Friday, May 19, 2017

IPO Chat Channel's Global FRAND Update (Europe)

The IPO IP Chat Channel will be presenting a webinar titled Global FRAND Update:  Europe next Wednesday, May 24, at 2 pm Eastern time.  Here is a link, for those who might be interested in registering, and here is a description:
This webinar will consider the impact of recent important judicial decisions and institutional initiatives in Europe concerning SEPs. The past decade has demonstrated that traditional patent jurisprudence is ill-suited for disputes involving standard-essential patents (SEPs) and the promise to license them at rates that are fair, reasonable, and non-discriminatory (FRAND). Yet the technological future, including the Internet of Things and 5G mobile networks, can only be built using standards and patent licenses. Courts and governments worldwide are straining to adapt.
Our panel includes a senior competition lawyer at both an owner of SEPs and an implementer, as well as a leading academic authority on standards. They will discuss: 
Unwired Planet v. Huawei, the recent decision of the U.K. High Court that determined the value of a worldwide FRAND portfolio license to Unwired Planet's patent portfolio, and decreed that Huawei must license it at that rate or face an injunction; 
The outcome of numerous SEP cases that have followed Huawei v. ZTE, the 2015 ruling from the European Court of Justice that spelled out conditions under which a SEP holder can ask for an injunction without infringing competition law; 
The April announcement by the European Union that it is preparing a "roadmap" to provide a framework for licensors and licensees of SEPs, and an early leak of its contents.
Panelists will include Professor Jorge Contreras, James Harlan (InterDigital), and Gil Ohana (Cisco).

Wednesday, May 17, 2017

Paris Court of Appeals Holds That Patentees Are Not Entitled to Disgorgement of Profits

Since France enacted legislation implementing the 2004 EC Enforcement Directive, French courts have been authorized to take into account the benefits to the infringer when awarding damages.  More specifically, the version of the relevant statute, article 615-7 of the Intellectual Property Code, in effect from October 29, 2007 to March 14, 2014, stated that  
For assessing damages and interest, the court takes into account the negative economic consequences, including loss of profit, suffered by the injured party, the profits realized by the infringer and the moral prejudice caused to the rightholder by the infringement.
However, the court may, alternatively, upon request by the injured party, award damages as a lump sum that shall not be less than the amount of royalties or fees that would have been due if the infringer had requested authorization for the use of the right infringed. 
(Translation courtesy of Margaret Wade and me.)  The version in effect since March 14, 2014, states (in my translation) that: 
For assessing damages and interest, the court takes into account distinctly:
the negative economic consequences, including loss of profit and the loss sustained by the injured party;
the moral prejudice incurred by the latter;
and the profits realized by the infringer, including the savings of intellectual, material, and promotional investments which the latter has derived from the the infringement. 
However, the court may, alternatively, upon request by the injured party, award damages as a lump sum.  This amount is higher than the royalties or fees that would have been due if the infringer had requested authorization for the use of the right infringed.  This amount is not exclusive of the compensation for the moral prejudice suffered by the injured party.
Whether either of these provisions permit French courts to award the prevailing plaintiff the infringer's profits from the infringement, or instead only to take those profits into account in crafting an award of compensatory damages, is a matter of some debate, though from what I've read the latter view seems to be dominant for now.  For previous discussion on this blog, see, e.g., here, here, and here.

Anyway, a recent judgment of the Cour d'appel de Paris, Carrera SARL et Texas de France SAS v. Muller et Cie, PIBD No. 1067, III, 170 (Dec. 9, 2016), appears to take the latter view, namely that IP owners are entitled to recover damages for their own losses but not an award of the infringer's profits as such.  If I'm reading the facts correctly, Muller (owner of European Patent 1,067,822 for a "Heating element manufacturing process for heating or cooking apparatus, such heating element and apparatus incorporating it") is a patent holding company that had licensed the patent, royalty-free, to six affiliated companies.  Muller prevailed on liability, but sought to recover an award of the defendants' profits rather than a reasonable royalty.  The district court awarded damages in the amounts of  €327,733 and €280,130, respectively, as well as damages for moral prejudice in the amount of €100,000.  Both parties appealed.  The Court of Appeals, however, rejected Muller's argument that  it could opt for an award of profits, stating (my translation):
Moreover, while it is true that the Law of October 29, 2007, implementing Directive 2004/48, invites the judge to take into account "the benefits realized by the infringer," it does not  authorize the confiscation of those benefits; and this taking into account is limited only to the portion relating to the losses suffered as a result of the exploitation, in order to attain a complete reparation for the loss." 
Furthermore, if I'm reading this correctly, since Muller disavowed any reliance on a reasonable royalties theory, the court also threw out that award; and it also disallowed the award for moral prejudice, concluding that there was no evidence that the infringement caused any harm to Muller's reputation.

