Judge James V. Selna of the Central District of California (“C.D. Cal.”) recently released the redacted, 115-page public version of his Memo of Facts and Law with his FRAND determination in the TCL v. Ericsson SEP dispute concerning 2G, 3G and 4G cellular technology in the European Telecommunications Standards Institute (“ETSI”) standards along with his Final Judgment And Injunction, which injunction has detailed terms like one would find in a licensing agreement.

Judge Selna ultimately ruled that Ericsson’s licensing conduct did not breach its FRAND commitment, but that Ericsson’s proposed licensing terms were not FRAND.  Judge Selna rejected the FRAND methodologies and resulting FRAND royalty rates proposed by both TCL and Ericsson.  Judge Selna did his own FRAND methodology based on the methods and evidence presented by the parties, following mainly a modified version of a “top down” approach proposed by TCL.  The FRAND rates determined by Judge Selna fell about half-way between TCL and Ericsson’s proposals, though direct comparison is difficult.  For example, for Ericsson’s 4G SEPs, the royalty rates from the parties and court varied as to scope (e.g., blended global rate versus regional rate) and required some conversion to compare (e.g., Judge Selna computed an effective “unpacked” royalty that accounted for lump-sum payments and royalty floors in Ericsson’s offers):

4G SEP Royalty Rate
(Percentage of Mobile Phone’s Net Price)

TCL’s Proposed 4G Global Rate 0.16% (Blended global rate)
Court’s 4G Rates (by region) 0.450% (U.S.)
0.314% (Rest of World; No 4G Sales in Europe)
Ericsson Effective U.S. 4G Rates
(Court calculated from Option A and B offers)
1.074% (Option A Effective U.S. Rate) or
1.988% (Option B Effective U.S. Rate)

We provide below a bullet-list summary of some key points from the decision as well as a (rather lengthy) detailed discussion of Judge Selna’s decision.  We consider this an important decision to read, and encourage you to do so, because it is one of the few decisions that describe a court’s analysis in determining a disputed FRAND royalty.  But we also believe this case provides only incremental development of the case law itself given the highly factual nature of the decision in this still developing area of law.  Judge Selna  acknowledged that trying to obtain “precision and absolute certainty” here was a “doomed undertaking.”  In other words: Learn from this decision, but do not assume it represents a definitive proper FRAND analysis and is representative of a FRAND royalty for all FRAND cases. Its one step in a continuing journey …

This is the fourth U.S. bench trial we are aware of that determined a FRAND royalty and, like all those bench trials, the court here rejected the FRAND methodology calculation proposed by all the parties and developed its own method and results. (See our May 1, 2013 post on Judge Robart’s RAND decision; our Oct. 3, 2013 post on Judge Holderman’s RAND decision; and our July 28, 2014 post on Judge Gilstrap’s RAND decision).   This is not surprising and is due in part to the developing area of law and the practical limitations of litigation.

Basically, FRAND cases–like other cases–are decided given the specific evidence and arguments presented by the parties.  Reasonable royalty law itself has some uncertainty that the courts are continuing to develope; FRAND royalty law is premised on that reasonable royalty law and is in a nascent state.  As litigants proceed in this developing area, they do not have as much guidance on the specific evidence and arguments they should make as compared to what’s required to establish infringement, validity or other well-settled areas of patent law.  And the amount of data and complexity’s they can pursue are limitless–especially for large patent portfolios–but the time and resources for litigation are limited.  Litigants can learn incrementally from this decision to develop better methodologies for litigating the next FRAND royalty case; but it might be a mistake to assume that this case (or any other single FRAND decision at this point) presents a fully proper FRAND analysis or a representative FRAND royalty for all cases.  Indeed, the parties themselves probably could re-litigate this same case and achieve a different result based on what they learned during the case and from the ultimate decision.

Further, Judge Selna indicated that his determination was made in the context of litigation and that real-world FRAND license negotiations may have different considerations, approaches and results.  In other words, it may be a mistake to take some portions of Judge Selna’s analysis as a requirement in real-world FRAND license negotiations.  Again:  This case provides incremental guidance and is not a water-shed moment.

Ericsson already has Appealed Judge Selna’s decision to the U.S. Court of Appeals for the Federal Circuit, including Judge Selna’s denial of Ericsson’s request for a jury trial.   That appeal will not go forward until after Judge Selna rules on the parties pending motions asking him to alter or amend his findings and judgment thereon.  Judge Selna set a Briefing Schedule and a February 12, 2018 hearing on those motions.  His decision on the motions may clarify issues and address any errors either party has identified.  Thereafter, a decision by the Federal Circuit in this case (typically within a year after the appeal starts) may provide more uniform guidance on litigating FRAND-committed SEPs.  However the case turns out, we appreciate the tremendous effort by Judge Selna and the parties in shedding more light on these developing SEP issues.

Summary Top-Level Points From Decision

Below are some generalized bullet point take-aways from Judge Selna’s instant decision and prior decisions in the case.  We encourage you to read our more detailed discussion and–more importantly–Judge Selna’s decision itself to understand the issues in full context:

  • Judge Selna indicated that the ETSI FRAND commitment did not require Ericsson to license non-essential patents, which would include patents not covered by the standard as well as patents that are in technical annexes to ETSI’s standard but were not “necessary to conform in order to claim compliance with the ETSI deliverable.”
  • Judge Selna indicated that a patent owner does not need to offer the same terms to the other similarly situated parties (e.g., Ericsson did not have to offer TCL a license on non-essential patents even though such patents were included in Ericsson’s license of SEPs to other companies; but the value of those non-essential patents should be taken into account when comparing what consideration was exchanged in different licenses).
  • Judge Selna indicated that Ericsson would not be bound by licensing terms determined in an arbitration with another party, because, he did not want to discourage use of arbitration as an efficient resolution vehicle.
  • Judge Selna indicated that competition law claims could not be based solely on the cost of litigating a patent infringement case (which petitioning the court by a patent owner is protected by the First Amendment under the Noerr-Pennington doctrine as long as its not a “sham” litigation) or business uncertainty from FRAND negotiation (which uncertainty is not an “economic injury” under California’s competition law).  Further, competition law claims could not rest solely on breach of the FRAND obligation.
  • Judge Selna indicated that the ETSI FRAND commitment should be interpreted under French law with TCL (the prospective licensee) being essentially a third-party beneficiary of an agreement between ETSI as the promissee and Ericsson (the patent owner) as the promisor.
  • Judge Selna indicated that ETSI rejected use of a “most favored nations” (or “most favored licensee”) condition for a FRAND license that would have required revisiting already entered licenses based on later licenses with others.
  • Judge Selna indicated that, in considering royalty rates for sales made within geographical/jurisdictional regions, the place were products are manufactured may set a floor for the lowest royalty rate in any country, because such products would be subject to the manufacturing country royalty rate (for infringement by “making” the invention) even if sold in another country with a lower royalty rate.
  • Judge Selna indicated that factors for determining which companies were “similarly situated” to TCL in comparing licensing terms for purposes of assessing the non-discrimination part of FRAND include (i) the geographic scope of the company (e.g., exclude “local kings” that only need a license in one country since TCL is a global company); (ii) the scope of the license required; and (iii) a reasonable sales volume.  Factors he found not relevant include (a) the company’s overall financial success or risk; (b) brand recognition and (c) device operating system.
  • Judge Selna indicated that comparable licenses have a role in the FRAND analysis and considering an array of licenses can substantially diminish concerns about whether individual licenses complied with FRAND, have asymmetric information or were unduly influenced by the threat of litigation.

Below also are some things to consider in applying Judge Selna’s decision:

  • Judge Selna’s decision was limited to determining FRAND in litigation based on the evidence before him.  He indicates several times in different portions of his decision that his approach and rulings do not necessarily apply outside litigation in the context of private party negotiations for a FRAND license.
  • Judge Selna’s reliance on a modified top down approach was justified in part by his finding that Ericsson had publicly and consistently supported a patent-counting, top-down approach using a specific aggregate royalty for all SEPs.  That approach may not apply to other SEP owners.


This case involves Ericsson’s large portfolio of patents alleged to be essential to ETSI’s 2G (GSM, GPRS and EDGE), 3G (W-CDMA) and 4G (LTE and LTE advanced) mobiile phone standards.  TCL is said to be the seventh largest manufacturer of mobile phones and other devices compliant with those standards.

Licensing Negotiations.  In March 2007, TCL affiliates entered 2G licenses with Ericsson having seven-year terms.  In 2011, the first year TCL had meaningful 3G phone sales, the parties started negotiating a 3G license.  From late-2012 to late-2014, Ericsson brought 11 lawsuits in 6 foreign countries against TCL on Ericsson’s SEPs.  In 2014, the parties started negotiating a 4G license.

TCL alleged that no offer or counteroffer made by Ericsson during their negotiations were on FRAND terms.  Ericsson reduced its royalty offers during the course of negotiations.  For example, Ericsson’s initial 4G offer in 2013 had a 3% royalty on the mobile phone price with a $3 “floor” (i.e., minimum royalty payment per device) and an $8 “cap” (i.e., maximum royalty payment per device).  During the course of negotiations, Ericsson reduced this 4% licensing amount in 2015 by about half to 1.5% royalty on the mobile phone price with a $2 floor and $4.50 cap.  Similarly, Ericsson reduced its initial 3G offer of July 2011 from a 2% royalty on the mobile phone price (with $2 floor and $6 cap) to a February 2015 offer of 1.2% royalty (with no floor or cap).

In March 2014, TCL filed the instant action when its initial March 2007 2G license with Ericsson was set to expire.  Ericsson filed a mirror-action case in E.D. Texas, which case was transferred to and consolidated with this case.

During the course of litigation, Ericsson made other licensing offers, leading to two alternative offers–Option A and Option B–considered by Judge Selna in his FRAND decision.  We discuss those offers in another section below.

In February 2017, Judge Selna held a 10-day bench trial–i.e., a trial before Judge Selna acting as the finder of fact, rather than using a jury as the fact finder.  On November 8, 2017, Judge Selna issued his decision in confidential form (i.e., it was not publicly available).

ETSI FRAND Commitment At Issue.  The FRAND commitment at issue arises from ETSI’s IPR Policy and letters of assurance (LOAs) that Ericsson submitted to ETSI that agreed to license its SEPs on FRAND terms under that IPR Policy.  Specifically, Section 6.1 of ETSI’s IPR Policy provides for a FRAND licensing commitment as follows:

When an ESSENTIAL IPR relating to a particular STANDARD or TECHNICAL SPECIFICATION is brought to the attention of ETSI, the Director-General of ETSI shall immediately request the owner to give within three months an irrevocable undertaking in writing that it is prepared to grant irrevocable licenses on fair, reasonable and non-discriminatory (“FRAND”) terms and conditions under such IPR to at least the following extent:

•  MANUFACTURE, including the right to make or have made customized components and sub-systems to the licensee’s own design for use in MANUFACTURE;

•  sell, lease, or otherwise dispose of EQUIPMENT so MANUFACTURED;

•  repair, use, or operate EQUIPMENT; and

•  use METHODS.

The above undertaking may be made subject to the condition that those who seek licences agree to reciprocate. [emphasis added]

Ericsson’s FRAND commitment concerned only “ESSENTIAL IPR” and not other non-essential patents.  Specifically, Section 15.6 of ETSI’s IPR Policy defines “ESSENTIAL” as follows:

“ESSENTIAL” as applied to IPR means that it is not possible on technical (but not commercial) grounds, taking into account normal technical practice and the state of the art generally available at the time of standardization, to make, sell, lease, otherwise dispose of, repair, use or operate EQUIPMENT or METHODS which comply with a STANDARD without infringing that IPR.  For the avoidance of doubt in exceptional cases where a STANDARD can only be implemented by technical solutions, all of which are infringements of IPRs, all such IPRs shall be considered ESSENTIAL.

 Prior Case Proceedings

During the course of this litigation, Judge Selna made several decisions prior to trial that helped shape the issues as ultimately tried and decided.  Following is a discussion of some of those decisions.

FRAND Royalty Does Not Include Implementation Patents.  In mid-2015, Judge Selna ruled that Ericsson’s implementation patents (i.e., patents that are not essential to practice the standards) would be outside the scope of the FRAND license considered in this case, because the ETSI FRAND commitment does not extend to patents that are not SEPs.  He rejected TCL’s argument that it was entitled to a license under Ericsson’s SEPs since Ericsson had licensed those patents to others when licensing its SEPs, stating

An ETSI obligation to offer an SEP on FRAND terms does not require the standard-essential patentholder to offer identical terms to all licensees.

Judge Selna also declined to enjoin Ericsson from pursuing infringement actions against TCL on Ericsson’s implementation patents (i.e., non-essential patents) because they would not be the subject of the instant FRAND dispute.

He did rule, however, that alleged comparable licenses to Ericsson’s SEPs that included non-essential patents should account for the value of the license to those non-essential patents because they are part of the consideration exchanged between the parties.

Agreed Global Anti-Suit Injunction.  The parties agreed that the instant action would result in a world-wide license for future use of Ericsson’s SEP technology as well as a release payment for past use that would globally resolve their disputes.  Accordingly, Judge Selna agreed to enjoin both parties from pursuing infringement actions any where in the world on the SEPs at issue in this litigation (often called an “anti-suit injunction”).  The parties submitted and Judge Selna entered an anti-suit injunction.

Will Not Consider Cross-License to TCL’s SEPs.  In November 2015, Judge Selna Ruled on Ericsson’s request that the FRAND adjudication include Ericsson obtaining a cross-license to TCL’s SEPs, rather than solely a one-way license to TCL on Ericsson’s SEPs.  Among other things, Ericsson argued that every offer it made to TCL included a cross-license to TCL’s SEPs, cross-licensing was consistent with the parties’ licensing practices, and a cross-licensing condition was permitted by ETSI’s IPR Policy.  Judge Selna, however, declined to consider a cross-license to TCL’s patents, because Ericsson failed to timely raise such cross-licensing in its Complaint or other pleadings.

Jury Would Have Determined All Disputed Material Terms; But Case Became a Bench Trial.  In its November 2015 ruling, Judge Selna ruled that a jury would determine all material terms over which the parties have a genuine dispute.  Although “boilerplate” or other terms might not be required to make a binding agreement to pay a monetary royalty, “the Court does not consider it fair to either party that the result of this adjudication could be a license or injunction that is simply silent as to terms that are material to such a license or injunction.”

In July 2016, Judge Selna Ruled that the jury’s tasks would include identifying material terms and “supply[ing] the FRAND term for each material term” based on competing options. [Order at 5]  The jury then would decide “whether all the terms, in the aggregate, meet the requirements of a FRAND license.”

The terms put before the jury would be limited.  TCL agreed that many of Ericsson’s proposed terms were not a basis for asserting that Ericsson breached its FRAND commitment. [Order at 6-7]    Further, Judge Selna ruled that Ericsson’s offer to license only SEPs, and not implementation or non-essential patents, did not breach Ericsson’s FRAND commitment. [Order at 7].  Judge Selna also ruled that it was not unreasonable or unfair for Ericsson to charge a royalty for a 2G license for 2G technology on a 3G chip that TCL purchases from a third-party where that third-party’s pass-through license did not include the 2G SEP technology. [Order at 8]

In a later order, discussed below, Judge Selna ruled that TCL could not obtain monetary damages for Ericsson’s alleged breach of its FRAND commitment.  He further ruled that this left only equitable remedies in the case and, therefore, this would be a bench trial, rather than a jury trial.  Ericsson objected and asserted a right to trial by jury, which is an issue that Ericsson’s has Appealed to the Federal Circuit. (See also Final Pretrial Order at 2)

Ericsson Not Bound By Licensing Terms To Huawei From Arbitration.  In May 2016, Judge Selna published his Order that denied TCL’s motion seeking summary judgment that Ericsson was bound to offer TCL the same licensing terms that were reached in an arbitration between Ericsson and Huawei.

In 2015, Ericsson and its licensee Huawei entered arbitration on three issues that they could not resolve in negotiations, including a FRAND royalty per mobile phone.  The arbitration panel found that some rates offered by Ericsson did not meet the non-discriminatory aspect of FRAND, but some did.  The arbitration panel also computed effective royalty rates paid by other licensees.

TCL argued that Ericsson was collaterally estopped from challenging the arbitration panel’s findings (specifically, TCL asserted “offensive nonmutual issue preclusion”).   Judge Selna found that equitable considerations counseled against applying collateral estoppel here.  Importantly, this case would result in a complete licensing arrangement and the FRAND commitment must consider the totality of monetary and non-monetary terms:

This Court has consistently held during the course of this litigation that the final result of this lawsuit will be a “complete, integrated, and enforceable relationship between TCL and Ericsson.”  The Court has recognized before, and recognizes again, that whether a licensing contract offer fulfills Ericsson’s FRAND obligations requires a holistic understanding of the offer and a weighing of the terms that include, but are not solely, the specific royalty rates …

At this stage, there were many material terms that the parties disputed.  In contrast, “Huawei and Ericsson agreed that resolution by arbitration of the three issues would produce a fair result given their agreement on all other issues,” but “[t]here is no like predicate here.”

Further, the arbitration panel itself “noted its limited task and the limited precedential value” of its findings:

In particular, the [arbitration] panel noted that even identical terms can take on a different character between discriminatory and nondiscriminatory depending on the nuances of circumstances.

Further, TCL had unfairly cherry-picked findings from the arbitration panel’s decision.  Specifically, TCL refused to accept geographic rate differences and a minimum royalty (or “floor”).  Judge Selna found it be unfair for collateral estoppel “to create a ceiling, but not a floor” on TCL’s payment obligations.

Finally, Judge Selna found that it would be against public policy to give the results of arbitration proceedings binding effect between the patent owner and other licensees, because that “discourages the use of arbitration as a means to resolve FRAND disputes and to finalize agreements that were mostly resolved by negotiation.”

In a later Motion In Limine Ruling, however, Judge Selna ruled that the resulting license between Huawei and Ericsson could be admitted at trial, but not the arbitration opinion itself or award:

The arbitration opinion reflects the thought process of three arbitrators employing a particular methodology to a particular (limited) record, without applying a burden of proof or other procedural safeguards of trial to their decision making.  Simply put, the task that the Huawei arbitrators were engaged in is significantly different than the task posed to the fact-finder in this case.

TCL could state that the Huawei license resulted from an adversarial arbitration proceeding.

Competition Law Claims Dismissed.  In August 2016, Judge Selna Dismissed TCL’s California Unfair Competition Law (“UCL”) claims based on the Noerr-Pennington doctrine, which provides First Amendment protection to a party petitioning the courts for remedies, such as infringement lawsuits.  The Noerr-Pennington doctrine precludes using the expense of defending injunction or exclusion order litigations as an “economic injury” for purposes of a competition law claim:

Although there is evidence in the record that TCL spent millions of dollars defending against actions by Ericsson seeking injunctions or exclusion orders, such injury cannot be the basis for TCL’s “economic injury” due to the Noerr-Pennington doctrine.

TCL had not established an exception to the Noerr-Pennington doctrine, such as showing that the cases filed by Ericsson were “sham” litigations.  TCL had not shown any other economic injury for purposes of a competition law claim.  For example, he rejected TCL’s argument that Ericsson’s conduct caused business uncertainty harm that would give rise to the competition law claim:

[B]usiness uncertainty alone does not count as “economic injury” under the UCL because it is not “money or property.”

Judge Selna also ruled that TCL had not established a competition law claim based on “unfair” activities by Ericsson.  TCL’s claim was premised on the Third Circuit’s decision in Broadcom v. Qualcomm, 501 F.3d 297 (3d Cir. 2007), which held that “a firm’s deceptive FRAND commitment to [a standard development organization] may constitute actionable anticompetitive conduct.”  In this case, following extenstinsive discovery, TCL did not provide “any evidence of an intentionally false promise to a standard setting organization.”  Further, even if Ericsson did breach its FRAND commitment, “TCL has submitted no evidence that Ericsson ever had an intent to do so.”

Judge Selna ruled that the competition law claim must rest on more than just breach of the FRAND obligation:

The crux of TCL’s argument seems to be that because TCL has evidence that Ericsson has breached FRAND, TCL has evidence that Ericsson has necessariliy done so in an anticompetitive way.  The Court has not been provided with any authority for this kind of bootstrapping of a breach of contract case, without more, into a violation of the policy or spirit of antitrust laws.  Instead, as the Broadcom court made the “intentionally false promise” and subsequent “breach of that promise” two separate elements, it is clear that there needs to be some other conduct by Ericsson than mere breach of its FRAND obligations.

