Thursday, September 26, 2019

Cert petition, amicus brief criticize Federal Circuit for vitiating damages apportionment requirement: Time Warner Cable v. Sprint

The United States Courts of Appeals for the Federal Circuit has always been viewed as being more sympathetic to patent holders than to alleged infringers--but not in every single aspect of patent law. For an example, some critics of former Chief Judge Randall Rader, who disparaged the PTAB as patent "death squads," acknowledged that his rulings on patent infringement damages provided some important clarifications beneficial to defendants.

Two documents that have recently been filed with the Supreme Court of the United States argue that various Federal Circuit rulings in recent years have gutted the apportionment requirement for patent damages claims:

  • On August 15, Time Warner Cable (which doesn't exist anymore in its original form, but a Charter Communications subsidiary is now continuing the litigation in question) filed a petition for writ of certiorari (request for Supreme Court review) in its patent infringement dispute with Sprint. A district court--affirmed by the Federal Circuit--found that Time Warner infringed a Sprint patent on connections between VoIP and pre-VoIP-era telecommunications networks, and awarded $140 million in damages, based on a reasonable-royalties theory that Time Warner argues failed to apportion between infringing and non-infringing features. Time Warner's petition raises two distinct issues: the one relating to apportionment and one about the written-description requirement.

  • Last week, Intel filed an amicus curiae brief in support of the first part (apportionment) of the cert petition.

Both the petition and the amicus brief place particular emphasis on a 135-year-old Supreme Court ruling: In Garretson v. Clark (1884), the top U.S. court stated that "the [prevailing] patentee [seeking damages] must in every case give evidence tending to separate or apportion the defendant's profits and the patentee's damages between the patented feature and the unpatented feature." And the highest court in the land added that "such evidence must be reliable and tangible, and not conjectural or speculative." In that case, the patent-in-suit read on an improved mop head, but not the cleaning device as a whole.

Interestingly, it was also in the late 19th century when it became law that a prevailing design patent holder was entitled to an unapportioned disgorgement of infringer's profits. A typical example of a design patent-infringing product at the time was a carpet. In the Apple-Samsung dispute, there was a strong policy argument that today's highly multifunctional products had to be analyzed and treated differently from 19th-century products. But in the Garretson utility patent case, apportionment already came into play even though it was a no-tech (not even a low-tech) product by today's standards. Undoubtedly, what was already warranted in the Garretson mop-head case is hugely more relevant in the smartphone era.

There's some indication that the Supreme Court may have felt last year that the question of apportionment at least potentially warranted another look: on April 4, 2018, the Supreme Court invited the Solicitor General to express the views of the federal government on the cert petition in EVE-USA, Inc. v. Mentor Graphics Corp.--but before the DOJ responded to this CVSG, the petition was withdrawn as a result of a settlement.

Yesterday, Sprint responded to Time Warner's petition and Intel's amicus brief.

With a view to any unfinished business left over from EVE-USA, Sprint argues that it was a lost-profits case, while the case pending now is about a reasonable royalty (as a damages theory). That's a weak point because the issue was and is apportionment.

Like pretty much any party opposing a cert petition, Sprint also claims that it's all about a factual determination rather than a need for important legal clarification. While I don't agree with Sprint as far as the apportionment question is concerned, they may have a point here with respect to the scope of the patent-in-suit in light of the written description. But that one is only a lower-priority second part of Time Warner's petition, and there is no support whatsoever from any amicus curiae for that part. I would have thought that more parties than one (Intel) would throw their weight behind the apportionment question, but one theory I have is that some of the organizations that typically care about such issues have business relationships with Sprint and/or work closely with Sprint on other policy issues.

The issue that Time Warner Cable and Intel complain about in their filings is that the Federal Circuit has in recent years deemed the apportionment requirement to be satisfied by other means than an apportionment in a strict sense. In a strict sense, apportionment would really mean to tell the jury what the commercial value of the non-infringing parts and features of a product is versus the deemed-infringing one(s)--and to then determine what percentage of that commercial value would constitute a reasonable royalty or lost profits. However, the Federal Circuit's disturbingly permissive approach has recently been to content itself with such alternative approaches as simply seeking a "low" royalty rate on an entire product--and there would be other examples that contrast with various Federal Circuit decisions earlier this decade or even before, all of which showed that the appeals court took the judiciary's gatekeeper role in the damages theories context very seriously.

Damages determinations are put before juries, and jurors easily get misled by damages theories based on the entire commercial value of a complex multifunctional product. Even the distinction between a mop head and the rest of a cleaning device can make a significant difference. The difference between a patent infringed by a $5, $10 or $20 chip versus the value of an entire smartphone is far more substantial. But this gets even worse when someone may assert a patent on a cup holder in a car, or on one feature of numerous computer programs shipped with a car.

Again and again and again, Sprint's opposition brief points to the fact that their trial evidence included, among other things, allegedly-comparable license agreements that should be deemed strong evidence of the market value of the patented invention--evidence that already involves an apportionment because the parties who negotiated the relevant agreements will have taken the need for apportionment into account.

Comparable license agreements bear considerable weight with judges and juries. With a view to their relevance to apportionment, Sprint's opposition brief implicitly suggests two things:

  1. Under what Sprint considers the correct standard, such license agreements should in and of themselves be deemed to satisfy the apportionment requirement. In other words, Sprint would like to use such top-down evidence to eliminate the need for a more conventional bottom-up apportionment where one attaches a part of the entire commercial value of a product to the infringing feature(s) and the remainder to the non-infringing ones.

  2. Sprint seems concerned that the apportionment standard suggested by the petitioner would make license agreements relating to entire products (and not just to particular components) inadmissible as evidence.

I disagree with #1, and believe #2 overshoots.

License agreements can't serve as a substitute for an apportionment analysis presented to the jury. The Garretson language doesn't appear to allow that kind of end-run around the apportionment requirement.

But that doesn't mean such license agreements would be inherently inadmissible under the standard proposed by Time Warner. If a jury is presented with apportionment evidence in the case before it, and if the jury is furthermore informed of how to analyze a third-party license agreement against this apportionment background, then it may still work. Let me give an example: a patent reads on a particular type of chip, and that kind of chip is commonly found in WiFi routers as well as smartphones. The patent holder negotiated a royalty of 0.5% of the end price of a WiFi router, and presents that agreement as evidence in a dispute with a smartphone maker. What clearly wouldn't make sense is to argue that a WiFi router manufacturer's decision to pay a 0.5% royalty means the same royalty rate should apply to smartphones. But the royalty can be normalized by calculating what royalty rate was actually paid with respect to the relevant component. If the relevant component in the accused smartphone is also properly identified, then the router license agreement may have a probative value that outweighs the risk of jury confusion.

It's too early to go into much detail on the merits. Right now, it's about certworthiness. The petition and Intel's amicus brief make a plausible case that the Federal Circuit has recently rubberstamped damages theories that don't appear to satisfy the apportionment requirement that one can reasonably read into the statute (35 U.S.C. § 284), the Garretson decision, and various post-Garretson decisions. Only the Supreme Court can get the Federal Circuit back on the right track by clarifying that Garretson, after 135 years, is still good law and even far more critical--in economic terms--than anyone could have imagined back in the 19th century.

Share with other professionals via LinkedIn: