Given the devastation done to the Cockburn report by the Court's order from Monday, the striking of these portions of the Cox and Leonard reports is not a great loss to Google.
IN THE UNITED STATES DISTRICT COURT
FOR THE NORTHERN DISTRICT OF CALIFORNIA
ORACLE AMERICA, INC.,
Plaintiff,
v.
GOOGLE INC.,
Defendant.
No. C 10-03561 WHA
ORDER GRANTING IN PART
AND DENYING IN PART
DAUBERT MOTION TO STRIKE
GOOGLE’S SUPPLEMENTAL
EXPERT DAMAGES REPORTS
________________________________________
INTRODUCTION
In this patent-infringement action, plaintiff moves to exclude portions of defendant’s
supplemental expert damages reports. For the reasons stated below, the motion is GRANTED IN
PART and DENIED IN PART.
STATEMENT
The background was set forth in previous orders (see Dkt. Nos. 230, 685, 785). In
response to Dr. Ian Cockburn’s third expert damages report, Google’s experts, Drs. Gregory
Leonard and Alan Cox were permitted to submit supplemental damages report that directly
responded to new material in Dr. Cockburn’s third report (Dkt. No. 702). Dr. Leonard’s
supplemental report on patent damages was based on the six patents previously asserted, the ’476,
’702, ’720, ’205, ’104, and ’520 patents. While briefing was underway, one was withdrawn with
prejudice. Another three rejected by the PTO examiner were withdrawn if the trial is held before
the administrative appeals are completed, a withdrawal whose effect will be considered below.
Therefore, there is a strong possibility that only the ’104 and ’520 patents will be asserted at trial,
and this order will only address issues pertaining to these two patents, without prejudice to
revisiting objections specific to the withdrawn patents if they later arise.
In his supplemental report on patent damages, Dr. Leonard conducts a so-called “forward
citation” analysis to arrive at his own ranking of the 22 patents from Dr. Cockburn’s report. This
was done to critique Dr. Cockburn’s “upper bound” opinion that three patents in suit were the
most valuable of the 22. Now that Dr. Cockburn’s “upper bound” opinion has been stricken and
the number of asserted patents reduced, the only remaining issue is whether Dr. Leonard can
opine that the ’104 patent ranks in the “middle or worse” of the top 22 patents. Specifically,
whether Dr. Leonard can opine that the ’104 patent ranks 17th among the top 22 patents under the
forward-citation analysis (Rpt at 7).
Dr. Leonard also criticizes Dr. Cockburn’s valuation of Sun’s Java mobile patent
portfolio, which included 569 Sun patents, for being too high. For support of his opinion, Dr.
Leonard references an Oracle acquisition-accounting document prepared in 2010 that purports to
value Sun’s Core Technology, which included 14,000 Sun patents (including the patents in suit),
at $505 million (Rpt at 9).
In his supplemental report on copyright damages, Dr. Cox decreases his calculated
damages for both copyright disgorgement and copyright lost profits despite the fact that Dr.
Cockburn’s third report did not revisit or revise his opinion on either subject. In Dr. Cox’s
original October 2011 report, he had calculated damages for copyright disgorgement at $39.6
million and lost profits at $5.7 million. Those calculations relied on Dr. Steven Shugan’s conjoint
analysis’ estimate of Android’s market share and revenue attributable to copyright infringement.
Specifically, Dr. Cox relied on the conjoint analysis’ estimate that approximately 14% of
Android’s market share and approximately 8% of Google’s revenue was due to Google’s
infringement of the 37 API packages. Now, in his supplemental report, Dr. Cox abandons his
prior reliance on the conjoint analysis and instead relies on Dr. Cockburn’s new group-and-value
apportionment of the 2006 bundle for patent and copyright reasonable royalty. Using the groupand-
value’s “lower bound” number of 5.1% for copyright allocation, Dr. Cox revises his
2
calculated damages of copyright disgorgement to $2.3 million and copyright lost profits to
$2.1 million (Rpt at 8–12).