Dismissed Damages Claims.  In another August 2016 ruling, Judge Selna Granted Ericsson’s motion for summary judgment that TCL could not show cognizable damages for TCL’s breach of contract or promissory estoppel.  TCL’s sole measure of such damages were its foreign attorney fees incurred in the foreign litigations.  TCL gave only ballpark figures of  $20M to $30M in witness testimony, but provided conflicting written invoices after discovery showing only $4.6M in fees.  Judge Selna ruled that TCL’s responses did not comply with TCL’s discovery obligations, so they would be excluded.

Did TCL Lose FRAND Rights Due To Its Negotiation Misconduct?  Judge Selna Ruled that, because he dismissed TCL’s damages claims, he did not to need to determine whether TCL lost FRAND commitments rights based on misconduct.  Judge Selna had tentatively ruled “that TCL could have temporarily ‘lost’ its rights as a third party beneficiary to Ericsson’s contract with ETSI during those periods of time when TCL was frustrating the ability of Ericsson to actually execute a FRAND license with TCL.”  If there are no damages for any breach of contract by Ericsson, then there is no need to determine whether there were times when Ericsson’s contractual obligations were suspended by TCL’s alleged bad negotiation conduct.

Ericsson’s Option A and Option B FRAND Offers

Judge Selna required Ericsson to file what it considered to be an offer on FRAND terms.  Ericsson ultimately submitted two options that the Court would consider.  Ericsson submitted two options of alleged FRAND terms that provided a worldwide license with running royalties as a percentage of the mobile phone selling price–Option A and Option B–based on offers Ericsson had made to TCL in April 2014 and February 2015, respectively.  In March 2016, Ericsson also proposed a third Option C that for the first time was based on a fixed royalty per unit rate, rather than a royalty as percentage of the mobile phone price.  Ericsson argued that such a per-unit royalty was more consistent with Ericsson’s revised, current licensing practices.  But Judge Selna Ruled that he would not consider the third Option C, because it was not timely submitted.

Judge Selna Ruled that evidence about licensing events after Ericsson’s Option A and Option B offers could be considered by the factfinder as being probative of those offers.  But he further ruled that some measures could be taken to ensure that Ericsson’s 2014 and 2015 conduct (when Option A and B Offers were made, respectively) would not be judged in a 2016 climate.  Ericsson could submit evidence, seek instructions or take other appropriate action to assert that “later-executed agreements were formed in a climate very different from the one in which the Option A and Option B offers were made.”

In a later Ruling, Judge Selna ruled that the fact finders determination that either Option A or Option B were FRAND would permit the court to enter such option as the final judgment in the case (and Ericsson suggested that TCL would chose between them if both were found to be FRAND).  To do otherwise–e.g., permit TCL to modify those options–would be unfair:

[A]ny other procedure following a determination that Option A or Option B are FRAND would be unfair to Ericsson, because it would create a “risk free” opportunity for TCL to improve on Ericsson’s licensing proposals through the mechanism of trial.

Below is a table summarizing the offers for Option A and Option B based on the court’s summary of them (see Opinion at 7-8).  Mobile Phone in table below includes feature phones, smartphones and tablets.

 ETSI Standard

Option A

Option B


Mobile Phone: 0.8% of net price for GSM/GPRS; 1.1% for EDGE

External Modem: 1.5% of net price with $0.40 floor

Personal Computer: $0.50 per unit for GPRS, $0.75 per unit for EDGE

Mobile Phone: 0.8% of net price for GSM/GPRS; 1.0% for EDGE

External Modem: $0.75 per unit

Personal Computer: Same as Option A


Mobile Phone: 1.5% of net price

External Modem: 1.5% of net price with $0.40 floor

Personal Computer: $2.25 for single mode; $2.75 for multimode

Mobile Phone: 1.2% of net price

External Modem: $0.75 per unit

Personal Computer: Same as Option A


Mobile Phones: 2.0% of net price

External Modem: $3 per unit if net price < $60; else $1 per unit.

Personal Computer: $3.5 per unit

Mobile Phone: 1.5% of net price with $2.00 floor and $4.50 cap

External Modem: 1.5% of net price with $2 per unit floor.

Personal Computer: Same as Option A

Geographic Differences China Royalty: ½ of rates above.
Annual Payment Mobile Phone: $30 million for first $3B in sales.  Effective 1% royalty rate for $3B in sales
Release Payment for Past Infringement Yes Yes

 Judge Selna’s FRAND Decision

Judge Selna summarized his task as involving three main steps:

  • Determine whether Ericsson met its FRAND obligation.  Judge Selma ultimately determined that Ericsson had met its FRAND obligation with respect to its course of conduct.  He decided he need not resolve whether Ericsson was required to make a FRAND offer during the negotiations.
  • Determine whether Ericsson’s final offers were FRAND offers: Offer A and Offer B.  Judge Selma ultimately determined that those offers were not FRAND.
  • If Ericsson’s offers were not FRAND, then  determine material FRAND terms using either (a) a “top-down” approach argued by TCL to determine Ericsson’s portion of a total aggregate royalty under the standards or (b) an “ex ante” approach argued by Ericsson that measures the value Ericsson’s patents add to a product.  Judge Selna ultimately rejected the approaches of both parties.  He used a modified Top Down approach based on evidence submitted by both parties.

Following is a detailed discussion of his decision and ultimate conclusions.

Judge Selna’s Interpretation of the ETSI IPR Policy

Judge Selna ruled that ETSI’s IPR Policy was governed by French law because (i) the parties did not dispute that French law applied and (ii) ETSI’s IPR Policy itself states that “[t]he construction, validity and performance of this IPR information statement and licensing declaration shall be governd by the laws of France.”

Judge Selna ruled that Ericsson’s licensing obligation arises from “ETSI’s acceptance of a patent holder’s patent as an SEP form[ing] a contract.”  [NOTE: This statement by Judge Selna probably should be clarified to mean ETSI’s acceptance of the patent holder’s declaration for patents that may be essential to the standard, because–as Judge Selna states in footnote 10–ETSI does not determine whether a patent actually is essential to a standard.]

Judge Selna ruled that TCL could enforce this contract under the French law doctrine stipulation pour autrui, which is akin to a third-party beneficiary under U.S. law:

ETSI is the promisee, the owner of a SEP who submits the IPR licensing declaration is the promisor, and the third-party beneficiaries are prospective licensees who benefit from the stipulation. [Memo at 9]

When interpreting a contract under French law, “[t]he main objective is to determine the common intent of the parties.”  Further, the contract should be interpreted to be “internally consistent” and in compliance with the law.

Judge Selna ruled that, under the IPR Policy, ETSI itself does not grant rights to IPR.  Rather, “the FRAND undertaking is to be expressly interpreted as an encumbrance on the IPR, where applicable under the laws of the jurisdiction.” [Memo at 11]. [NOTE: Classifying the FRAND commitment–which is not itself a license–as an “encumbrance” on a patent may be relevant to whether the FRAND commitment survives transfer of the patent to a new owner, such as occurs with an outright license of the patent; but that was not an issue in this case.]

Judge Selna stated that patent “hold up” was one of the concerns behind ETSI’s IPR Policy:

In formulating its IPR Policy ETSI was concerned, among other things, with addressing the problem of “hold up.”  Hold up occurs when a patent holder seeks to extract more for the use of his patent than the value which his patent adds to a standard. … If a firm takes a license and incorporiates that technology in its product, it cannot easily take an alternative path in developing and marketing its product.  This lock-in “tilts the negotiating balance in favour of the IPR owner,” such that “the term ‘fair and reasonable’ for royalty becomes whatever anyone cares to demand,” increasing the risk that “[s]mall enterprises get pushed out of the market.” [Memo at 11-12].

Judge Selna also stated that another concern was “price discrimination among potential licensees.” [Memo at 12].  As one example, there was a “serious risk of distortion of market forces against [small and medium-sized enterprises] and in favour of large multinationals.”  Judge Selna stated that ETSI IPR Policy “forbids discrimination based on nationality or ETSI membership,” but also was concerned about other types of discrimination, such as “protecting small and medium-sized enterprises.”

Judge Selna observed that some ETSI participants considered a more specific definition of what constitutes FRAND, but ETSI could not reach consensus among all participants of whether and to what extent ETSI should more specifically define FRAND. [Memo at 12-13].  This has “left the resolution of FRAND-related disputes to the national courts.”

Judge Selna did find some guidance from ETSI’s initial consideration–and later rejection–of a “most favored nations (or here licensee) concept” found in a 1993 version of ETSI’s IPR Policy that would have required re-opening and re-negotiating existing licenses based on subsequent licensing of others; that provision was deleted in 1994.  That deleted most-favored licensee provision stated:

[A licensor must] promptly notify a licensee of any licence granted by it to a third party for the same IPRs under comparable circumstances giving rise to terms and conditions that are clearly more favourable, in their entirety, than those granted to the licensee and allowing the licensee to require replacement of the terms and conditions of its licence, in their entirety, either with those of the third party licence, or with such other terms and conditions as the parties may agree. [Memo at 13].

A “doomed undertaking”

At the outset, Judge Selna indicates that the royalty analysis could not be precise, would be subject to criticism and would be limited by the evidence and arguments presented by the parties:

Before turning to the royalty setting analysis advanced by the parties’ experts, the Court makes one central observation as the fact finder in this case.  The search for precision and absolute certainty is a doomed undertaking.  The complexity of the analyses and the number of variable components inevitably lead to criticism.  Indeed, there are facial limitations in the analyses themselves.  The Court’s effort has been to determine whether each expert’s work has a reasonable level of reliability and convincing force that allows the Court to make a judgment whether to accept the ultimate conclusion advanced. [Memo at 14].

As two examples, he notes that one expert only used U.S. patents and another expert only used English language patents.

Judge Selna’s candor here is appreciated and practical.  It is one reason why one may be mistaken by taking the analysis and ultimate FRAND royalty determination in this case as being applicable to other cases.  For this practical reason, the remainder of our summary of the Court’s decision will focus more on general principles or learning moments from the decision and may not dig fully into the details of the analysis (which you can find within the decision itself).

TCL’s Top Down Method of SEP Valuation

Judge Selna reviewed a “top down” royalty methodology that TCL presented.  Judge Selna found it “significant[]” that Ericsson did not present its own top down model.  The top down approach apportions the relative value of a specific patent owner’s SEP portfolio from a projected aggregate royalty for all SEPs for that standard. Basically, “compute[] a fraction of the aggregate royalty where the numerator is the value of the SEPs owned by Ericsson for that standard, and the denominator is the total value of all SEPs in that standard”:

Royalty = (Aggregate Royalty)  x  (Ericsson SEP value/All SEP value)

Judge Selna observed that the top down approach was appealing because “it prevents royalty stacking,” which occurs “when each individual SEP holder demands a royalty which when totaled exceeds the value of all the SEPs in a standard.” [Memo at 15]  Further, it “can also prevent hold-up because it prevents SEP owners from charging a premium for the value added by standardization.” [NOTE: Judge Selna’s statement here and elsewhere about royalty stacking and patent holdup does not cite any particular evidence that royalty stacking or holdup exists in this case.  Generic statements about royalty stacking and patent holdup–which may cause the fact finder to err on the side of awarding a lower royalty–might not be allowed were this a jury trial. (see our December 5, 2014 post on the Federal Circuit’s decision in Ericsson v. D-Link, 773 F.3d 1201 (Fed. Cir. 2014)).  We posit no opinion, however, whether generic concepts of royalty stacking or holdup erroneously had any substantive influence on Judge Selna’s analysis here.]

Judge Selna stated that TCL’s top down approach considered the essentiality, importance and contribution of Ericsson’s patents that accounted for the value of expired or acquired patents as well as regional (geographical jurisdiction) differences among the patents.  But the approach could not address discrimination and “is not necessarily a substitute for a market-based approach that considers comparable licenses.” [Memo at 15].

TCL’s top down approach had nine steps:

  1. Select maximum aggregate royalty for all SEPs for the standard (6% of mobile phone price for 4G; 5% for 2G/3G)
  2. Determine total number of SEPs for the standard.
  3. Rank Ericsson’s claim charted patents on an essentiality scale of 1 to 3 (1 = No Evidence Claim Is Not Essential, 2 = Might Not Be Essential Depending on Claim Construction and 3 = Not Essential Under Any Reasonable Claim Construction)
  4. Evaluate importance and contribution of each Ericsson SEP patent family (TCL experts found a “relatively low value” for Ericsson’s SEPs)
  5. Adjust value of Ericsson SEPs based on the importance/contribution of Step 4
  6. Apply a forward-citation analysis as a check against the determined value of Ericsson’s SEPs, which “attempts to determine the value of U.S. patents based on the frequency with which they are cited in later patent applications.”
  7. Adjust value based on fluctuation of SEPs due to patents expiring or being acquired.
  8. Adjust value based on geographical location (Ericsson’s U.S. patents were strongest and foreign patents had weaker value).
  9. Use TCL’s sales data to weight royalty be region and blend regional royalties together “to create a single global royalty rate for each standard (TCL found 0.16% of price of 4G mobile phones and 0.21% of price of 2G/3G mobile phones). [Memo at 15-16]

Judge Selna rejected TCL’s final results based on problems in TCL’s Steps 4-6 and 9 above.  But he did use some of TCL’s data to derive his own royalty under a modified top down approach based on “a simple patent counting system which treats every patent as possessing identical value” and then makes adjustments based on some reliable values from TCL’s analysis.  Judge Selna applied two general formulas for his top down analysis to determine Ericsson’s Royalty Rate (i.e., percentage of device’s net selling price):

An important justification for Judge Selna’s use of the top down approach in this case was his finding that “Ericsson has long argued that a fair and reasonable royalty rate for a SEP license can be determined using a top down approach, or what the Court calls a simple patent counting system.” [Memo at 18]  For example, in 2008, Ericsson stated on its website that its license comply with the “prevalent industry interpretation of FRAND, i.e. the basis is a reasonable maximum aggregate royalty rate to which each patent holder is entitled a proportion according to its relative share of all standard essential IPR.”  He found that Ericsson repeatedly confirmed during discovery in this case that Ericsson’s licensing policy was based on such a top down approach. [NOTE: Pure patent counting that would apportion royalties based purely on the number of SEPs owned by a patent holder and treat all patents as though they have equal value has been highly criticized.  Judge Selna justified his top down approach here because he found that the patent owner made public statements supporting this approach.  This justification may limit applying such a top down approach to SEPs owned by others.]

Aggregate Royalty Rate: 5% for 2G/3G and 6% or 10% for 4G.  TCL relied on an aggregate royalty burden suggested in prior Ericsson statements and did not independently calculate what would be an aggregate royalty burden. [Memo at 19].  Among other things, those Ericsson statements were made around the time the standards were adopted and when Ericsson sold handsets and, thus, had an incentive to strike a balance because it would be both a licensor and licensee.  Judge Selna would have considered a higher aggregate royalty based on improvements made to the standard after Ericsson’s prior statements, but Ericsson did not present “evidence that showed the value of subsequent additions to each standard.” [Memo at 19-20]

For the 2G/3G standards, Ericsson was part of a joint press release that suggested pure patent counting “whereby essential patents for W-CDMA are licensed at rates that are proportional to the number of essential patents owned by each company … at a modest single digit level.”  In that press release, Nokia endorsed a 5% royalty and NTT DoCoMo suggested “keeping cumulative royalty rate below 5%.”  Although the 5% royalty rate might not apply to other groups “not a part of this press release,” it was fair to use the 5% royalty rate in this case, which “successfully induced manufacturers to adopt the 3G W-CDMA standard.” [Memo at 21-22].

For the 4G standards, Ericsson’s website stated that Ericsson expected a 1.5% royalty on a mobile phone’s net price because its SEPs represent 20-25% of the total SEP value to the standard and a reasonable maximum aggregate royalty would be 6-8% of the mobile device net price.  Ericsson also had suggested a “single-digit” aggregate royalty burden for LTE in its discovery responses to TCL in this case.  Judge Selna found that Ericsson’s statements were not merely “a hope or prediction, but a pledge to the market that if the market adopted Ericsson’s championed standard, the total aggregate royalties would be calculated as described above.” [Memo at 24] [Note: Judge Selna justified using the specific aggregate rates he determined here because he found that the patent owner publicly endorsed those rates to induce adoption of the standard.  A different aggregate rate might apply for other SEP holders in this standard.]

Judge Selna rejected Ericsson’s reliance on surveys (showing an aggregate royalty from 28.8% to 37.3%) and a summation of publicly announced royalties from nine companies (totaling 14.8%), because he would expect declared royalties to be high (since it would be an upper limit or sealing in license negotiations) and it was not shown that the other patent holders actually had licensed patents at those rates.

Based on the foregoing, Judge Selna found “some merit” in applying the top down approach for Ericsson’s patents in this case, even though the approach “is not perfect”, for five reasons:

  1. It relies on statements by Ericsson and others to induce adoption of the standard when the risk of patent hold-up was low;
  2. The statements were made before the standard was adopted, so SEP owners had incentive to be reasonable and reducing risk of hold-up or royalty stacking;
  3. Ericsson had an incentive to be fair because it was both a licensor and licensee;
  4. Ericsson still stands by this methodology; and
  5. It provides at least a maximum cealing for a FRAND rate, “because increasing the royalty rate after the standard has been adopted, without showing that the increase is due to additions to the standard, is the definition of hold-up.” [Memo at 25-26]

Judge Selna also found that this approach “hews to the principle of setting rates to reflect Ericsson’s own estimate of the total value the licensed technology contributed to the product.”

Based on the foregoing, Judge Selna “applies the 5% figure to 2G/3G, and applies both 6% and 10% to 4G.” [Memo at 26]

Ericsson’s Proportional Share of SEPs.  Ericsson’s Proportional Share is the ratio of the number of Ericsson SEPs over the total number of SEPs for the standard. [Memo at 26].   This required the court to determine how many SEPs are in each standard and how many such SEPs were owned by Ericsson.

To assess essentiality, Judge Selna used the definition of ESSENTIAL from ETSI’s IPR Policy (presented in the background above).  The only dispute on the definition of ESSENTIAL was whether patents that covered the technical informative annexes to the 3G standard were “ESSENTIAL”.  Ericsson argued it covered those technical annexes because ETSI’s definition of standard includes “any standard adopted by ETSI including options therein.”  TCL argued, however, the ETSI’s directives state that informative annexes “shall not contain provisions to which it is necessary to conform in order to claim compliance with the ETSI deliverable.” [Memo at 27]  Judge Selna agreed with TCL and held that patents that covered only technical features in an informative annex “while part of the standard, are not standard-essential patents.”  This led Judge Selna to exclude at least one family of Ericsson’s patents from the list of Ericsson’s SEPs.

Total Number of SEPs.  Judge Selna considered an essentiality study by TCL experts to determine the number of industry-wide SEPs.  This study had several steps [NOTE: Sometimes the decision was unclear in places whether the numbers represent individual patents or patent families that have several patents]:

  1. Looked at all IPR declarations recorded in ETSI’s database as of September 2015, leading to identification of over 153,000 patents and applications.
  2. Excluded patent families with only expired patents or that were not in English, leaving 11,469 patent families.  Judge Selna found this was erroneous, but the error had a relatively small impact so “[t]he Court is satisfied that the subset actually examined was a reasonable surrogate for the whole.” [Memo. at 28]
  3. Excluded patent families that did not have claims directed to user equipment (e.g., mobile phones), leaving 7,106 patent families. [NOTE: This step is not explained.  Cellular standards generally concern an overall system — e.g., how a mobile phone and cellular infrastructure interact for overall system efficiency.  And a mobile phone may need a license for patent claims directed to a system, even if the claim is not specifically directed to just the mobile phone, because the system claim may be infringed by the user of the mobile phone (see, e.g., NTP v. RIM, 418 F.3d 1282 (Fed. Cir. 2005)).  But the patent owner’s licensing program might cover such infringement in agreements with infrastructure or other licensees.  Its not clear whether and to what extent this issue was considered in the analysis.].
  4. Sorted the remaining 7,106 patent families based on which standard they were declared for (2G, 3G and/or 4G) and by patent holder for the 15 largest patent holders.
  5. Analyzed essentiality for a random sample of 1/3 the patents for each standard per top 15 patent holder–i.e., for 2,600 patent families.  Double-checked 17% of those essentiality determinations and adjusted the results based on a 9.5% disagreement of whether those patents were essential.
  6. Resulting Global SEP Patent Families per Standard:
    2G – 446 SEP families
    3G – 1,166 SEP families
    4G – 1,796 SEP families
  7. Resulting SEP Patent Families with U.S. Patents:
    2G – 413 SEP families
    3G – 1,076 SEP families
    4G – 1,673 SEP families

Ericsson made several challenges to the process above, but did not propose alternative numbers. [Memo at 30].  Ericsson’s challenges were:

  • TCL’s analysis consultants “spent on average 20 minutes and charged only $100 per patent”; in contrast, the Via Licensing pool charges $10,000 per patent to determine essentiality.  Judge Selna found that the effort here was sufficient to provide a workable size and did not need the precision of a licensing pool.
  • TCL’s analysis consultants did not read the entire patents, which would inform means-plus-function limitations, have terms defined by the patentee as its own lexicographer, and could have disclaimers of claim scope.  Judge Selna found this was a “more salient” concern, but (presumably absent specific evidence of error) the court would not speculate whether this actually was a “substantial” or “entirely trivial” concern.
  • Some of TCL’s analysis consultants were not qualified.  Judge Selna found that Ericsson’s cross-examination did not convince him they were unqualified.
  • TCL’s analysis consultants knew that TCL was paying them for the analysis.  Judge Selna found that a blind test was not required, and potential bias exists for every expert in the case.

Judge Selna concluded that the flaws in analysis were not significant enough to entirely reject TCL’s consultants’ analysis.  He would use a estimate of how much TCL’s analysis over-designated patents to be essential (11.4%) as a way to downwardly adjust the total number of patents that TCL designated as SEPs.

Based on the foregoing, Judge Selna found that the total number of SEPs in each standard was as follows:

2G – 365 Total SEPs
3G – 953 Total SEPs
4G – 1481 Total SEPs

Total Number of Ericsson SEPs.  Judge Selna next considered the total number of SEPs owned by Ericsson.  Ericsson identified 235 patent families, but provided claim charts for only 192 of them.  TCL disputed whether some were essential to the standards.  Judge Selna ultimately considered both Ericsson and TCL’s asserted number of Ericsson SEPs:

Number of Ericsson SEP Patent Families: 2G 3G 4G
TCL’s Argument: 29 33 74
Ericsson’s Argument: 29+2 33+14 74 + 51

Judge Selna then considered TCL’s analysis of the impact on Ericsson’s proportional share of SEPs based on new patents being added as SEPs to the standard.   TCL’s expert concluded there would be no significant change “because Ericsson can only obtain additional patents when the standard grows.” [Memo at 35].  Judge Selna was “skeptical that [such] model is the best way to estimate the growth of the 4G standard.”  Nonetheless, Judge Selna assumed that “Ericsson’s newly acquired SEPs will be offest by SEPs being added to the standard” such that the court will assume Ericsson maintains a constant proportional share of SEPs. [Memo at 35].

Judge Selna then considered adjustments based on patents expiring, because no license is required for patents for use after they expire under U.S. law and a FRAND commitment “cannot permit what domestic patent law prohibits.” [Memo at 36]. [NOTE: The Supreme Court’s Kimble v. Marvel decision held that a patent owner cannot charge royalties for use of an invention that occurs after a patent expires.  The Supreme Court, however, further suggested that a patent owner can structure payments that occur after a patent expires based on a royalty for use of the invention that occurred during the patent term — e.g., spread-out payments for the convenience of the parties.  (See our June 22, 2015 post)]  Judge Selna used the date of closing arguments (May 18, 2017) as the date at which patents must be active (not expired) in order to be part of a forward-looking license.  Importantly, Judge Selna found that expired SEPs should not be removed from the count of total SEPs, because the aggregate rate was based on the assumption that all SEPs were active (not expired):

To remove expired patents from the denominator [i.e., total number of SEPs] (without decreasing the total aggregate royalty) would result in transferring the value from expired inventions to the remaining patents in the standard instead of to the public. [Memo at 36]

Judge Selna considered how many months each of Ericsson’s SEPs would be active (not expired) during the 5-year period that would be used as the licenses term.  He added all of those months and then divided by the total number of SEPs to get an average that represents the total effective number of unexpired SEPs Ericsson will own during the 5-year license period.  This led to the following effective number of Ericsson SEPs:

Effective Number of Active Ericsson SEPs 2G 3G 4G
TCL’s Numbers 12 19.65 69.88
Ericsson’s Numbers 12 24.65 111.51

Having now determined the effective number of Ericsson SEPs and the total number of SEPs, Judge Selna calculated Ericsson’s Proportional Share of SEPs using TCL and Ericsson’s different numbers for Ericsson SEPs:

Ericsson’s Proportional Share of SEPs 2G 3G 4G
TCL’s Numbers


(12 of 365 SEPs)


(19.65 of 953 SEPs)


(69.88 of 1481 SEPs)

Ericsson’s Numbers


Same as TCL proposed


(24.65  of 953 SEPs)


(111.51 of 1481 SEPs)

Adjust For Relative Strength of Ericsson SEPs.  Judge Selna next considers TCL’s analysis of the relative strength of Ericsson’s SEPs, but “found it too flawed to be used to calculate the overall rates.” [Memo at 38]  Considering the relative strength was important because “even in the universe of standard essential patents, many are relatively trivial, while some are key features of the standard.”

TCL rated the “importance” and the “contribution” of each SEP.  Specifically, TCL rated the “importance” of an SEP on a scale from 1 to 3, with 1 being “important or technically valuable”, 2 being “moderately important” and 3 being “only marginally important.”  TCL also rated on a scale from 1 (no alternative) to 4 (no improvement over alternative) the “contribution” of the SEPs, which would be small–even for “important” SEPs–if “there were other almost as useful options ETSI could have chosen when the standard was adopted.”  TCL did not limit alternatives to only those actually considered by ETSI, but included alternatives identified in the SEPs or their file histories and “any other technical solutions that would have been known to a person of ordinary skill in the art.” [Note: Recall that Judge Holderman in the Innovatio litigation limited alternatives to only those that the standard setting body actually considered. (see our Oct. 3, 2013 post).]

TCL found only 13 of 146 of Ericsson SEP families had a high important score and contribution score (i.e., a score of 1 or 2).

TCL attempted to determine Ericsson’s SEP value share by reliance on a cross-industry academic paper (by Dr. Jonathan Putnam) asserting that “most patents are worth very little” and divided patents into tiers of value in a given industry:

Top 10% of patents makeup 65% of all value of patented technology in the industry;
Next 10% of patents worth 14.6% of all patent value;
Bottom 50% of patents are 4.8% of all patent value

TCL computed Ericsson’s SEP value share by applying that academic paper’s results to TCL’s finding that only 13 of Ericsson’s SEP families were in the Top 10% Tier (because both the “important” and “contribution” scores were high).  TCL also cross-checked this with a forward citation analysis, “which attempted to determine the strength of patents by examining how often they are cited in future patent applications.” [Memo at 40].  Below are the results of TCL’s analysis:

2G – 6.7% of U.S. patents (Academic Paper) or 8.1% (Forward Citation Analysis)
3G – 4.0% of U.S. patents (Academic Paper) or 5.7% (Forward Citation Analysis)
4G – 3.1% of U.S. patents (Academic Paper) or 4.0% (Forward Citation Analysis)

Judge Selna, however, found several flaws with TCL’s analysis:

  1. TCL did not consider the relative value of all SEPs in the standard, so there was nothing to compare against TCL’s rankings of Ericsson’s SEPs.
  2. TCL’s alternative technology analysis did not consider if they would be inconsistent with each other, would perform worse, would maintain a viable functioning standard or would require a license to other Ericsson patents.
  3. TCL’s alternative technology analysis was legally flawed, because it did not consider who owns the alternative and the cost of the alternative: “How much proposed alternatives will affect the value of a patent depends on a number of variables, including whether the alternative is unpatented, expired, part of the previous standard, owned by another company that lets manufacturers use it for free or at a low rate, [owned by] an entity that aggressively protects its intellectual property, or [owned] by the company itself.” [Memo at 41-42]
  4. TCL’s “importance score” did not impact TCL’s estimated value.  And TCL’s classification of Ericsson’s patents “contribution” had no meaningful explanation between rankings to use as a basis for determining  if an SEP was in the top 10%.
  5. TCL’s expert’s reliance on the Dr. Putnam academic paper that 10% of patents provide 65% of all patented value conflicted with his testimony in the Innovatio litigation based on a different paper that 10% of WiFi SEPs provided 84% of all patented value.
  6. TCL’s forward-citation analysis contradicted its importance/contribution analysis, does not appear to have been used by any other court and “the Court is not convinced on this record that it provides a meaningful way to value SEPs.” [Memo at 42-43]

Based on the foregoing flaws, Judge Selna would not accept TCL’s valuation numbers.  He did find it useful generally, however, as showing that “Ericsson’s patent portfolio is certainly not as strong or essential as it has claimed.”

Value of Ericsson’s Non-U.S. SEPs.  Judge Selna found that Ericsson’s SEP portfolio was weaker outside the U.S., and a FRAND royalty should account for those geographic differences (although it was okay and “a matter of commercial reality [that] firms regularly adopt a single world-wide rate”). [Memo at 43].  An important caveat is that, because patents can be enforced in the country where products are made (and not just on where they are imported, sold or used), the strength of SEPs in the country where the licensee makes its products “sets a global floor for the manufacturer’s FRAND rate.”[Memo at 44].  In this case, because TCL makes its products in China, the strength of Ericsson’s Chinese SEPs determines the lowest FRAND rate for any product TCL sells globally.  Further, as a practical matter of administering justice, “[w]here geographic disparities are relatively insubstantial or unsupported by the evidence, the Court disregards them in favor of a more understandable, administrable, and enforceable royalty structure.” Accordingly, Judge Selna considered three regional areas: United States, Europe and ROW (i.e., the rest of the world).  Judge Selna relied on valuation from TCL’s expert to derive the following value share as compared to Ericsson’s U.S. SEP value share:

Value Share Relative to Value of U.S. SEPs 2G 3G 4G
U.S.  100% 100% 100%
Europe  72.2% 87.9% (None)
ROW (rest of the world) 54.9% 74.8% 69.8%

Plugging-In The Numbers.  Finally, Judge Selna plugged-in all of the different numbers discussed above into the formula he discussed above to determine Ericsson’s Royalty Rate on its SEP portfolio, i.e.:

This lead to several different combinations of values for each iteration of the standard (2G, 3G and 4G) across three regional areas (U.S., Europe, and ROW) using both TCL and Ericsson’s SEP patent counts and both a 6% and 10% aggregate rate for 4G — i.e., there were many different and alternative royalty rates computed.  We show below one calculation as an example of plugging-in Judge Selna’s determined numbers into the equation above–i.e., Ericsson’s Royalty Rate for its 2G SEPs as a percentage of the devices average selling price (“ASP”):

  • Aggregate Royalty for 2G:  5% of device’s average selling price (“ASP”)
  • Effective Number of Ericsson SEPs for 2G: 12
  • Total number of 2G SEPs: 365
  • Regional Strength Ratio of Ericsson 2G SEPs: U.S. (100%), Europe (72.20%) and ROW (54.90%)