I thank Pierre Véron for a productive discussion of this case, prior to my publishing this post.

Monday, May 15, 2017

Two More Upcoming Events on FRAND, SEPs, and Remedies

1.  As noted on Patently-O recently, the Université de Liège's Competition and Innovation Institute will be putting on a Conference on Innovation, Research and Competition in Brussels May 29-30, under the direction of Professor Nicolas Petit.  Here is a link to the program, which will focus on issues relating to standard-essential patents.  Speakers will include, among others, German Federal Supreme Court Judge Klaus Grabinski and economists Joseph Farrell and Stephen Haber (who have very different views on patent holdup).  Looks fascinating--wish I could attend.

2.  The Centre of Innovation, Intellectual Property and Competition (CIIPC) at National Law University, Delhi, is putting on the Second Annual Roundtable on Innovation, Intellectual Property and Competition Law in Bengaluru (Bangalore) India, on July 1-2.  Here is a link.  The program is not up yet, but I understand that the topics will include IP and competition law, FRAND, and remedies, among other things, and that speakers will include Arvind Subramanian (Chief Economic Advisor to the Government of India)  and Antony Taubman, the Director of the WTO's Intellectual Property Division.  This one also sounds great.

Thursday, May 11, 2017

PwC 2017 Patent Litigation Study Is Now Out

Every year around this time I look forward to reading PricewaterhouseCoopers' annual study of U.S. patent litigation.   The firm's 2017 Patent Litigation Study:  Change on the Horizon? is now out and available here.  The study's methodology, including its adjustment of median damages to 2016 dollars, is set out at p.29.  Here are some highlights: 

1.  Patent infringement litigation declined to 5,100 filings in 2016, 9% fewer than in 2015 (p.4).  (I'd note that Lex Machina's 2016 Patent Litigation Year in Review, for which if you're interested you should contact Lex Machina, reports 4,537 filings in 2016, down 15% from 2016.  I'm guessing the difference may be due at least in part to different cutoff dates.  Lex Machina reports filings through December 31, 2016, while PwC uses a September year-end (see p.4).  More generally, I'd recommend that interested readers get a copy of the Lex Machina report for purposes of comparison; see also my paper with John Golden, which at pp. 14-15 and accompanying endnotes discusses some of the differences between the two firms' analyses and methodologies relating to damages, for previous years.)

2.  Notwithstanding a record damages award of $2.5 billion in Idenix v. Gilead (noted previously here), median damages declined from $10.2 million in 2015 to $6.1 million in 2016 (p.5).  Between 1997 and 2016, however, median damages have "ranged from $2.0 million to $17.0 million, with an overall median award of $5.8 million over the last 20 years" (p.9).  "Excluding damages awarded before trial (i.e., summary judgment and default judgment), the overall median award over the last 20 years jumps to $8.0 million" (id.).  Jury awards outpace bench awards $9.5 million to $0.6 million from 2012-16, a factor of about 15 (p.10).  For the years 2007-16, reasonable royalties account for 61% of the awards to practicing entities, lost profits 21%, and mixed awards make up 19% (p.11).  

3.  Excluding ANDA-related cases, 80% of U.S. patent trials are now conducted before juries (p.6). 

4.   Patent owners' overall success rates from 1997-2016 stand at 36% for practicing entities and 25% for NPEs.  Among cases that go to trial, however, success rates are 77% and 70%, respectively.  Overall, success rates at trial before juries are 74%, before judges 52% (pp. 14-15).  Though NPEs are less likely to win overall, for those that do win the median damages awards from 2012-16 were $15.7 million, compared to $4.1 for practicing entities (p.16).  From 1997-2016, median damages awards for university or other nonprofit NPEs were $16.3 million, for "company" NPEs $13.0 million, and for individual inventors $6.7 million.  Success rates for universities and nonprofit NPEs were also the highest (52%), followed by 28% for company NPEs and 18% for individuals (pp. 16-17).    

