Tuesday, March 26, 2013

Intel: 'Injunctions [over] FRAND-encumbered SEPs create a risk of coerced windfall settlements'

Intel's amicus curiae brief supporting Apple with respect to Judge Posner's FRAND ruling was the first one to be filed and today became the first one to be publicly accessible. I have read it, and I like it. A lot of its content is consistent with other pro-FRAND positions I've previously reported on and stated myself, but I wanted to publish the document for all those who care to read it in detail and also wanted to highlight (further below) a few particularly well-worded passages. Here's the brief (this post continues below the document):

13-03-20 Intel Amicus Brief in Support of Apple on FRAND

Intel supports Judge Posner's denial of injunctive relief as well as his position on damages.

In the headline of this post I already quoted how Intel describes the risk of patentee overcompensation resulting from negotiations under the threat of an injunction:

"Injunctions with respect to FRAND-encumbered SEPs create a risk of coerced windfall settlements that would distort competition, undermine the standard-setting process, and injure consumers."

In connection with the second risk -- "undermine the standard-setting process" -- Intel notes that at the end of the days it's not even in interest of SEP holders to allow abuse:

"Taken to its extreme, the issuance of injunctions against standards implementers would in fact be contrary to the interests of all SEP holders because it would undermine the widespread adoption and utilization of interoperability standards in furtherance of which the SEP holders made the FRAND commitments."

Intel is right. There are two reasons, however, for which certain SEP holders won't be convinced by this argument to behave more reasonably. Pressure to maximize short-term revenues is huge (most SEP holders in this industry are publicly-traded companies) and next quarter's numbers will in many cases be a higher priority than long-term sustainability of the standard-setting system. And if companies like Google (Motorola) and Samsung are desperately trying to get away with infringement of non-SEPs (by hoping to bring about a stalemate through SEP injunctions), they just have their own competitive interests in mind and not the greater good of the standardization system. Therefore, it's up to the courts to protect the system, with long-term considerations in mind, against self-serving abuse.

Intel's brief also describes very well the effect that incorporation of a patented technique into a standard has on the viability of alternative solutions:

"Before a standard's adoption, the royalties that the patentee could demand from licensees reflect only the value of its patent relative to other methods of achieving the same technological objective. Once an SSO [standard-setting organization] adopts an interoperability standard that incorporates a particular patent, however, substitutes for the technology typically lose commercial viability, which--absent FRAND commitments--enables SEP holders to extract supra-competitive royalties from firms that must implement the standard."

"When SSOs’ interoperability standards attain substantial commercial acceptance, [...] substitute technologies are often effectively foreclosed."

Others have said this before, but saying that "substitutes [...] lose commercial viability" is a really good wording. If people have misconceptions about the value of the innovation behind and protected by SEPs, it's usually because they focus on the importance of what a technique does (for example, a handover between two base stations in a cellular network) rather than the fact that there may be, and often are, countless other ways to achieve the same effect, just that all those alternative solutions wouldn't be interoperable because the standard prescribes, for example, only one handover mechanism.

Intel summarizes its position on FRAND royalties (and related damages) as follows:

"Third, in order that the royalty be 'reasonable,' FRAND royalty-setting must be based only on the value of the component that implements the standard; reflect the ex ante value of the SEPs (i.e., their value before they were incorporated into the standard); and take into account all the other technologies (patented or otherwise) incorporated into the component."

The first part -- "based only on the value of the component that implements the standard" -- is an argument Apple emphasizes all the time in the policy and antitrust contexts (as well as in litigation). The usual counterargument (made, for example, by Nokia in a submission to the ITU) is that a lower percentage on a broader royalty base will have the same bottom-line result as a higher percentage applied to a narrower base ("as 1% of a base of 100 leads to the same result as 10% of a base of 10, which royalty structure is actually used has no economic relevance, as long as the result reflects the appropriate value for a license"). I believe the whole process of courts determining FRAND royalties (be it for the purpose of calculating past damages or for a FRAND rate determination with a view to the future) would work a lot better if all patent holders were required to base their demands on what is in each case the smallest salable unit implementing the relevant standard. Everything else serves no useful purpose. It just allows SEP holders to muddy the water and complicate the process.

Intel's brief also supports Judge Posner's decisions to exclude FRAND damages claims. In this context Intel endorses an amicus brief filed in the same appeal back in December 2012 by Cisco, HP and other companies, which brief addressed damages methodologies in general (not limited to royalties for FRAND-pledged SEPs). Intel says that "Judge Posner's exclusion of Motorola's damages experts should be affirmed as a textbook example of proper gate-keeping by a trial judge". Like Cisco, HP and friends, Intel also criticizes the 15 Georgia-Pacific factors for patent damages:

"[T]hat outmoded and unworkable approach produces economically unsound results, confuses juries, and distorts reasonable royalty calculations, to the detriment of the patent system and the economy as a whole."

In Intel's opinion (and in mine, by the way), "Georgia-Pacific should have no application in the FRAND context, and this Court should insist on damages models that better reflect actual business decisions and market realities."

One of the alternatives Intel considers preferable over Georgia-Pacific in FRAND cases is "the next-best-alternative methodology, which Judge Posner properly adopted and applied when he excluded testimony from one of Motorola's damages experts".

The next-best-alternative methodology is closely related to the ex ante approach Google criticizes in its opening brief. As I wrote in my commentary on that brief, I don't think the point in time -- which Google focuses on -- is all that important. What really matters is that there's a scenario in which a technology is licensed on its merits and has to compete with alternatives (especially the next best alternative), and there's one in which a group of industry players formed a consortium and defined a standard that precludes those alternatives. The temporal aspect isn't nearly as important as the competition-related one. If the reasonableness analysis is based on the hypothetical assumption that alternatives are still commercially viable at the time of infringement, and if the intrinsic value of the patented technology is properly separated from the value of standardization itself, then the result will always be a FRAND royalty. There may even be additional technical alternatives available at that time that weren't known at the time of standard-setting.

If you'd like to be updated on the smartphone patent disputes and other intellectual property matters I cover, please subscribe to my RSS feed (in the right-hand column) and/or follow me on Twitter @FOSSpatents and Google+.

Share with other professionals via LinkedIn: