Probabilistic Discounts and FRAND/Reasonable Royalties

Interesting post today by David Long on the Essential Patents Blog about a recent decision by Judge Rodney Gilstrap in a pending patent infringement case, St. Lawrence Comm'ns LLC v. ZTE Corp., permitting St. Lawrence's expert to offer an opinion that the royalty rate that St. Lawrence had previously negotiated with Samsung for the use of the patents in suit in the pending case against ZTE and Motorola should be increased by 50% to reflect a "settlement discount" and 18% to account for an "invalidity discount."  (The percentages come from St. Lawrence's negotiation with yet another firm, LG, which had proposed to take a license subject to such discounts.)  In other words, if I understand correctly, the expert's opinion is intended to account for the fact that negotiated royalties usually reflect a discount reflecting the probability of validity and infringement, whereas a reasonable royalty awarded at trial is supposed to reflect the bargain the parties would have struck ex ante knowing the patent to be valid and infringed.  (That counterintuitive assumption is necessary to avoid a double discounting problem, as I have explained, e.g., here).  Nevertheless, to my knowledge it hasn't been common up to now for courts to allow an expert to make an explicit modification to a comparable license rate to reflect this assumption.  Jonathan Masur discusses the issue extensively in his article The Use and Misuse of Patent Licenses, 110 Nw. U. L. Rev. 115, 120 (2015), and Masur and Erik Hovenkamp return to it in their article Reliable Problems from Unreliable Patent Damages, 25 Tex. Intell. Prop. L.J. __, __ (forthcoming 2017), https://papers.ssrn.com/abstract_id=2825236.  

Returning to the St. Lawrence case, Judge Gilstrap also will permit the expert to base his opinion on, among other things, licenses for which the defendant Motorola is a licensor rather than a licensee, stating that an expert may rely on evidence "that does not explicitly fall within the Georgia-Pacific framework" (p.5).  Georgia-Pacific factor two explicitly refers only to licenses paid by the defendant for the use of other patents, but the court concludes that the expert "adequately explains the relevance of this evidence" (p.6).  On the other hand, the court will exclude the expert from testifying that the royalty should be adjusted upward to account for the fact that it would last 2.5 years longer than the Samsung royalty, inasmuch as the running royalty that the expert is proposing here necessarily would account for the longer duration, and the court perceives no other evidence to support an adjustment in this regard.

Finally, the court will permit the expert to use licenses negotiated under threat of an injunction in Germany.  According to the court, "Mr. Weinstein does not use these licenses as a starting point for his opinion. Rather, he uses these additional licenses to justify an adjustment of the hypothetical royalty rate. In doing so, he explicitly takes into account how an injunction could place have placed pressure on the licensees and affected the rates of these additional licenses. As with many of its other challenges, Motorola’s argument goes to the weight and not admissibility of the testimony" (p.8).

There are a lot of issues to mull over here, and for anyone interested in these issues I recommend the Essential Patents post, as well as the Masur and Hovenkamp papers.

 

In other news, yesterday in Smartflash LLC v. Apple Inc., the Federal Circuit reversed a judgment in a case over which Judge Gilstrap had presided that had originally resulted in a jury award against Apple in the amount of $533 million.  (See previous discussion, here and here.  Judge Gilstrap had already granted a conditional new trial on damages, however, according to this article.)  The ground for reversal, which had been anticipated by concurrent PTAB decisions against which Smartflash had lodged an appeal, was lack of patentable subject matter.