5.   Median time to trial has crept up to about 2.5 years, with stays pending PTAB proceedings probably responsible for some of the delay (p.7).

6.  The study also presents data on the distribution of cases, damages awards, and success rates by industry (pp. 18-20), and ranks district courts according to a number of metrics at pp.22-24.  The report concludes with a section on appeals, reporting that 75% of decisions are appealed and over half of all appeals result in a reversal at least in part (pp. 25-27).

RIP William Baumol

I learned this morning that the economist William Baumol died last week, at the age of 95.  Here are his obituaries in the Washington Post and the New York Times.  As the obits note, Professor Baumol was one of the leading economists of his generation.  I mention him here in part because of his influential paper (coauthored with Daniel Swanson) Reasonable and Nondiscriminatory (RAND) Royalties, Standards Selection, and Control of Market Power, 73 Antitrust L.J. 1 (2005).  Although Norman Siebrasse and I have criticized some aspects of the Swanson-Baumol paper in our article The Value of the Standard, 101 Minn. L. Rev. 1159 (2017) (see here), the paper remains an important one  that anyone involved in this field needs to be familiar with.  In addition, as the obits point out, Professor Baumol was also well-known for his theory, nicknamed "Baumol's Disease," of why improvements in productivity do not affect the cost of labor-intensive services such as child care and live musical performances, the relative price of which therefore tends to increase over time.  He will be missed.

Wednesday, May 10, 2017

The Comparative Patent Remedies Blog Is Now Four Years Old

I should have mentioned this over the weekend, but the blog is now four years old.  I launched it on May 6, 2013, and it's still going strong at the four-year mark.  I'm pleased to say that the blog has attracted readers from all over the world, and that (to date) it's had over 400,000 page views.  Thank you to my readers, and I hope you continue to enjoy the blog.

Monday, May 8, 2017

Recovery of Attorneys' Fees and Costs in Canada

Aside from my February 13 post on Apotex Inc. v. ADIR, 2017 FCA 23 (Feb. 2, 2017), and a very brief mention one month later of Airbus Helicopters S.A.S. v. Bell Helicopter Texteron Canada Limité, I haven't been reporting as promptly as I should on developments in the law of patent remedies in Canada.  I hope over the next few weeks to catch up--in addition to Airbus, there is another recent opinion, Dow Chemical Co. v. Nova Chems. Corp., that addresses several important damages issues--but I'll start today with a discussion of a different case from this past January, involving the recovery of costs.  (Costs may include attorneys' fees as well as other disbursements, such as expert witness fees).  

As I discuss in my book (p.209 n.147), like many countries Canada routinely awards attorneys' fees to the prevailing party, in patent and other litigation--though the common wisdom is that the fees awarded are typically only a fraction of the amount actually incurred.  (As I understand it, the thinking is that fee awards aren't intended to be fully compensatory, but rather to strike a balance between compensating the prevailing party and not unduly discouraging people from attempting to vindicate what they believe to be their rights.)  Things can be a bit more complicated in some instances, however, because Rule 400(1) of Canada's Federal Courts Rules states that "The Court shall have full discretionary power over the amount and allocation of costs and the determination of by whom they are to be paid," and then in Rule 400(3) sets forth various factors that may be taken into account in exercising this discretion.  Among these are "any conduct of a party that tended to shorten or unnecessarily lengthen the duration of the proceeding," and "whether any step in the proceeding was . . . improper, vexatious or unnecessary, or . . .taken through negligence, mistake or excessive caution . . . ."  Rule 400(4) further directs that "[t]he Court may fix all or part of any costs by reference to Tariff B"--a table of rates for various services, available here, and Rule 407 states that "Unless the Court orders otherwise, party-and-party costs shall be assessed in accordance with column III of the table to Tariff B."  However, Rule 406 also permits the court to exercise its discretion to award costs against a successful party, and also to award costs on a "solicitor-and-client" basis, which I understand means, essentially, at a higher (possibly full indemnification) rate.   

