Magistrate Judge Grewal in N.D. Cal. recently issued an Order excluding the testimony of Golden Bridge Technology’s damages expert because it was based on a flawed methodology for determing a fair, reasonable and non-discriminatory (FRAND) royalty rate for the asserted patent alleged to be essential to the 3GPP WCDMA standard.  The primary problem appears to be the expert’s failure to allocate value to the specific patent apart from non-patented product features and failure to rely on real-world industry practices (as opposed to theoretical or asperational practices):

In calculating the royalty base, Schulze did not even try to link demand for the accused product to the patented feature, and failed to apportion value between the patented feature and the vast number of non-patented features in the accused products.  Schulze [the damages expert] had no basis to ignore the fundamental teaching of the entire market value rule, which permits a royalty based on the entire market value of an accused product only where “the patent-related feature is the basis for customer demand.”  In calculating the royalty rate, Schulze impermissibly relied on licenses without any showing of comparability, relied on a maximum, cumulative royalty rate without any showing that anyone had committed to such a notion and failed to allocate particular value to the invention claimed in the ’793 patent[in-suit].

Judge Grewal stated that, in this case, the typical reasonable royalty analysis involved additional consideration of FRAND licensing terms resulting in a modified Georgia-Pacific hypothetical negotiation that seeks to prevent “patent holdup” or “royalty stacking” by ensuring that “a FRAND license should compensate a patentee for their technical contribution to the technology embodied in a standard, but should not compensate them for mere inclusion in the standard.”  Judge Grewal thus focused on ensuring that the royalty was based on the technical contribution of the patent (without further discussing or giving weight to concerns about “patent holdup” or “royalty stacking”).

Flawed Royalty Base.  Judge Grewal ruled that the expert failed to show an industry practice in using the entire unit price — e.g., price of Apple’s iPad or iPhone — as the royalty base, failed to apportion the royalty base among patented and non-patented features and failed to show that the patented feature drove demand for the entire unit as required under the entire market value rule.  This conflicted with the Federal Circuit’s Lucent decision that “stands for the proposition that, in the absence of evidence that the infringing feature drives demand, the royalty base must be somehow apportioned to reflect the value of the patent-related feature in the absence of the non-infringing features.”  The expert’s attempt to rely on industry practice in using a per unit price relied on aspirational papers of a per unit price without showing “actual practice in the field.”  Further, the entire unit is not the smallest saleable unit, because “throughout this litigation, GBT [the patentee] has taken the position that the entire infringing functionality lies in the baseband processor, not the accused product as a whole.”  Even if the entire unit were the smallest saleable unit, that would “not relieve a patentee of the burden of apportioning the base,” which the damages expert failed to do here.

Flawed Royalty Rate.  Judge Grewal also found that the damages expert’s royalty rate analysis was flawed.  First, the expert erroneously relied on portfolio licenses that the patentee secured in settling several infringement cases that did not involve the patent at issue here.  The expert did not assess the comparability of those licenses by comparing (1) the value of those portfolio patents to those accused products or (2) the value of the patent-in-suit to the accused products in this case.  Further, the expert did not account for the lump-sum settlement payment in those licenses as having included future sales.

Second, the expert erroneously relied on there being some maximum, cumulative royalty rate for all WCDMA patents even though “he identifies no evidence that any party ever agreed to such a [maximum, cumulative] rate.”

Third, the expert erroneously treated the value of the ’793 Patent-in-suit to be the same as the value of all other patents essential to the WCDMA standard.  The expert appears to have done this by doing a straight division of the number of SEPs into an assumed estimated cumulative royalty rate for all SEPs, thus giving each SEP the same value:

The heart of Schulze’s rate analysis dealt with [Georgia-Pacific] factor 12.  Based on his study of the 1999 UMTS Intellectual Property Association proposed maximum global royalty rate, and various academic studies regarding cumulative rates for standard-essential patents, Schulze opined that the participants in the market for 3G technology would have settled on a certain cumulative royalty rate for all 3G standard-essential patents.  He then calculated a royalty for the ’793 patent[in-suit] alone as follows.  Looking to the declared WCDMA standard essential patent families provided by Sipro Lab Telecom, and Sipro’s classificaton of those families, he determined the percentage of the families related to terminals like the accused products.  Schulze then multiplied the nuimber of WCDMA standard essential patents identifed in one article by the percentage to determine the number of patent famlies, including the ’793 patent, that read on terminals.  Dividing his estimated cumulative royalty rate by number of terminal-related eseential patents, he arrived at a rate for the ’793 patent.  Regarding factor thirteen, Schulze assumed that each standard essential patent shares an equivalent value to each other standard essential patent within the overall standard.

Judge Grewal found such methodology flawed, stating that “the case law is clear that mere patent counting and dividing is not enough”, which does not account for the particular value (vel non) of the patent-in-suit:

Third, Schulze erred in assuming the value of the ’793 patent was no different than the value of each of the other WCDMA standard-essential patents considered.  GBT identifies no case law supporting the notion that a claimed standard-essential patent gets a free pass on the fundamental notion that a patent damages methodology must be tied to the relevant facts and circumstances of the case at issue.  If anything, the case law is clear that mere patent counting and dividing is not enough.

Judge Grewal, therefore, excluded the patentee’s damages expert opinions “in their current form”, but allowed the expert “another shot” to tender a new report with opinions consistent with the court’s ruling.