Judge Selna’s plugging-in of the various numbers and options into the equation led to the results shown in the following table:

Standard Aggregate Rate Effective Number Ericsson SEPs U.S. SEPs Europe SEPs ROW SEPs
2G 5% 12 0.16402% 0.11842% 0.090049%
3G using TCL SEP count 5% 19.65
(per TCL)
0.10309% 0.090618% 0.07711%
3G using Ericsson SEP count 5% 24.65
(Ericsson count)
0.12932% 0.11367% 0.09674%
4G using 6% Rate and TCL SEP count 6% 69.88
(TCL count)
0.28297% 0.19751%
4G using 6% Rate and Ericsson SEP count 6% 111.51
(Ericsson count)
0.45145% 0.31517%
4G using 10% Rate and TCL SEP count 10% 69.88
(TCL count)
0.471611% 0.32918%
4G using 10% Rate and Ericsson SEP count 10% 111.51
(Ericsson count)
0.752576% 0.52529%

Judge Selna then compared the calculated rates for U.S. SEPs to what he called the “unpacked” rates from Ericsson’s Option A and Option B offers (“unpacked” meaning the specified royalty rate plus a quantification that accounts for non-monetary consideration).  Below is a comparison of the highest rates from the top down approach (in table above) to the unpacked Option A and B rates:

Royalty Rate for U.S. SEPs 3G U.S. SEPs 4G U.S. SEPs
Top Level Rate Range for U.S. SEPs 0.10309% to 0.12932% 0.28297% to 0.752576%
Unpacked Option A Rate 1.3168% 1.4659%
Unpacked Option B Rate 1.5000% 2.5918%

Even though he had “reservations” about the Top Level analysis, the disparity between those calculations and Ericsson’s Unpacked Option A and B offer rates shows that those options are “not fair or reasonable offers” in view of top down measure. [Memo at 50].

Ericsson’s Ex Standard Method

As an alternative to the modified Top Down Method discussed above, Judge Selna considered an ex-Standard method used by Ericsson, and ultimately found that–as with TCL’s Top Down approach–the analysis was “flawed at multiple steps” and rejected its conclusions. [Memo at 50].  The method sought to analyze patent value on a feature-by-feature basis along with the contribution of Ericsson’s patented technology to that feature — e.g., valued an improved battery life feature of the 4G standard and determined what percentage is attributable to the contribution Ericsson’s patented technology.  This led to an estimated value of Ericsson’s 4G SEPs for just two features (improved battery life and faster data rate) to be at least $6.15 per mobile device even though “no Ericsson licensee has ever paid anywhere close to $6.15 per phone for a license to Ericsson’s 4G patents.”  [Memo at 52].  From a broad perspective, Judge Selna found it was “not logical that two features could have a value in excess of Ericsson’s entire portfolio.” [Memo at 54]

Ericsson’s experts performed three main steps, presented below along with sub-steps:

  1. Identify specific technical contributions (or features) of the standard covered by Ericsson’s SEPs.
    1. Subdivided Ericsson’s SEPs into 10 technology sub-areas.
    2. For each technology sub-area, identify next best alternatives that would not infringe Ericsson’s SEPs.
    3. Identify benefits of Ericsson’s technology over the alternatives.
    4. Calculate value of these benefits, such as improved batter life, faster data speeds/throughput, fewer connection delays/less latency, better uplink peak-to-average ratios, increased spectral efficiency, and coverage improvements.
  2. Estimate the value for each technical contribution (or feature) in total (i.e., for all SEP or other contributions to the that feature) over the next best available non-infringing alternative.
    1. Chose two exemplar benefits — improved battery life and faster data speeds — to compute dollar value of as compared to the next best alternatives.
    2. Determine dollar value of benefit–e.g., assigned $15.90 value to standard’s sleep mode technology considering a 54% battery life improvement and a survey of U.S. smartphone users (“the IP&R Survey”); assigned $26.24 to $33.00 value to standard’s faster data speed technology and surveys by IP&R and Accenture.
  3. For each feature, apportion out Ericsson’s share of the value of that feature based on Ericsson’s SEPs covering the feature.
    1. Relied on a study by Signals Research Group of Ericsson’s technical contributions (documents submitted to the standards body during standard development which may or may not included patented technology) that were “approved” and included in the standard (“the Signal Approved Contribution Counting Study”).
    2. Relied on the percentage of Ericsson’s share of all SEPs on the standard, rather than Ericsson’s share of SEPs on the specifically identified technology sub-area.
    3. Apportioned Ericsson’s share of value of features based on the Signal Approved Contribution Study–e.g., $2.32 for improved battery life and $3.83 to $4.82 for faster data speed.

Judge Selna expressed concern that Ericsson’s experts had not determined Ericsson’s share of Sleep Mode patents or the number of Ericsson’s technical contributions that were directed to sleep mode. [Memo at 51].

Judge Selna found (in a later section of his decision) that the Signal Approved Contribution Count Study (See Step 3.I above) had two major flaws. [Memo at 75].  First, contributions include non-patented technology and there was no evidence correlating the contribution to intellectual property rights.  Second, the contribution counting did not account for transferred or expired patents.  Ericsson transferred ownership of about 200 U.S. patents in the past decade.  And failing to account for expired patents could lead to the royalty rate being based on expired patents. [Memo at 75].  Accordingly, Judge Selna ruled:

While contribution counting [i.e., counting of written technical contributions submitted to a standard setting body and included in the standard] may have its uses, it cannot be used to determine a FRAND rate for a patent portfolio, or unpack a cross-license [i.e., convert non-monetary consideration of a license into a dollar value to assess effective royalty as a total dollar amount]. [Memo at 75]

Judge Selna found the analysis flawed because it did not show that Ericsson’s SEPs were responsible for the percentage of value of the features he identified:

Ericsson is only entitled to 14.6% of the value [of] longer battery life or faster connections if it can show that it owns 14.6% of the patents that cover those inventions.  [Ericsson’s expert] did not attempt to show that Ericsson is responsible for 146% of the specific features he valued. [Memo at 53] [Note:  Judge Selna’s statement about “owns 14.6% of the patents” is unclear on whether it contemplates pure patent counting that treats all patents equally  (e.g., owns about 14 of 100 patents on the feature) or contemplates a substantive measure (e.g., owns one or more patents whose combined technical contributions are 14.6% of the value of the feature).]

Judge Selna found that Ericsson’s consumer surveys (see Step 2.II above) were not reliable for several reasons:

  • A survey focused on a single feature at a time, which focus can overstate the importance of that feature to actual purchase decisions.
  • A survey asking consumers about willingness to pay for an individual feature may be unreliable and susceptable to influences.
  • A survey focused on mobile network services, not necessarily handset feature.
  • Ericsson did not provide information on the survey methodology or specific questions asked. [Memo at 53-54]

Judge Selna thus found Ericsson’s Ex-Standard analysis unreliable and did  not rely on it for his royalty determination.

Non-Discrimination Analysis Based on Comparable Licenses

After determining a fair and reasonable royalty (i.e., the “FR” of FRAND) by his modified Top Down approach above, Judge Selna then looked at the non-discriminatory (or “ND”) part of the FRAND commitment by comparing Ericsson’s Option A and Option B offers to licenses that Ericsson gave to others. [Memo at 54]  Here, “[t]he parties agree that like, or close to, like rates must be offered to firms which are similarly situated.”  A key dispute was which licensees of Ericsson’s SEPs were “similarly situated” to TCL:

Licensees of Ericsson SEPs Similarly Situated to TCL
(Disputed companies underlined)
TCL’s Argument

Apple, Samsung, Huawei, LG, HTC and ZTE*

*Note: TCL considered ZTE similarly situated but dropped ZTE because its expert could not “unpack” the ZTE license. [Memo at 58 n.30]

Ericsson’s Argument

Karbonn, CoolPad, Huawei,* LG, HTC and ZTE

*Note: Huawei was not on Ericsson’s original list, but Ericsson apparently later agreed that Huawei was similarly situated to TCL. [See Memo at 58]

Judge Selna’s Finding TCL’s List : Apple, Samsung, Huawei, LG, HTC and ZTE

Licensees “Similarly Situated” To TCL.  Judge Selna concluded that he would consider as candidates of similarly situated firms “all firms reasonably well-established in the world market,” which would include companies with large market share as well as emerging companies. [Memo at 57]  He was concerned that too much picking and choosing of criteria for who is “similarly situated” could lead to such a narrow list that it would effectively read out non-discrimination requirement.

Judge Selna noted that all of the parties’ experts considered “firms using the same technology and at a similar level in the value chain.”  But they departed in other ways.  He was concerned that relying on only head-to-head competitors was too narrow.  But considering every company that uses the same technology is too broad–e.g., it “would impose the same rate on large global firms and local niche manufacturers.”

Judge Selna viewed “similarly situated” more broadly in the mobile phone market given the dynamic change of competitors in the past decade.  For example, half of the six companies with the largest share of the U.S. market–and Ericsson’s handset division itself–were “shuttered or divested.”  Although TCL entered the market only six years ago, TCL is now the fourth largest manufacturer in the U.S. market.

Judge Selna found the following factors to be relevant in determining what would be a similarly situated company:

  • geographic scope of the company
  • licenses the company required
  • a reasonable sales volume

In contrast, Judge Selna found that the following factors would not be relevant to a “similarly situated” company analysis:

  • the company’s overall financial success or risk
  • brand recognition
  • device operating system
  • existence of retail stores [Memo at 58]

Judge Selna found geographic scope to be the most important factor in this case.  He distinguished global companies from “local kings” (i.e., “a company that sells most or all of its devices in a single country, often the same country where it is headquartered and manufactures the devices.”). [Memo at 59].  Unlike a global company, local kings’ sales are largely in one country.  They also get a different license from Ericsson, because they only need a license in a single country; in contrast, a global company may get a blended rate given different patents and value thereof in different countries.

Judge Selna thus did not include Karbonn and Coolpad suggested by Ericsson, because they are local kings. [Memo at 59-60]

Judge Selna did include Apple and Samsung suggested by TCL as similarly situated to TCL given that they all had a global market, required licenses to 2G, 3G and 4G and were in the top ten of market share: Samsung 1st, Apple 2nd and TCL 7th, between agreed “similarly situated” LG and ZTE and above agreed “similarly situated” HTC that was not in the top 10.  Although Apple’s devices sold for a higher price than TCL’s devices, Ericsson agreed that the higher prices were due to the Apple brand and not the value added by Ericsson’s SEPs.  Judge Selna did not use other factors Ericsson argued to distinguish the companies (see the list above of irrelevant factors), finding that “the prohibition on discrimination would mean very little if the largest, most profitable firms could always be a category unto themselves simply because they were the largest and most profitable firms.” [Memo at 61]

“Unpacking” Ericsson’s Licenses.  Judge Selna considered how to “unpack” licenses that Ericsson entered with the companies he found “similarly situated” to TCL.

Unpacking a license involves evaluating all of its terms and other consideration so that the Court can calculate the effective one-way rate that each licensee pays Ericsson for its handset SEPs. [Memo at 62]

Judge Selna noted that the competing experts’ unpacking analysis reached “conclusions for each firm [that] largely agreed and were rarely widely disparate.”  He further described common terms used in the process:

  • Cross-License“, “Two-Way License” or “Reciprocal License” — “the licensee grants Ericsson the right to use its infrastructure SEPs in exchange for a smaller payment.”
  • Cash Payment” — Money payment in the form of a lump sum or running royalties.
  • Lump Sum” — “a fixed payment or series of fixed payments regardless of how many units the licensee sells.”
  • Running Royalty” — “the licensee pays a royalty for each qualifying unit, usually either as a percentage of the unit’s net selling price, or on a dollar-per-unit basis.”  Sometimes the running royalty is subject to a cap (i.e., maximum dollar amount per unit to pay) and floor (i.e., minimum dollar amount per unit to pay).
  • Net Balancing Payment” — Cash or other consideration given by the licensee to Ericsson in addition to giving Ericsson a cross-license to the licensee’s patents in order to account for the different value of the patent portfolio’s being cross-licensed.  Net Balancing Payment equals (Ericsson One-Way Royalty Rate times Licensee’s Revenues on Its Licensed Products) minus (Licensee’s One-Way Royalty Rate times Ericsson’s Revenues on Its Licensed Products).
  • Portfolio Strength Ratio” or “PSR” — When considering cross-licensing value, the ratio of Ericsson’s One-Way Royalty Rate to Licensee’s One-Way Royalty Rate.
  • Release Payment” — One-time payment for past unlicensed sales.

Judge Selna considered what the parties agreed to as the “unpacking formula”, which starts with the following basic mathematical premise:

If the license includes consideration beyond just a one-way royalty payment, then Judge Selna would compute an effective one-way royalty rate to capture that additional value.  For example, when an Ericsson license with a company includes a cross-license, Judge Selna uses the “Net Balancing Payment” formula and PSR (both defined above) to compute the effective royalty rate for that license agreement (i.e., essentially use value of cash royalty payments and value of licensee’s cross-license to Ericsson).  The formula requires determining four inputs to determine Ericsson’s effective One-Way Royalty Rate–i.e., Net Balancing Payment, Licensee’s Revenues on Licensed Products, Ericsson’s Revenue From Cross-Licensed Products and the Portfolio Strength Ratio (“PSR”), as shown in the formula below:

Judge Selna considered common issues that arose in applying the unpacking formula:

  1. Release Payment For Past Unlicensed Sales.  The Ericsson license agreements did not spell out any basis to allocate a lump sum payment between past and future sales.  Judge Selna reasoned that  using a lump sum payment to trigger the release was more a timing issue, rather than an agreement to pay different rates for past and future sale or an agreement that the lump sum was entirely for past sales. [Memo at 65]  Judge Selna would treat released sales as part of the overall forward looking licensing terms that, along with future sales, would be paid for as part of the overall transaction: Parties “generally care much more about the total amount they have to pay and the total value they receive, rather than whether a payment is labeled as a release from past liability or for the future license.” [Memo at 66].  Although it “is certainly possible that parties could specifically agree to different royalty rates for released and prospective sales … that is not the case for any of the licenses the Court unpacked.”
  2. Apportioning Lump Sum Payments Between Multiple Standards.  For the approach he was taking, a single lump-sum payment for a license covering multiple standards (e.g., 2G, 3G and 4G) must be unpacked to determine an effective royalty rate on a standard-by-standard basis.  In doing this, “the more assumptions the experts made, the more the license reflects the expert’s decisions rather than the parties’ agreed upon royalties rates.” [Memo at 67]  Further complications arise if the licenses involved non-essential patents or devices other than mobile phones, such as external modems or personal computers.  In this case, however, TCL did not attribute any value to the non-essential patents.  Further, both parties attributed the entire lump sum payments to just mobile phones, which attributes no payment for other devices; accordingly, Judge Selna “will treat TCL the same way” when setting royalty rates. [Memo at 67 n.37]  Judge Selna ultimately found some problems with Ericsson’s expert’s apportionmentof value between the 2G, 3G and 4G standards, so he “generally adopts” TCL’s expert’s apportioning of the Net Balancing Payments between those standards. [Memo at 68]
  3. Dollar-Per-Unit Rates, Caps and Floors.  Judge Selna declined to use a dollar-per-unit royalty or use royalty caps or floors–even though Ericsson had done so in some licenses–for four reasons.  First, dollar-per-unit is at odds with industry practices and Ericsson’s own past practice.  Second, “a percentage-based royalty better aligns the incentives of the SEP-holder and the licensee,” which “furthers ETSI’s express policy objectives of both rewarding SEP holders and making their intellectual property available to the public.  Third, Ericsson itself confirmed in several ways, including discovery interrogatory responses, that “royalties should be a percentage running royalty.”  Finally, the record does not support SEPs having “a fixed, determinable value which would justify a fixed dollar-per-unit rate or a percentage rate as modified by floors or caps.”  Further, the caps and floors were “solely the product of negotiations, not any sort of analysis of whether they are fair or reasonable.” [Memo at 69]  Importantly, Judge Selna did not find that such practices were problematic or could not be used by parties in managing or accounting for risks, but wanted his calculations in this litigation to be based on principles of general application:

To be sure, in the course of private negotiations, parties may enter into a variety of licensing schemes that reflect each party’s unique assessment of the risk of a particular arrangement.  However, the Court prefers to conduct its FRAND analysis on principles of general application which do not require the Court to discern the peculiarities of those risk assessments. [Memo at 69]

Judge Selna’s next step of applying the unpacking formula was resolving the parties’ disputes about the inputs to that formula–i.e., the PSR, the present value of the net balancing payments, the present value of Ericsson’s revenue on cross-licensed products and the present value of the licensee’s revenue on licensed products.

  1. Discounting Future Values Into Present Values.  The parties used different rates to discount future value in to present value.  Ericsson used a 10% to 12% discount rate, which it used in its internal business case memo made at the time of an agreement to indicate the risk associated with a licensee.  TCL’s expert used different rates depending on the type of payment or license: Treasury Bill rate for past sales, 10% to 12% for future revenue or payments; or prime rate for lump sum payments.  Judge Selna ultimately adopted the following discount rates to future values to obtain present values:
    Future Revenue Projections: 10% discount rate
    Future Fixed Payments: 5% discount rate (a lower rate because fixed payments are more certain and valuable than a royalty based on a percentage of the product selling price)
    Past Revenue: 0.45% (Treasury Bill Rate) (upward adjustment since selling of product before paying royalty is like “an interest-free loan from the SEP-holder”).
  2. Estimating Revenue.  The parties used two sources for projected revenue: (1) projections from Ericsson’s internal business case memo for each license and (2) third-party International Data Corporation (“IDC”) data from actually reported sales.  Judge Selna found the third-party IDC data “a valuable check on a party’s internal and unvalidated projections” because (1) both parties relied on it; (2) the revenue projections in Ericsson’s business case memos “dramatically underestimated” the actual revenue in the IDC data; (2-a) “IDC’s business model relies on providing accurate data”; (3) “IDC data reflects actual sales, not the projections of one party to the license.” [Memo at 72-73]  Judge Selna found it was not unfair to Ericsson that the IDC data showed better actual revenue from sales than Ericsson had projected (i.e., more licensed products were licensed under a lump sum than Ericsson projected), because (a) the FRAND obligation “does not incorporate an SEP-holder’s projections; it applies to the actual terms and conditions” and (b) benefitting from the certainty of a lump sum payment comes with the risk that actual sales will be greater than projected sales. [Memo at 73]
  3. The Appropriate Portfolio Strength Ratio (“PSR”).    TCL used PSRs derived from its patent counting study (generally the number of licensed Ericsson SEPs divided by the number of cross-licensed SEPs for particular licensee); Ericsson used PSRs from its contribution counting study (as discussed above, Judge Selna rejected the contribution study as being “a good proxy for the strength of [an] SEP portfolio[]”).  Judge Selda noted two (problematic) assumptions with either PSR calculation: (1) “an SEP portfolio’s strength is directly proportional to its size” and (2) “each patent or contribution is treated equally, regardless of individual value of the invention, or whether it is for a handset, infrastructure device, or both.”  He noted that “treating each patent equally was an express feature of the [top down] methodology advocated by Ericsson and others.”  But “it is less clear that assumption is valid in the context of negotiations between sophisticated parties,” which is the case for the comparable licenses being considered here. [Memo at 74]  He ultimately followed TCL’s PSR calculations given his rejection of Ericsson’s contribution study (discusussed above).

Results from Unpacking Ericsson’s Other Licenses.  Judge Selna’s analysis in how he unpacked Ericsson’s license with various companies is fairly redacted on critical values and we refer you to those sections for the details of his analysis:

  • Apple License (Memo at 76-79)
  • Samsung License (Memo at 79-82)
  • Huawei License (Memo at 82-85)
  • HTC License (Memo at 85-86)
  • ZTE License (Memo at 86-89).

Results from Unpacking Ericsson Option A and Option B Offers to TCL.  The two licensing options that Ericsson gave TCL included some terms that were not based purely on a percentage royalty of the sales price — e.g., a minimum per unit royalty fixed royalty would apply for low-cost products.  Judge Selna computed an unpacked rate as a straight one-way license without a discount, which he applied to TCL’s 2014 and 2015 actual sales data.  The table below shows his results for Option A and Option B [NOTE: Judge Selna apparently only unpacked Option A and Option B licensing terms into an effective running royalty for mobile phones; he simply restates the Option A and Option B terms for other devices (external modem and personal computer).]:

 ETSI Standard

Unpacked Option A

Unpacked Option B


Mobile Phone: 1.079% of net price (effective rate based on $30 million payment for first $3 billion sales)

External Modem: 1.5% of net price with $0.40 floor

Personal Computer: $0.50 per unit for GPRS, $0.75 per unit for EDGE

Mobile Phone: 0.8701% of net price (effective rate based on blending of 0.8% of net price for GSM/GPRS; 1.0% for EDGE)

External Modem: $0.75 per unit

Personal Computer: Same as Option A


Mobile Phone:1.0535% of net price

External Modem: 1.5% of net price with $0.40 floor

Personal Computer: $2.25 for single mode; $2.75 for multimode

Mobile Phone: 1.2% of net price

External Modem: $0.75 per unit

Personal Computer: Same as Option A


Mobile Phones: 1.0738% of net price

External Modem: $3 per unit if net price < $60; else $1 per unit.