Anyway, in Mediatube Corp. v. Bell Canada, 2017 FC 6 (Jan. 4, 2017), the Federal Court concluded that the patent in suit was valid but not infringed.  Both parties nonetheless sought to recover costs.  First, the plaintiff sought to recover (1) costs incurred as a result of the defendant's corrections of its initial disclosures, which the plaintiffs alleged led to delay in resolving the dispute; (2) "aggravated costs" due to the defendants' calling the plaintiffs "patent trolls," which in the plaintiffs' view was both inaccurate and an accusation of fraud or dishonesty; and (3) costs arising from the defendants' citation of 735 prior art references in their statement of defense and counterclaim of invalidity (and thereafter identifying, three months before trial was set to begin, only 62 references that they would rely upon at trial).  The court rejected these arguments, the first on the grounds that (1) the defendants acted in good faith, (2) the timing wouldn't have made a "difference between proceeding with the trial and reaching a settlement," and (3) the noninfringement argument was "so compelling," given "the absence of four distinct essential elements of claim 1 from Bell's systems," that the judge was "not prepared to accept that the plaintiffs had a reasonable belief that they had a good arguable case before receiving the Corrected Information" (pars. 232-34).  As for the "patent troll" accusation, the court concluded that the description was more a matter of opinion and in any event not tantamount to an accusation of fraud or dishonesty (para. 240).  As for the prior art disclosures, the court was "not convinced that the plaintiffs were put to unwarranted expense" (para. 245).

Second, the defendants sought elevated costs and the imposition of costs on a solicitor-and-client basis, principally on the ground that the infringement action lacked merit.  The court agreed, in part.  First, it concluded that the defendants were entitled to costs for claims made against defendant Bell Aliant, some of which were withdrawn pretrial, others during trial.  With regard to these expenses, the court did not impose solicitor-and-client costs, but it accepted that costs should be elevated in the amount of 50% for the maintenance of the claims that were withdrawn during trial, which the plaintiffs pressed "in the face of information that Bell Aliant was not infringing" (para. 253).  Similarly, the court awarded elevated costs of 50% for claims that the plaintiffs made against Bell Canada, stating that "the contortions of the claims in issue that would be required in order to find infringement indicate that the plaintiffs should have known that Bell Canada did not infringe" (para.256).  Finally, the court stated that it would award solicitor-and-client costs relating to the plaintiffs' allegation that defendants had misappropriated confidential information and therefore could be liable for punitive damages, on the ground that unfounded accusations of fraud or dishonesty are an appropriate basis for the imposition of fees on this basis (paras. 254-67).  For further discussion of this case, see Norman Siebrasse's write-up on Sufficient Description.

Without speculating whether such an award would be merited in a case like Mediatube, I've sometimes wondered whether courts should have the authority to award attorneys' fees multipliers, in cases in which an action or a defense lacked merit and an award of attorneys' fees alone might not suffice to deter such conduct in the future (perhaps because of the disparity in resources between the parties).  Ultimately I'm not sure that's a good idea, but the issue did arise, in a sense, in a Federal Circuit case I blogged about last year (see here).  Of course, the award of elevated costs in Mediatube probably won't actually consist of the actual fees multiplied by 150%, since as noted above fees in Canada (other than, perhaps, the solicitor-and-client basis costs which the court did not elevate) are usually awarded pursuant to Tariff B and in general are believed to be well below the amount actually incurred.  Still and all, an interesting issue.

For further discussion of awards of attorneys' fees in Canadian patent cases, see Mark Biernacki & Kevin Unrau, Federal Court Awards Legal Costs of $6.5 Million in Patent Infringement Case (available here, and noting a trend toward higher fee awards in Canada, above the default Tariff B column III); Federal Court of Appeal and Federal Rules Committee, Review of the Rules on Costs Discussion Paper (Oct. 5, 2015) (available here, discussing among other matters the potential inadequacy of Tariff B).  I thank Norman Siebrasse for calling these papers to my attention; see also Norman V. Siebrasse, $2.9 Million Award Is 30% of Actual Legal Fees, Sufficient Description, Feb. 5, 2016 (available here). 

Friday, May 5, 2017

More Upcoming Events on Patent Remedies

1.  The Intellectual Property Owner's Association (IPO) is holding a Damages and Injunctions Committee Conference in Chicago on June 7, 2017.  Here is a link to the website, and here is a description of the program:
A one-day program, organized by IPO’s Damages & Injunctions Committee. Panel discussions will address strategies and procedural considerations for litigating damages/remedies issues; economic tools useful in apportioning for value of the IP at issue; recent developments in damages/remedies law; and the continuing evolution of patent damages law associated with apportionment for the footprint of the patent.
It looks like there will be lots of interesting speakers, including (among others) Judge Cathy Ann Bencivengo; attorneys Morgan Chu and Ed Reines; economists John Jarosz and Alan Cox; and Professors Sarah Burstein, Jonathan Masur, and Jeffrey Prince.  I'd love to attend myself if I could, but I'll be enjoying the midnight sun in Finland.

2.  The Sedona Conference will be hosting a webinar titled "Early Consideration of Patent Damages" on June 13, 2017, from 1-2:30 p.m. Eastern Time.  Here is a link to the website, and here is a description:
Damages issues have received substantial attention from the Federal Circuit over the past several years, as the Court has attempted to provide guidance to litigants and the district courts in the face of public concerns about eight- and nine-figure patent damages awards. This attention has led has led to the rapid evolution of a complex area of law, and - perhaps counterintuitively - unpredictability about the viability of damages theories. To assist litigants and the courts, the Sedona Conference's Working Group 9 on Patent Damages and Remedies (WG9) developed a Proposed Model Local Rule for Damages Contentions, which provides for an early exchange of damages-related information to enable the parties to develop and disclose their damages theories earlier than might otherwise occur, and thus to allow for orderly Daubert and pretrial processes.
In furtherance of this effort, WG9 has also developed a framework for early damages-focused hearings (to be published shortly), to allow for a pre-Daubert discussion of damages theories with the court in order to clarify and narrow and damages-related disputes, and to simplify Daubert and pretrial proceedings. These efforts of WG9 have garnered significant attention from the patent community. Indeed, the Northern District of California has adopted a rule reflecting a substantial portion of WG9's Proposed Model Rule on Damages Contentions, and other courts have considered The Sedona Conference's Principles and Best Practice recommendations in individual patent litigation matters as well.
This webinar broadens the discussion so that litigants and courts around the country may gain the exposure to these proposals and consider how they can be incorporated into appropriate cases to bring efficiencies to patent litigation.
This is the first in a series of monthly/bi-monthly webinars that WG9/WG10 (Patent Litigation Best Practices) will be hosting the rest of 2017, in conjunction with no less than nine forthcoming publications each providing Principles and Best Practice consensus, non-partisan recommendations, representative of all stakeholders in the patent system in true The Sedona Conference fashion.
The webinar is scheduled for 90 minutes, during which time you may text your questions to the panel, who will endeavor to address all that time allows.
Moderated by:
Andrea Weiss Jeffries,
WilmerHale, Los Angeles, CA

Hon. Cathy Ann Bencivengo, U.S. District Judge, S.D. Cal., San Diego, CA
Melissa Finocchio, Intellectual Ventures, Bellevue, WA
John C. Jarosz, Analysis Group, Inc., Washington, DC
Hon. John Love, U.S. Magistrate Judge, E.D. Tex., Tyler, TX

Wednesday, May 3, 2017

Is a FRAND Royalty the Same as a Reasonable Royalty?

My opinion, as I've stated before, is yes:  that courts, when called upon to award or determine the amount of a FRAND royalty (as in the Microsoft and Innovatio cases in the U.S., the Apple v. Samsung  case in Japan, or the recent Unwired Planet case in England) they should employ the same standards they would use when awarding a "reasonable" royalty as damages for patent infringement.  (Of course, FRAND cases are much more complex than other, more run-of-the-mill cases, and therefore might require more innovative approaches to coming up with a number, as we've seen in these cases.  All I mean is that, as Mr. Justice Birss states in Unwired Planet, is that ". . . arriving at a FRAND royalty rate is not different conceptually from assessing what a reasonable royalty would be in a patent damages enquiry albeit the particular factors applicable in setting a FRAND royalty for a licence to be FRAND and their application may differ from assessing damages" (para. 169).  This is consistent with the view expressed by Professors Contreras and Gilbert as well in their article Jorge L. Contreras & Richard J. Gilbert, A Unified Framework for RAND and Other Reasonable Royalties, 30 Berkeley Tech. L.J. 1451, 1487 (2015), and Norman Siebrasse and I express agreement with it on a policy basis in our paper The Value of the Standard, 101 Minn. L. Rev. 1159, 1204 (2017).  On this view, the principal significance of a FRAND commitment is that the patent owner who violates it may be liable for breach of contract or promissory estoppel or under some other body of law, and would have a weaker argument than might otherwise be the case in support of obtaining injunctive relief.  By contrast, the alternative view, that a FRAND royalty is less than a reasonable royalty, would seem to me to require courts to figure out exactly how much less it should be and why, and I don't see any principled basis for making such a determination.  (Of course, some SSOs require members to license their patents royalty-free, but that's not what we're talking about here.)