ANALYSIS
An expert witness may provide opinion testimony “if (1) the testimony is based upon
sufficient facts or data, (2) the testimony is the product of reliable principles and methods, and (3)
the witness has applied the principles and methods reliably to the facts of the case.” FRE 702.
District courts thus “are charged with a ‘gatekeeping role,’ the objective of which is to ensure that
expert testimony admitted into evidence is both reliable and relevant.” Sundance, Inc. v.
DeMonte Fabricating Ltd., 550 F.3d 1356, 1360 (Fed. Cir. 2008).
1. DR. LEONARD’S REPORT ON PATENT DAMAGES.
A. Forward-Citation Ranking
of the Top 22 Patents.
As a critique of the “upper bound” opinion in Dr. Cockburn’s group-and-value approach,
Dr. Leonard conducts a forward-citation ranking of the patents within the top 22. Specifically,
Dr. Leonard uses the citation count of each of the top 22 patents as a proxy for the value of that
patent. First, Dr. Leonard counts the number of times each patent has been cited by any another
U.S. patent. Next, he “control[s] for patent age by comparing each patent’s forward citation
count to the number of forward citations for other patents in the same technology class and of
approximately the same age. Specifically, [he] considered patents that were issue within three
years before or after the subject patent’s issue date” (Leonard Decl. ¶ 4). Finally, he ranks the top
22 patents by their age-adjusted citation counts to get the following result (Rpt at 7):
[T]he ’104/’205/’720 patents in suit rank 10th (’720), 11th
(’205), and 17th (’104) among the 22 patents — again, in
the middle of the pack or worse . . . . This, in turn, suggests
that only the lower bound on Dr. Cockburn’s range has any
support.
With respect to the ’104 patent, Dr. Leonard’s ranking is fatally flawed. Dr. Leonard fails
to account for the fact the ’104 patent was re-issued twice, and thus fails to include citation counts
to its predecessor patents. The predecessor to the ’104 patent was US Patent RE36204, which
was applied for in November 1996 and issued in April 1999, and has one citation. The
predecessor to USRE36204 was US Patent 5367685, which was applied for in December 1992
3
and issued in November 1994, and has 73 citations (Cockburn Decl. ¶ 5). Dr. Leonard fails count
the 74 predecessor citations and instead counts only three citations for the ’104 patent
(Richardson Decl. Exh. C). Had those predecessor citations been counted, the rank of the ’104
patent would have jumped from 17th to the top (see Cockburn Decl. ¶ 6).
Google responds to this criticism by arguing that because the ’104 patent has been
reissued with different claims, then counting citations to predecessor patents would have been
inappropriate. This argument is not persuasive. Patents are not cited for their claim language;
instead, patents are cited if they disclose important ideas material to a later application’s
patentability. That is why the citations are to the entire patent, which is largely composed of
specifications and drawings, not claims. The predecessor patents to the ’104 patent had the same
specifications and drawings. Not counting citations to these predecessor patents is error.
Therefore, Dr. Leonard’s ranking of the ’104 patent and any opinion based on that ranking
is STRICKEN.
B. Oracle’s 2010 Accounting Document.
As an alternative base for a reasonable royalty allocation, Dr. Leonard relies on an
accounting document prepared by Oracle in connection with its January 2010 acquisition of Sun.
In April 2010, a few months after Sun’s acquisition, Oracle’s accounting department prepared an
“Estimation of the Fair Value of Certain Assets and Liabilities of Sun Microsystems, Inc. as of
January 26, 2010” report. The report was prepared to assist “Oracle Corporation management in
allocating the Sun purchase price for financial report purposes.” One section of this document
calculated the fair value of the Sun Core Technology, which included all of Sun’s 14,000 patents
in 2010 (including the patents in suit), at $505 million (Zimmer Decl. Exh. F). Dr. Leonard
opines that this amount should serve as the base for apportionment of the patents in suit in the
2006 hypothetical negotiation.
According to the accounting document, $505 million was the calculated “fair value” for
14,000 patents under a “profit allocation method.” Fair value was defined as “the price that
would be received to sell an asset or paid to transfer a liability in orderly transaction between
market participants at the measurement date [January 2010].” The profit-allocation method
4
estimated the value “by capitalizing the profit saved because the company owns the technology.
In other words, the owner realizes a benefit from owning the intangible asset rather than paying a
rent or profit for the use of the asset” (Zimmer Decl. Exh. F). To summarize, the $505 million
was an estimate of a reasonable royalty for an unrestricted, perpetual license (although not
necessarily an exclusive license) of all 14,000 patents in Sun’s 2010 patent portfolio. Although
this valuation was on Sun’s 2010 patent portfolio, four years after our 2006 hypothetical
negotiation, it is nevertheless relevant to the reasonable royalty analysis.
Post-infringement information can be helpful in assessing whether a royalty is reasonable.
Lucent Technologies, Inc. v. Gateway, Inc., 298 F.3d at 1313–14 (2009). Here, the 2010
document is relevant because the patents in suit were also owned by Sun in 2010. Thus, the $505
million valuation would have included the patents in suit. This dollar amount, in turn, can shed
light on the reasonableness of the royalty estimates by the parties’ experts. Rephrased in the
words of our Supreme Court, the 2010 document can “bring out and expose to light the elements
of value that were there from the beginning.” Sinclair Refining Co. v. Jenkins Petroleum Co., 289
U.S. 689, 691-92 (1933). Although the number of total patents owned by Sun likely changed
between 2006 and 2010, this does not make the 2010 valuation irrelevant as there is no evidence
that Sun’s patent portfolio unforeseeably surged or plunged in value between 2006 and 2010.
Oracle argues that the report should be excluded because the profit allocation method does
not measure the full value of Sun’s intellectual property in 2006, including profits earned through
licensing to third parties such as Google. This may be true. Nevertheless, the profit allocation
method is relevant because it estimates the value of a license to Sun’s patent portfolio. Since the
2006 negotiation between Google and Sun was for a license to the patents in suit, not a complete
sale of the patents in suit, the $505 million calculation is relevant and will not be excluded.
C. Opinion That Reasonable Royalty Should Be Limited to the Value that
Google Was Expecting to Receive.
Oracle requests that the Court strike Dr. Leonard’s opinion that the reasonable royalty
should be limited to Google’s expected revenue from a license. In his supplemental report, Dr.
Leonard opines that “[i]t is the value that Google was expecting to receive that matters for the
5
reasonable royalty analysis” (Rpt at 8). This statement is too close to an inappropriate suggestion
of law.
The infringer’s revenue from the infringement does not cap the amount of a reasonable
royalty. Both patentee’s and infringer’s expected revenues and losses should be considered in
calculating a reasonable royalty. The court of appeals has held that “although an infringer’s
anticipated profit from use of the patented invention is among the factors to be considered in
determining a reasonable royalty, the law does not require that an infringer be permitted to make
a profit.” Monsanto Co. v. Ralph, 382 F.3d 1374, 1384 (Fed. Cir. 2004) (emphasis added).
Permitting Dr. Leonard to offer testimony that is contrary to law would intrude on the Court’s
role in instructing the jury as to the relevant considerations for calculating a reasonable royalty.
Therefore, Dr. Leonard’s opinion that “[i]t is the value that Google was expecting to receive that
matters for the reasonable royalty analysis” is STRICKEN.
Subject to the rules of evidence, Dr. Leonard can opine that “Google would have decided
how much it was willing to pay Sun based on its expectations of the value of the partnership to
Google” (Opp. at 11), just as Dr. Cockburn can opine that Sun would not have licensed its
intellectual property for an allegedly incompatible Android without full compensation for
its losses.
2. DR. COX’S REPORT ON COPYRIGHT DAMAGES.
The order allowing Google to submit supplemental damages reports limited the revisions
to only those “directly responsive to new material by Dr. Cockburn” (Dkt. No. 702). Dr. Cox’s
calculation of copyright disgorgement and copyright lost profits in his supplemental report goes
beyond this instruction because Dr. Cockburn did not revisit either of those topics in his third
damages report.
In his original October 2011 report, Dr. Cox chose not do perform his own Android
market-share analysis. Instead, his calculations of copyright disgorgement and copyright lost
profit relied on Dr. Shugan’s estimate of incremental Android market shares and revenue
attributable to the 37 API packages. Specifically, Dr. Cox took the results from Dr. Shugan’s
conjoint analysis to come up with 8.1% for the incremental Android revenue wrongfully earned
6
and 13.45% for the market share of Android attributable to the copyright infringement (Dearborn
Decl. Exh. A at 58, 38–40).
None of the information relied on by Dr. Cox in his October 2011 report changed in Dr.
Cockburn’s third damages report: Dr. Shugan’s conjoint analysis was not changed. No new
estimates of Android market share and revenue attributable to the copyright infringement were
calculated. Therefore, Dr. Cox revisions in his supplemental report were not “directly responsive
to new material by Dr. Cockburn.”
Instead of directly responding to new material in the third damages report, Dr. Cox simply
takes numbers from Dr. Cockburn’s new group-and-value approach, which did not calculate
Android revenue or market share, and uses them in a new methodology for calculating copyright
disgorgement. Such a revision was not allowed by the order permitting supplemental reports.
Furthermore, in other calculations, Dr. Cox simply substitutes numbers from Dr. Shugan’s
conjoint analysis with lower numbers from Dr. Cockburn’s new group-and-value approach.
These are inappropriate substitutions. Dr. Cockburn’s group-and-value approach and Dr.
Shugan’s conjoint analysis were two completely different calculations. Dr. Shugan’s conjoint
analysis purported to estimate 2008-2011 Android market share and revenue. The
group-and-value approach purported to apportion the 2006 license bundle. The problem with
equating the two calculations has been stated many times and will not be repeated here (Dkt. No.
685, 785).
In his October 2011 report, Dr. Cox chose to rely on the market share estimates from
Dr. Shugan’s conjoint analysis. He cannot backpedal from that reliance now. This remains true
even though a recent order has stricken the conjoint analysis’ determination of market share (Dkt.
No. 785). This is because Google itself argued that the conjoint analysis’ determination of market
share was unreliable in its Daubert motion against Dr. Cockburn’s third damages report (Dkt. No.
718 at 14–17). Google has put itself in this difficult position by arguing that the conjoint analysis
was unreliable while knowing that its own expert was relying on the results. Google’s expert
cannot abandon his calculations now.
7
In sum, because Dr. Cox’s calculation of damages for copyright disgorgement and lost
profit in his supplemental report are not directly responsive to new material by Dr. Cockburn,
these calculations are STRICKEN.
CONCLUSION
For the reasons stated, Dr. Cox’s calculation of damages for copyright disgorgement and
copyright lost profit in his supplemental report is STRICKEN. Dr. Leonard’s statement that “[i]t is
the value that Google was expecting to receive that matters for the reasonable royalty analysis” is
STRICKEN. Dr. Leonard’s forward-citation ranking of the ’104 patent is STRICKEN. The motion
to strike Dr. Leonard’s ranking of the ’205 and’720 patents is denied without prejudice. The
motion to exclude the 2010 acquisition document is DENIED. Nothing stricken by this order
affects the experts’ original reports, which are not the subject of this Daubert motion.
IT IS SO ORDERED.
Dated: March 15, 2012.
/s/ William Alsup
WILLIAM ALSUP
UNITED STATES DISTRICT JUDGE
8