Personal Computer: $3.5 per unit

Mobile Phone: 1.9878% of net price (effective rate for actual sales accounting for $2.00 floor and $4.50 cap)

External Modem: 1.5% of net price with $2 per unit floor.

Personal Computer: Same as Option A


Competitive Harm.  Ericsson argued that the relevant harm from discrimination would be harm to competition (which is actionable under competition law), not merely harm to a competitor (which is not actionable under competition law). Judge Selna noted U.S. Sherman Act jurisprudence provides no guidance on ETSI’s non-discrimination part of FRAND.    Judge Selna “finds that harm to the competitor firm offered discriminatory rates is sufficient.”  [Memo at 91 (emphasis added)]. [NOTE: This wording  may be disputed in this and other cases.  Judge Selna does not cite any case supporting this statement, does not express an analysis for this statement and qualifies his statement as “finds”, which in legal writing usually signals a fact determination based on the evidence of a specific case.  The “legal” or “factual” nature of Judge Selna’s statement here may be significant in determining whether this statement was a precedential legal ruling to be addressed by other parties in other cases or a factual finding specific to these parties in this case.]

Ericsson’s Offers to TCL Were Discriminatory.  Judge Selna compares the unpacked Option A and Option B offers to TCL with the unpacked rates he computed above for the licensees of Ericsson’s patents deemed “similarly situated” to TCL.

Importantly, Judge Selna used the Apple and Huawei licenses only as a “informative” reasonableness checks, but not direct comparision, because both of those licenses came into effect after Ericsson’s Option A and Option B offers were made to TCL. [Memo at 91].  One key reason is that ETSI rejected a “most favored licensee” provision as part of FRAND, which otherwise would have required accounting for licenses entered after the instant license:

[T]he concept of “most favored nation,” or here “most favored licensee,” was never part of the ETSI FRAND equation, and in fact was rejected. [Memo at 91]

Judge Selna found that the results of his analysis and the results across company were reasonably in the same ball park, as shown in his charts below (where the court redacted the names of licensees and we substituted the names of experts with the names of their sponsoring parties “TCL” or “Ericsson”):

Judge Selna then compared his unpacked rates for other Ericsson licenses with Ericsson’s Option A and Option B offers to TCL:

Judge Selna also stated that discrimination is not measured solely on the date it was made “because license agreements last for multiple years.”  He did a comparison of Option A and Option B offers with rates paid by other licensees over a period of time, but that table is too redacted to be useful.  [NOTE:  Judge Selna’s charts above are based in part on calculations that used actual sales data, rather than what the parties contemplated and projected at the time they entered licensing agreements.  This may lead to disputes as to whether what the patent owner believed were non-discriminatory offers at the time a license was entered based on  projected sales would become discriminatory later if actual sales did not meet those projections.  That may  not have been Judge Selna’s intent given, among other things, his statement earlier that ETSI does not have a most-favored nations requirement that would require re-visiting past licenses if actual sales differed from projected sales.  Further, he acknowledges that his calculations were not perfect and should be compared more as a matter of degree, than of surgical precision.]

Based on the foregoing and what he found to be “radically divergent rates” between what Ericsson offered TCL and others, Judge Selna found that TCL showed that the terms were discriminatory and, thus, were not FRAND:

The Court readily acknowledges that these unpackings are not perfect.  However, by any measure, Option A and Option B [offers from Ericsson to TCL] are radically divergent from the rates which Ericsson agreed to accept from licensees similarly situated to TCL.  TCL has carried its burden and demonstrated that Option A and Option B are discriminatory and do not meet FRAND terms. [Memo at 94]

Setting a Prospective FRAND RATE for the Ericsson – TCL License

Judge Selna sought to determine the prospective FRAND royalty rate, as a percentage of the device selling price, based on the results from his Top Down Analysis (that produced U.S. patent rates) and the results from his comparable license analysis (that produced global rates).  [Memo at 95].  He first converted his comparable license rates from a global rate to a U.S. patent rate so that they could be compared with the U.S. rate from his Top Down analysis.  He derived a formula for converting the global rate to a U.S. rate based on the concept that the total global royalties paid equals the sum of royalties paid in each region.  Recall that he had separated the global license into three regions: U.S., Europe and ROW (rest of the world).  Recall also that he determined the value share of Ericsson’s SEPs in Europe and ROW relative to Ericsson’s U.S. SEPs, as shown in our table above labeled “Value Share Relative to Value of U.S. SEPs”.  Accordingly, he would take the sales in a region as shown (pulled from the IDC data) and multiply that times the value share relative value for that region to assess a comparable U.S. royalty for that region.  For example, multiply ROW Revenues (from IDC data) times 69.8% (the ROW value share relative to U.S. SEPs for 4G) to derive a comparable U.S. royalty payment for ROW sales. [Memo 95-98]

Judge Selna then compared these results with the results he calculated from the Top Down approach.  He dropped the two lowest and two highest royalties from those combined results to yield a chart of potential U.S. royalties for Ericsson’s U.S. SEPs, as shown in the court’s chart below for 4G rates (with redline added by us showing Judge Selna’s ultimate adoption of 0.45% as a FRAND Rate for Ericssons U.S. SEPs):

Although he could not determine a FRAND royalty “with exactitude”, Juge Selna found that, “with abundant and largely congruent data before the Court, the Court finds that 0.45% is an appropriate FRAND for Ericsson’s 4G SEP portfolio in the United States.” [Memo at 99].  Judge Selna then used the “Value Share Relative to Value of U.S. SEPs” to compute a corresponding ROW rate of 0.314%  (i.e., 0.45% U.S. 4G Rate times 69.8% 4G ROW Value Share).

Applying a similar analysis (with exceptions here and there) to other 2G and 3G standards, Judge Selna determined the final prospective (i.e., future sales) FRAND rates for Ericsson’s SEPs as shown in the table below [Memo at 104]:

Release Payment for TCL’s Past Unlicensed Sales

Having determined a prospective royalty above for TCL’s future sales, Judge Selna next determined TCL’s release payment to Ericsson for TCL’s past sales of unlicensed products.  Ericsson had the burden to prove entitlement to the release payment and the royalty rate to apply.  Ericsson argued that Judge Selna should use the prospective royalty rate applied to future sales (which Judge Selna determined above).  TCL argued, among other things, that some past sales should be time-barred (presumably based on the U.S. patent statute limit of damages to infringing activities no more than six years prior to the filing of a patent infringement suit).  Judge Selna adopted Ericsson’s proposal to use the prospective royalty rate and found that none of TCL’s past sales were time-barred.  Judge Selna also used the 0.56% discount rate for past unlicensed sales that he used in the comparable license analysis above “to reflect the fact that TCL received the benefit of Ericsson’s SEPs well before it must pay for them.”  He then concluded that “TCL must pay Ericsson $16,449,071 as a release payment for unlicensed sales from 2007-2015.” [Memo at 105-106]

Judge Selna did note a few issues in his calculation.  For example, a multi-mode device with 3G/2G technology should have a reduced 2G rate “because [the multi-mode devices]only use 2G functionality when they cannot connect to a 3G network.”

Judge Selna’s Conclusions of Law

Following his factually intensive analysis discussed above, Judge Selna provided a more succinct statement of his conclusions of law.  We summarize some of his finding below to the extent that we may not already cover have covered them above.

NonDiscrimination.  No U.S. cases have definitively addressed non-discrimination and only one expert in the case opined on what non-discrimination means in the  ETSI FRAND commitment.  Judge Selna did conclude that rates can differ between licensees and that “TCL cannot claim that anything other than the nominally lowest rate in marketplace is per se discriminatory,” stating:

The Court concludes there is no single rate that is necessarily FRAND, and different rates offered to different licensees may well be FRAND given the economics of the specific license. [Memo at 109]

In a later portion of his decision, Judge Selna ruled that “Ericsson’s use of floors in its rates is itself discriminatory” based on the evidence (or lack thereof) presented in the case:

Ericsson’s use of floors in its rates is itself discriminatory.  In the absence of a credible showing that Ericsson’s SEPs add  a measurable incremental value, there is no basis for essentially discriminating on the basis of the average selling price where a floor would result in a higher effective rate for lower priced phones. [Memo at 113]

Comparable Licenses Have Role.  Judge Selna ruled that other licenses are a proper, but not exclusive, measure of whether an offer is FRAND.  He rejected “TCL’s seeming blanket rejection of comparable licenses.” [Memo at 109]  Rather, “actual licenses reflect the economic value of the patented technology in the market place.”  Considering an array of licenses can “substantially diminish” concerns about any particular license:

The Court finds that by looking at an array of licenses, concerns about FRAND compliance of any particular license, asymmetric information, and litigation pressures are substantially diminished. [Memo at 110]

Georgia-Pacific Methodology.  Ericsson’s comparative license approach overlapped with Georgia-Pacific, but “the Court did not find useful a full-blown Georgia-Pacific analysis in the unique context of a FRAND dispute.” [Memo at 110]

FRAND Obligation.  Judge Selna found that “Ericsson negotiated in good faith and did not commit a breach of contract by virtue of its conduct.” [Memo at 111].  The parties disagreed whether Ericsson’s FRAND commitment required Ericsson “to offer rates which are in fact FRAND.”  Judge Selna decided not to resolve that issue, because (1) there would be no damages if this was a breach given his prior summary judgment ruling (discussed above) and (2) finding a breach as a basis for awarding specific performance (i.e., an injunction with adjudicated licensing terms) would be unnecessary and superfluous since the court could order specific performance as part of the parties’ declaratory judgment claims. [Memo at 112]

Injunction Implementing Adjudicated License

The parties jointly submitted, and Judge Selna adopted, a Final Judgment And Injunction to implement his decision discussed above.  The Final Judgment and Injunction involves a five-year license period and has many common license terms and otherwise reads like a license agreement, though some Appendix are not public and  may contain significant portions of the licensing agreement.  Ericsson has appealed the final judgment and injunction to the Federal Circuit.