Nevertheless, I wonder whether other courts will necessarily adopt the view that a FRAND royalty and a reasonable royalty are the same thing.  For example, as Professor Peter Picht has pointed out to me, last year in Unwired Planet v. Huawei, Case No. 4b O 120/14 (19 Jan. 2016), the Düsseldorf District Court stated that, although a SEP owner is entitled only to a FRAND royalty for the period of time following its refusal to make a FRAND offer, it could recover its full damages for any period prior to that.  Here is my translation of the relevant language (paras. 413, 414,a nd 416) from the opinion:
. . . The refusal of the patent owner, to accede to the lawful demand of the patent infringer to conclude a license contract, can be a competition law violation and can give rise to the infringer's own damages claims against the patent owner.  Subsequent to its unlawful refusal, the patent owner cannot demand its entire damages; rather, the amount thereafter is limited to the value of a reasonable royalty.
. . . Moreover, the FRAND declaration does not have the effect of constraining the amount of the damages at the outset to the amount of a FRAND royalty.  Otherwise, this would mean that the effect of a FRAND declaration would be to confer upon any market participant a claim for the conclusion of a license on FRAND terms.  The court does not accept this reasoning.  Moreover, a claim for the full amount of any subsisting damages can be limited only by a counterclaim on the part of the infringer.  The conditions for such a counterclaim must be presented and proven by the infringer. . . .
Further . . . the infirnger of a SEP, like any other patent infringer, is obligated to inform itself prior to use of any existing patents and to seek a license.  If it does not do so, it must then pay the (full) damages claim.  Only if the patent owner refuses to grant a license on FRAND terms is it in violation of TFEU article 102, and the infringer is thereafter obligated only in the amount of a reasonable royalty.
I have to admit, I found this language a bit confusing the first time through.  The court seems to be saying that the patent owner is entitled to its full damages for the period of time preceding its refusal to grant a FRAND license, but only a reasonable royalty thereafter.  (I'm fairly certain that my translation is reasonably accurate.)  But then if a FRAND and a reasonable royalty are synonymous, what are the "full damages" the court is distinguishing them from?  Maybe the court has in mind a situation in which the patent owner would be entitled to lost profits or the disgorgement of the defendant's profits for the period of time preceding the owner's refusal; that might make sense, if the facts were such that an award of lost profits otherwise would be available, or if one believes that disgorgement can be an appropriate remedy in this sort of case (see post here).  Still, some further clarification would be helpful.

On the other hand, maybe there are some ways in which a FRAND and a reasonable royalty could be different in practice.  First, if a court-awarded reasonable royalty is supposed to reflect the ex ante bargain the parties would have struck knowing the patent to be valid and infringed, a court-ordered reasonable royalty might exceed what an ex ante FRAND royalty (which would have reflected a discount for the probability of noninfringement and invalidity) would have been.  This could lead to an anomaly, whereby the failure to make a FRAND offer initially might mean that, under Huawei v. ZTE, the patent owner loses the right to an injunction, but then is awarded a royalty that exceeds the amount of the (excessive) ex ante FRAND royalty.  Second, maybe a FRAND royalty could exceed a reasonable royalty if the former incorporates a meaningful "nondiscrimination" component and the latter does not.  Third, as in Unwired Planet v. Huawei, a FRAND license might be for a (global) portfolio of patents, whereas a reasonable royalty in a normal patent infringement case would typically just be for the national component of one or a handful of patents.  But this wouldn't necessarily suggest any difference in the basic methodology of calculation, I don't think.

I would welcome further thoughts on these issues.

Monday, May 1, 2017

Nichia v. Everlight: Does the Patentee Have to Prove All Four eBay Factors to Merit Injunctive Relief?

