methodology or specific royalty rates.
Not mandating a specific methodology is consistent with developed case law, the Federal Circuit’sEricsson v. D-Link decision rejecting a specific methodology for all RAND-encumbered patents, but relying instead on the particular facts presented in any given case:
To be clear, we do not hold that there is a modified version of the Georgia-Pacific factors that should be used for all RAND-encumbered patents. Indeed, to the extent D-Link argues that the trial court was required to give instructions that mirrored the analysis in Innovatio or Microsoft, we specifically reject that argument. We believe it is unwise to create a new set of Georgia-Pacific-like factors for all cases involving RAND-encumbered patents. Although we recognize the desire for bright line rules and the need for district courts to start somewhere, courts must consider the facts of record when instructing the jury and should avoid rote reference to any particular damages formula.
Although the factors are not exclusive or required to be considered, DOJ found that they “focus attention on considerations that may be likely to lead to the appropriate valuation.” This “focus”, however, might lead to the precise “rote reference to any particular damages formula” that the Federal Circuit counseled to avoid.
DOJ found that the “smallest saleable Compliant Implementation” factor “may be appropriate … particularly when the product is complex and incorporates many patented technologies.” DOJ supports this by citing to three Federal Circuit decisions on the smallest saleable patent practicing unit (see our blog posts on LaserDynamics, Virnetx and Ericsson). But those cases make clear that the “smallest saleable patent practicing unit” is solely a creature of U.S. jury litigation under evidentiary Rule 403 that seeks to limit confusion of lay jurors considering a hypothetical negotiation and, thus, may not be applicable to real world, bilateral negotiations between sophisticated market participants (see our Feb. 3, 2015 post). DOJ does not mention this limited applicability of the smallest saleable patent practicing unit, which challenges of the IPR Policy may argue raises questions about reliance on DOJ’s analysis. But DOJ may have taken some comfort in that this factor is not mandatory nor does it preclude considering the end product, stating:
This factor does not mandate the use of the smallest saleable Compliant Implementation as the correct [royalty] base. For example, the provision does not exclude evidence of the role of the relevant patented functionality in driving demand for the end product, or bar using end-product licenses to help determine the appropriate value to attribute to the technology.
DOJ also considered the second non-mandatory factor regarding apportioning value of all SEPs on a standard “addresses royalty stacking, which may hamper implementation of a standard,” relying in part on the Federal Circuit’s Ericssson decision that discussed royalty stacking concerns. As with patent hold-up, the Federal Circuit’s recent Ericsson decision eschewed reliance on theoretical royalty stacking concerns in favor of actual evidence of such royalty stacking, stating that “[t]he mere fact that thousands of patents are declared to be essential to a standard does not mean that a standard-compliant company will necessarily have to pay a royalty to each SEP holder.” As with patent holdup, DOJ does not state that it considered any actual evidence of royalty stacking or that IEEE had either, and the numerous litigations involving IEEE standards have not provided such evidence. So challengers of the revised IPR Policy may argues this raises questions about relying on DOJ’s position here that is premised on a problem for which there is no evidence exists.
Regarding the third non-mandatory factor on what constitutes comparable licenses, DOJ found comfort that “[t]he Update does not prevent consideration of licensing agreements other than those specifically identified therein.” The ability to consider a broader swath of comparable licenses is consistent with U.S. case law, which permits differences from the instant license negotiation to be factors a jury can consider without necessarily excluding those comparable licenses altogether (see our Sep. 17, 2014 post on Virnetx and Dec. 5, 2014 post on Ericsson).
Any “Compliant Implementation”
DOJ considered a revision based on the definition of “Compliant Implementation” as “meaning that a patent holder making an IEEE RAND Commitment cannot refuse to license its patents for use in IEEE-SA standards at certain level of production.” DOJ observes that this may “entail a departure from historical licensing practices”, but “the Update does not mandate specific licensing terms at different levels of production,” stating:
For example, the royalty rate need not necessarily be the same at all levels of production. In each case, the RAND royalty should reflect the value of the patented technology. If a patented invention’s value is not reflected in the current price of upstream implementations, due to historical licensing practices, some adjustments may be necessary, and the Update does not prevent such adjustments.
DOJ’s point here is not entirely clear. Does DOJ mean that separate licensing agreements can be entered and accumulate at each of level of the vertical supply chain so that, by the end of the supply chain, the cumulative total royalty provides full compensation for the patented technology (and can that be done given patent exhaustion concerns — see, e.g., our June 25, 2014 post considering Federal Circuit staying action against retailers while case proceeded against manufacturer). Or does DOJ suggest, per its last sentence above, that the price of a component used to implement a standard should be increased to account for the benefits that an invention provides to each particular end product in which the component is used.
For example, how would DOJ’s proposal work in the hypothetical presented in our Feb. 3, 2015 postfor patented technology that provides power savings in a wireless standard that is implemented in a microchip used (1) in a stationary end product that derives no benefit from the patented power-savings because it is powered from a wall outlet and (2) a mobile end product that substantially benefits from the patented power-saving:
- Does the microchip vendor raise the price of the microchip across the board to pay the entire patent licensing fee so that manufacturers of the stationary end product (that do not need the patented technology) pay the same licensing fee as manufacturers of the mobile end product (that greatly benefit from the patented technology)?