That seems to be what the Federal Circuit is saying in Nichia Corp. v. Everlight Ams., Inc., a precedential opinion authored by Judge Stoll that issued on Friday of last week.  Nichia, the plaintiff, "is the world’s largest supplier of LEDs" (p.2), and it owns three patents relating to "package designs and methods of manufacturing LED devices" (p.3).  The defendant, Everlight, "buys chips from suppliers and packages them into LEDs," which it sells "directly to customers and through its subsidiaries" (p.2)  The district court found the patents valid and infringed but denied a permanent injunction, despite the fact that the parties are competitors, on the ground that Nichia hadn't proven irreparable harm or that remedies at law (i.e., damages) were inadequate:
Specifically, the court found that “[t]he record shows an absence of meaningful competition.” . . .  It found that Nichia had “failed to establish past irreparable harm, or the likelihood of irreparable harm in the future based on lost sales” or “based on price erosion.” . . . It found that Nichia’s licensing of the patents to major competitors suggested that harm from “infringement of the patents-in-suit is not irreparable.” . . . And it found that Nichia’s licensing practices have made “multiple low-priced non-infringing alternatives from competitors available to replace the accused Everlight products if such products were not available” (pp. 21-22). 
Nichia contested those findings, but the Federal Circuit concluded that the district judge didn't abuse his discretion in reaching them--e.g., "despite the parties’ stipulation that they competed in the same market, . . . the court found that Nichia failed to prove that this competition was meaningful," p.23; "[t]he court found that Nichia 'failed to establish that [Everlight was] responsible for causing a single lost sale in the U.S.'" (id.); the court rejected Nichia's claim of price erosion relating to a sale of LEDs to GE (p.24).  Moreover:
Nichia argues that the court wrongly found that its licensing activities precluded a finding of irreparable harm. Nichia argues that the court’s decision is contrary to settled law, as it applies a categorical rule that licenses preclude irreparable harm.
To the extent the court adopted a categorical rule, we agree with Nichia; such a rule would run afoul of our precedent. . . 
But to the extent that the court found that Nichia’s prior licenses weighed against a finding of  irreparable harm, we countenance that approach. While evidence of licensing activities cannot establish a lack of irreparable harm per se, that evidence can carry weight in the irreparable-
harm inquiry (pp. 24-25).
None of this is inconsistent with prior post-eBay U.S. case law, I don't think, but then we get to the kicker at the end:
We discern no clear error in the district court’s finding that Nichia failed to prove that it would suffer irreparable harm absent the injunction. On that traditional equitable factor, Nichia did not bear its burden. See W. T. Grant Co., 345 U.S. at 633. Because Nichia failed to establish one of the four equitable factors, the court did not abuse its discretion in denying Nichia’s request for an injunction (p.25).
As I've noted before, a literal reading of eBay, Inc. v. MercExchange, L.L.C., 547 U.S. 388 (2006), might seem to require the prevailing patent owner to prove all four of the factors Justice Thomas identified as being relevant to a grant of injunctive relief.  To quote from the opinion (p.391):
According to well-established principles of equity, a plaintiff seeking a permanent injunction must satisfy a four-factor test before a court may grant such relief. A plaintiff must demonstrate: (1) that it has suffered an irreparable injury; (2) that remedies available at law, such as monetary damages, are inadequate to compensate for that injury; (3) that, considering the balance of hardships between the plaintiff and defendant, a remedy in equity is warranted; and (4) that the public interest would not be disserved by a permanent injunction.
As I write in my book (p.103):
. . . a literal reading of Justice Thomas’s statement that “a plaintiff must demonstrate” the four factors would appear to depart from tradition not only insofar as it expressly allocates the burden to the plaintiff on all four factors but arguably requires the plaintiff to prevail on all four factors. . . . Some district courts have not read the opinion quite so literally, however. The district court on remand from the Supreme Court’s eBay decision, for example, denied a permanent injunction despite concluding that the third factor (balance of hardships) favored neither party.115/  Nevertheless, in at least one post-eBay opinion the Federal Circuit seems to have taken Justice Thomas at his word, stating that “not all patentees will be able to show injury, and even those who do must still satisfy the other three factors.”116/
115/  See eBay, 500 F. Supp. 2d at 584. See also I-Flow Corp. v. Apex Med. Techs., Inc., No. 07cv1200, 2010 WL 141402, at *1–2 (S.D. Cal. Jan. 8, 2010) (granting a permanent injunction even though only three of the four eBay factors favored an injunction; the balance of the hardships factor was neutral); cf. Smith & Nephew, Inc. v. Synthes (U.S.A.), 466 F. Supp. 2d 978, 982–85 (W.D. Tenn. 2006) (explaining that “[t]he four-factor eBay test is a balancing test under which the plaintiff must demonstrate that the totality of circumstances weighs in its favor” and granting an injunction even though two factors, the balance of the hardships and the public interest, were “speculative” but nevertheless also in plaintiff ’s favor) (emphasis added).
116/  i4i Ltd. P’ship. v. Microsoft Corp., 598 F.3d 831, 862 (Fed. Cir. 2010), aff ’d on other grounds , 131 S. Ct. 2238 (2011). In Robert Bosch LLC v. Pylon Mfg. Corp ., 659 F.3d 1142 (Fed. Cir. 2011), discussed infra at note 121 and accompanying text, the court held that an injunction should have issued where the first three factors favored the patentee and the fourth was neutral. Whether this more recent holding means that patentees do not have to prevail on all four factors is still unclear, however, given the odd phrasing of the fourth factor (that the public interest “would not be disserved” by entry of the injunction).
I wonder if there's any chance the court would agree to rehear this case en banc to resolve the matter?  (Though whether there is a conflict in Federal Circuit case law may be a matter of opinion.  Bosch, as noted in footnote 116 above, may not be very strong support for the proposition that the patentee doesn't need to prevail on all four factors, particularly in light of later statements by the court in Apple Inc. v. Samsung Elecs. Co., 809 F.3d 633, 646-47 (Fed. Cir. 2015), that the fourth factor usually weighs in favor of patentees because of the public interest in enforcing patents.)  In my view, as suggested above, requiring the patentee to prove all four factors is hardly an "equitable" or "traditional" approach.  (Doesn't it, in a sense, turn the inquiry into a bright-line rule of the sort the court says is undesirable?)  To be sure, I wouldn't want us to go back to the days of mandatory injunctive relief (though I would prefer that courts focus more on the underlying economic reasons for granting or denying injunctions, e.g., granting them reduces the burden on the courts to try to determine the value of an intangible asset, while denying them reduces the risk of holdup and other costs not related to the value of the invention).  And it may be that an injunction wouldn't have been a good idea in this particular case on its facts.  But reading the Supreme Court opinion in such a literal manner seems unwise to me.  It also leads U.S. law even farther afield from the standards that apply in other countries, where injunctive relief still mostly remains the norm.

For what it's worth, the Supreme Court uses very similar mandatory-sounding language in a post-eBay opinion on preliminary injunctions, Winter v. Natural Resources Defense Council, 555 U.S. 7 (2008) ("A plaintiff seeking a preliminary injunction must establish that he is likely to succeed on the merits, that he is likely to suffer irreparable harm in the absence of preliminary relief, that the balance of equities tips in his favor, and that an injunction is in the public interest").  Again, I'd be surprised if the Court really meant that plaintiffs must prevail on all four factors to obtain a preliminary injunction, but who knows?  For further discussion, see Mark Gergen, John M. Golden, & Henry E. Smith, The Supreme Court’s Accidental Revolution? The Test for Permanent Injunctions, 112 Columbia Law Review 203, 210-11 & n.35 (2012) (stating, inter alia, that "[c]ircuits have split over whether the preliminary injunction analog of the eBay test articulated in Winter . . .  abrogates a previously common approach under which a particularly strong showing on one factor (e.g., irreparable harm) can mean that an otherwise inadequate showing on another factor (e.g., likelihood of success) suffices").

Some of the discussion among my colleague law professors about this case over the past few days has suggested that maybe irreparable harm is now a sine qua non for injunctive relief, but that an injunction could still issue under appropriate circumstances even if the evidence doesn't favor the plaintiff on some of the other factors.  Maybe that's a reasonable way of looking at the matter, post-eBay.  Whether it makes sense or not is another matter, however; and whatever the correct interpretation is, in my view the way the Supreme Court has articulated the standard is clumsy and prone to potential misinterpretation.

For Dennis Crouch's insightful discussion of this case on Patently-O, see here.