- Or does the microchip vendor sell the same microchip at different prices based on what end product within which the microchip will be implemented to account for the different value patented technology provides to different end products?
- Or does the microchip vendor sell a different microchip to different manufacturers to exclude technology not needed by the end product?
None of these may be practical options. This may be particularly true with the latter option given (1) costs in having many different microchip versions, (2) some nominal need for the patented technology in an end product (i.e., its needed, but is not as valuable as it is in other products) and (3) end products may be required to have the patented functionality–even if no benefit to that product–in order to state that they are fully compliant with the wireless standard (important where the fundamental premise of having standards is interoperability among disparate products that comply with the standard). These various considerations may indeed be why, as DOJ and courts have observed, actual license negotiations often are based on licensing end products, rather than components thereof.
DOJ considered a provision of the proposed IPR Policy amendment that “permits a licensor to require a potential licensee to grant back a license to its own patents essential to the same standard.” But the proposed amendment would “prohibit licensors from demanding licenses to applicants’ patents that are not essential to the same standard … and from forcing an applicant to take a license to patent claims that are not essential to the referenced standard.” This is intended to prohibit “forc[ing]” a cross-license to a licensee’s differentiating patents, which might “decrease incentives to innovate.” DOJ found that this provision still “leaves parties free to negotiate these types of terms voluntarily.”
DOJ ultimately concludes that the proposed IPR Policy amendment has “potential to benefit”, and cannot conclude it is “likely to harm competition”, stating:
The Department concludes that the Update has the potential to benefit competition and consumers by facilitating licensing negotiations, mitigating hold up and royalty stacking, and promoting competition among technologies for inclusion in standards. The Department cannot conclude that the Update is likely to harm competition. Further, to the extent there are any potential competitive harms, the Department concludes that the Update’s potential procompetitive benefits likely outweigh those harms. Accordingly, the Department has no present intention to take antitrust enforcement action against the conduct you have described.
DOJ was careful to note that its conclusion did not express views as to applying the amended IPR Policy retroactively to existing LOAs, where IEEE indicated the proposed IPR Policy would not be retroactive:
The Department’s analysis in this letter applies only to the Update;’s impact on future LOAs; the Department offers no statement regarding its intentions concerning the application of the Update retroactively to previously submitted LOAs.
The import of DOJ’s action here and the proposed IEEE IPR Policy amendment is yet to be seen. If the IEEE Board of Directors does adopt the amended IPR Policy, DOJ’s letter here may not prevent others from challenging the IPR Policy change on antitrust or other grounds and DOJ itself indicated it may take action in the future if the changed policy “produces anticompetitive effects.” As discussed above, those challenging the IPR Policy change may raise questions about the DOJ’s positions here given conflicts with existing law, lack of evidence that problems sought to be solved even exist and other issues. This may give the IEEE Board of Directors pause about whether to adopt or amend the proposed IPR Policy amendment when they consider the issue.
Further, the impact of the IPR Policy if adopted is not clear. It plainly has anti-patent overtones, which is disappointing given how much we all have benefited from the innovations fueled by patents every step of the way over many decades in telecommunications and other standards-driven technology. An important part of DOJs conclusion was the choice patent owners have whether to submit a LOA having the new IPR Policy provisions and flexibility to not be bound by various factors when negotiating a RAND-encumbered patent under that new IPR Policy. But, in practice, will refusing to submit such an LOA be a viable competitive option for a patent holder? For example, in the CSIRO case, Judge Davis observed that the patent holder who at first gave an LOA later expressly refused to provide an LOA for later versions of the standard, yet IEEE did not change its standard to avoid that patent. (see our July 28, 2014 post for summary of the CSIRO decision, currently on appeal to the Federal Circuit). And if a large number of innovating patent holders refuse to submit the new LOA, will IEEE find that the cost of constant consideration of whether and how to revise standards to design around those patents as well as foregoing innovative technology warrants more flexibility in the LOA? But the uncertainty of such an experiment might be better tested in a brand new standard setting organization, rather than an established ubiquitous one like IEEE .
We also will see if other standard setting organizations consider a similar IPR Policy change or await the aftermath of this one if adopted. DOJ itself noted that its role was not to assess whether the IPR Policy is right for IEEE and that having different IPR Policies among different standard setting organizations can be beneficial:
The Department’s task in the business review process is to advise the requesting party of the Department’s present antitrust enforcement intentions regarding the proposed conduct. It is not the Department’s role to assess whether IEEE’s policy choices are right for IEEE as a standards-setting organization (“SSO”). SSOs develop and adjust patent policies to best meet their particular needs. It is unlikely that there is a one-size-fits-all approach for all SSOs, and, indeed, variation among SSOs’ patent policies could be beneficial to the overall standards-setting process. Other SSOs, therefore, may decide to implement patent policies that differ from the Update.